article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the main challenge for existing and planned underwater / ice neutrino telescopes @xcite,@xcite,@xcite,@xcite , @xcite,@xcite,@xcite,@xcite is detection of neutrino of extraterrestrial origin .
but neutrino is neutral weakly interacting particle and one can not observe it directly but has to detect it by means of lepton that appears in @xmath0 interaction and passes a distance in medium before being detected .
the case with @xmath1 is considered , first of all , because muons possess the best ability to propagate large distances in contrast to @xmath2 ( due to their short radiation length ) and @xmath3 ( due to their short life time ) in a wide range of energies . on the other hand , atmospheric muons
( that represent a penetrating component of atmospheric showers resulting from interaction of primary cosmic radiation with the earth atmosphere ) are the principal background for neutrino signal . to eliminate it one has to predict correctly the detector response on atmospheric muon flux .
also one should keep in mind the fact that atmospheric muons are of scientific interest themselves : for instance , from the point of view of charm production @xcite . the last ( but not the least ) important
point is detector calibration which has to be done , again by means of atmospheric muons that are the only more or less known intensive calibration source for deep underwater / ice neutrino telescopes .
thus , when analyzing experimental data obtained with underwater / ice detectors one has to apply some model for muon transport through thick layers of matter , using an algorithm that gives the muon energy @xmath4 at the end of distance @xmath5 providing that its initial energy is equal to @xmath6 .
commonly , monte carlo ( mc ) technique is used for this purpose @xcite , @xcite,@xcite,@xcite,@xcite,@xcite being the most adequate to essentially stochastic nature of muon energy losses .
such a mc algorithm necessarily incorporates some uncertainties into the final result .
these uncertainties can be divided into two parts : * insurmountable uncertainties which relates to * * finite accuracy of formulae for muon cross - sections ; * * data on medium density and composition .
* `` inner '' errors that are produced by simulation algorithm itself due to * * finite accuracy of numerical procedures ; * * simplifications that are done to get reasonable computation time . as a function of @xmath7 as measured and parametrized by zeus collaboration ( markers and dashed line , correspondingly ) , and parametrized by bezrukov - bugaev ( solid line ) . , title="fig:",width=302 ] [ fig : sigma_exp ] ( solid line ) and ( b ) zeus parameterization ( dashed line )
. , title="fig:",width=302 ] [ fig : enlopn ] in case @xmath8 an algorithm is obviously too inaccurate and corrupts result more than uncertainties with the muon cross - sections and medium composition .
@xmath9 is better but leads to unreasonable wasting of human and computer efforts , the tool for the muon transport is too `` thin '' comparing to insurmountable uncertainties of group @xmath10 .
thus , the case @xmath11 seems to be an optimum equilibrium .
this report is aimed to get some quantitative parameters of muon transport algorithm which condition @xmath11 leads to .
the accuracy of existing formulae for muon cross - sections was analyzed in number of works @xcite,@xcite,@xcite .
the conclusion is that it is not higher than @xmath121% for @xmath13 1 tev and becomes the worse the higher muon energy is .
we do not give here the complete analysis but consider only two examples that concern the muon photonuclear interaction which contributes the largest uncertainty to the total muon energy losses comparing with other kinds of muon interactions that represent purely electromagnetic ( and hence , better known ) processes : ionization , bremsstrahlung , and direct @xmath14-pair production .
for last twenty years one has been using the formula for photonuclear muon cross - section that was developed in the frame of generalized vector dominance model ( gvdm ) @xcite .
in such an approach the differential cross - section ( and , consequently , the muon energy losses due to photonuclear interactions ) is proportional to the total cross - section for absorption of a real photon by a nucleon , @xmath15 .
there are several parameterizations for @xmath15 that were obtained by trying to get the best fit to experimental data @xcite,@xcite,@xcite,@xcite , @xcite .
two of them are shown in fig.1 along with the last results from the zeus experiment @xcite .
@xmath15 is presented as a function of @xmath7 , where @xmath16 is the center - of - mass energy of the @xmath17 system .
parameterizations @xcite ( bb ) and @xcite ( zeus ) both fit more or less the experimental data but even within `` experimentally known '' range @xmath18 10@xmath19 gev@xmath20 the difference between two parameterizations reaches 4% . at @xmath21
= 10@xmath22 gev@xmath20 the bb curve is higher than the zeus one by 15% .
the same data are presented in fig.2 in terms of muon energy losses due to photonuclear interaction in pure water .
difference is kept within several percents below @xmath23 100 tev and then grows up to @xmath12 10% at @xmath24 = 1 eev .
parameterizations for @xmath15 published in @xcite,@xcite are lower comparing to zeus parameterization since they are based on old results which differ remarkably from the latest zeus data . on the other hand , the recent parameterization @xcite is higher comparing to bb .
-pair production @xcite,@xcite,@xcite,@xcite ( dotted lines ) are shown , as well .
solid lines stand for total energy loss ( brem + pair + nuc , ionization not included ) with- @xcite,@xcite and without @xcite accounting for qcd part in photonuclear interaction .
, title="fig:",width=302 ] [ fig : qcd ] it became clear in last few years that at very high lepton energies the essential part of photonuclear cross - section is due to nongvdm contribution . it appears , that this additional part is well described by qcd perturbation theory .
recent new calculations @xcite , @xcite , @xcite show an essential increase of the muon energy losses due to photonuclear interactions . in @xcite the allm formulae @xcite ( that is based on regge approach and on h1 and zeus data )
is used for parameterization of nucleon structure function @xmath25 . in @xcite
the ckmt regge model @xcite is applied for a description of @xmath25 at low and intermediate @xmath26 ( squared four - momentum transfer ) and the fit of parton distribution functions given by mrs group @xcite at high @xmath26 .
authors of @xcite use for calculation of qcd perturbative part the color dipole model in version of @xcite , with parameters founded from latest desy data .
.differences in terms of the muon energy losses in standard rock due to photonuclear interaction ( in percents ) between three works that account for qcd part @xcite,@xcite,@xcite and bezrukov - bugaev formula based on gvdm @xcite . [ cols="^,^,^,^",options="header " , ] the results of these three works are presented in fig.3 and tables 1,2 in terms of the muon energy losses in standard rock . @xcite and @xcite are close to each other while the muon energy losses obtained in @xcite are essentially lower .
we believe that this is due to the fact that old data @xcite were used in @xcite to fit the parameters of model .
anyway , the total muon energy losses become higher when accounting for perturbative qcd part in photonuclear interactions by 1@xmath272% at @xmath24 = 100 tev and by 20@xmath2730% at @xmath24 = 1 eev -leptons as photonuclear interaction dominates in @xmath28 s energy losses at @xmath29 100 tev ] .
fig.4 shows changes in muon survival probabilities in sea water and standard rock which incorporation of nongvdm qcd - corrections results in and @xmath28-lepton transport @xcite that is available by request to authors . ] .
it is seen that for extremely high energies ( @xmath30 1 pev ) effect is very significant .
[ fig : tau2 ] energy losses are proportional , in particular , to medium density .
the density of water ( ice ) is known or , in any case , can be measured precisely .
but since the main technique to select the neutrino induced events is looking for up - going muons from the lower hemisphere , one should know also the density and composition of sea bed rock .
when this is unknown one has to use a standard rock model ( @xmath31 = 2.65 g/@xmath32 , @xmath33 = 22 , and @xmath34 = 11 ) . at the same time the measured rock densities at gran sasso and frejus underground laboratories
are equal , for instance , to @xmath35 = 2.70 g/@xmath32 and @xmath36 = 2.74 g/@xmath32 , respectively @xcite,@xcite which is 2@xmath273% higher .
investigation of the baikal lake bed ( nt-200 neutrino telescope ) showed a complex structure with several layers of different composition and density that varies with increase the depth from @xmath37 = 1.70 g/@xmath32 to @xmath37 = 2.90 g/@xmath32 ( @xmath3825% variations ) @xcite .
thus , error that is incorporated to muon transport calculations by uncertainties with the rock density / composition lies , at least , at a few percents level . all this have led us to a conclusion that we never know the muon energy losses better than with 1%-accuracy ( actually , worse ) and , consequently it makes no sense to waste efforts trying to reproduce energy losses with accuracy better than 10@xmath39 with an algorithm for the muon transport .
so , there is a reasonable quantitative criterium for a muon propagation algorithm goodness : algorithm whose accuracy is better than 1% can be accepted as a good one , otherwise it is bad .
among all simplifications that may be done when developing an algorithm for the muon propagation ( and that affect the algorithm accuracy ) one is obligatory .
as number of muon interactions per unit of path is practically infinite it is impossible to simulate all acts and one has to set some threshold for relative energy transfer @xmath40 ( @xmath41 = @xmath42 , where @xmath43 is energy which is passed by muon of energy @xmath24 to either real or virtual photon in a single interaction ) above which muon energy losses are treated by the direct simulation of @xmath44 for each interaction ( `` stochastic '' part ) and below which energy losses are treated by means of stopping - power formula ( `` soft '' part ) that can be obtained by integration of differential cross - sections for the muon interactions : @xmath45 vs @xmath40 obtained as a result of mc simulation for monoenergetic muon beams with initial energies @xmath46 = 1 tev ( 1st column of plots ) , 10 tev ( 2nd column ) , 100 tev ( 3rd column ) and 10 pev ( 4th column ) .
each column contains six plots which correspond to six depths @xmath5 .
see text for more details .
, title="fig:",width=328 ] [ fig : fig1 ] ( @xmath33 is a mass number of the target nucleus , @xmath47 is the avogadro number ) .
number of interactions to be simulated per unit of muon path depends on @xmath40 , growing roughly as @xmath48 along with computation time .
so , it would be desirable to choose @xmath40 as large as possible .
but , on the other hand , increase of @xmath40 affects the simulation accuracy since some part of statistical fluctuations in energy losses goes out of simulation and is not accounted for .
thus , the question is _ how large value of @xmath40 may be chosen to keep result within 1%-accuracy _ ?
this problem was discussed in literature @xcite , @xcite , @xcite , @xcite but , in our opinion , more careful analysis is lacking .
let s firstly consider propagation of monoenergetic muon beams .
fig.5 shows results on propagation ( survival probabilities @xmath49 vs @xmath40 ) for four beams with initial energies 1 tev , 10 tev , 100 tev and 10 pev through slant depths 3.2 km , 12 km , 23 km and 40 km , respectively , in pure water .
closed circles represent results obtained with knock - on electrons ( that appear as a result of ionization with large energy transferred to an atomic electron ) included in direct simulation , open circles stand for completely continuous ionization that is treated simply by bethe - bloch formula .
solid lines on each panel correspond to @xmath40 = 10@xmath50 and all energy losses multiplied by factors 0.99 ( upper line ) and 1.01 ( lower line ) , thus these lines limit `` @xmath381%-energy - losses - error - band '' .
horizontal dotted lines correspond to @xmath40 = 10@xmath50 and cross section for absorption of a real photon by a nucleon parametrized according to @xcite instead of parameterization @xcite that was mainly applied at presented simulations .
one can conclude as follows : * in most cases with except for lower row and the left column ( that corresponds to low survival probabilities and low muon initial energies , respectively ) survival probabilities lie mainly within @xmath381%-band .
* the difference between survival probabilities for two models of ionization - partially fluctuating ( above @xmath40 ) and completely `` soft '' - is the less appreciable the larger muon energy is .
* parameterizations @xcite and @xcite do not differ noticeable from each other in terms of survival probabilities . * for @xmath51 [email protected] there is almost no dependence of survival probability on @xmath40 with except for very last part of muon path where survival probability becomes small .
now let s go to the more realistic case with atmospheric muons .
fig.6 represents intensities of vertical atmospheric muon flux at 8 depths @xmath5 of pure water as functions of @xmath40 as obtained by simulation with muons sampled according to sea level spectrum @xcite : @xmath52 meaning of closed and open circles , as well as solid and dotted lines is the same as for fig.5 .
dashed lines on panels for @xmath53 5 km correspond to intensities which were computed for all energy losses treated as completely `` continuous '' ( no fluctuations included ) .
dash - dotted lines show intensities of vertical muon flux simulated with muons sampled according to the gaisser sea level spectrum @xcite : @xmath54 as obtained by mc simulation .
see text for more details .
, title="fig:",width=328 ] [ fig : fig2 ] general conclusions are qualitatively the same as for mono - energetic muon beams but quantitatively the influence of @xmath40 and model of ionization is much weaker
. one can conclude the following : * computed muon flux is strongly affected by accounting for fluctuations in energy losses : muon flux intensity computed by means of stopping - power formula is less comparing with mc simulated flux by @xmath1210 % at 3 km and by @xmath12 20% at 5 km . at the depth of 20 km muon flux computed with ignorance of fluctuations is only 10 % of simulated flux . *
1%-uncertainty in muon cross sections plays the principal role for resulting error ( all simulated data lie within @xmath381%-band with no exceptions ) .
this error has a tendency to grow with depth from @xmath382.5% at 1 km to @xmath5515% at 20 km .
* difference between muon spectra @xcite and @xcite leads to uncertainty from at 1 km to + 16% at 20 km . * error which appears due to simplified , entirely `` continuous '' ionization lies , commonly , at the level of 2@xmath273% . *
dependence of simulated muon flux intensity upon @xmath40 is the most weak one comparing with other error sources .
thus , we can make the following conclusions : to reproduce muon energy losses with an accuracy better than 10@xmath39 it is quite enough to account only for fluctuations in energy losses with fraction of energy lost being as large as @xmath56 [email protected] ( commonly , values @xmath40
= 10@xmath5710@xmath39 are incorporated in majority of algorithms for the muon propagation by now @xcite , @xcite , @xcite ) .
at least , in case with atmospheric muons only radiative losses may be simulated while ionization may be treated as an entirely continuous process without any lost of accuracy . in spite of remarkable dependency of simulated survival probabilities upon accounting or not accounting for fluctuations in ionization ( see fig.5 )
it seems to be correctly to treat ionization by completely continuous way also for neutrino - produced muons , as ( a ) differences in survival probabilities appear only for relatively low energies ; ( b ) they are of few percents only ; ( c ) they are of both signs and so , are expected to be partially self - compensated ; ( d ) the uncertainties with @xmath58 cross - sections are still much larger , in any case @xcite,@xcite,@xcite,@xcite . with such parameters
there are only several muon interactions to be simulated per 1 km w.e .
_ , music _ [ version for pure water with bremsstrahlung cross - sections by kehlner - kokoulin - petrukhin , 04/1999 ] _ , and mum _ [ version 1.4 , 12/2001 ] _ in pure water .
the energy range 25 gev@xmath271 pev is divided into two parts to keep linear scale : the upper plot stands for energy losses vs @xmath24 = 25 gev@xmath271 tev , the lower one does for energy losses divided by energy vs @xmath24 = 1 tev@xmath271 pev . the model for propmu energy losses
is taken from @xcite as referenced in @xcite .
curves for mum s and music s energy loss models are very close to each other and therefore are hardly distinguished .
, title="fig:",width=328 ] [ fig : water ] _ and 2 versions of propmu _ [ versions 2.01 , 18/03/93 and 2.1 , 01/2000 ] _ with different values of @xmath40 and different random generators . ,
title="fig:",width=328 ] [ fig : water_add ] we would like to emphasize that uncertainties with atmospheric muon flux intensity that are caused by only @xmath381% variations in energy losses are much higher than 1% ( see item ( b ) above ) .
this is due to sharp power - law shape of surface atmospheric muon spectrum and due to the fact that source of atmospheric muons is far away from an underwater / ice detector and their flux may only decrease when passing from the sea level down to detector depth .
the source of muons that are produced in @xmath59 interactions is uniformly distributed over water / ice and/or rock both out- and inside the array .
so , uncertainties with energy losses should affect the results for neutrino - generated muons weaker .
indeed , intensity of the muon flux which accompanies the neutrino flux in a medium is proportional to the muon range that , in turn , is inversely proportional to energy losses and , consequently , an error in energy losses leads to an equal error with an opposite sign in simulated flux of muons which are born in @xmath58-interactions .
any muon mc propagation algorithm consists of a set of procedures on numerical solution of equations , interpolation and integration .
all these procedures are of finite accuracy and , consequently , the incoming model for muon interactions is somewhat corrupted by them .
having a set of formulae for the muon cross - sections for bremsstrahlung , @xmath14-pair production , photonuclear process , and knock - on electron production it is easy to obtain the formula for the total averaged muon energy losses by integration of differential cross - sections over all range of kinematically allowed @xmath41 : @xmath60 but energy losses as they are simulated by an algorithm will differ from ( 4 ) due to `` corrupting '' application of numerical procedures .
two algorithms with the same formulae for cross - sections at the input will reproduce the muon energy losses differently .
the difference between simulated and calculated energy losses collects all errors that are contributed by each step of algorithm and thus , is a good quantitative criterium for its _ inner accuracy _ , whose contribution to the resulting error must not exceed 1% , as it was shown in the sec .
= 2.65 g/@xmath61 , @xmath33 = 22 , and @xmath34 = 11 ) , mum _ [ version 1.4 , 12/2001 ] _ and propmu _ [ version 2.1 , 01/2000]_. , title="fig:",width=328,height=381 ] [ fig : strock ] 1 km , 3 km , 6 km and 10 km , as simulated with propmu , music and mum for atmospheric muons sampled according to sea - level spectrum @xcite . for a comparison , parameterizations @xcite ( kbs ) and @xcite ( okada )
are given , as well .
on panel for @xmath62 1 km the spectrum which resulted from simulation with mum with all cross - sections multiplied by 0.9 is also presented ( mum@xmath63 , open circles ) . , title="fig:",width=328 ] [ fig : dif1 ] the trivial test may be suggested to check inner accuracy of any muon transport algorithm : let s simulate propagation of @xmath64 muons ( we used @xmath64 = 10@xmath65 for test reported below ) with energy @xmath66 over a short distance
@xmath67 `` measuring '' the final energy @xmath68 for each @xmath69th muon at the end .
then , simulated energy loss may be computed as @xmath70 comparing ( 5 ) with ( 4 ) it is possible to determine a goodness of the given algorithm applying , for example , criterium @xmath71 1% .
fig.7 , fig.8 , and fig.9 show results of such a comparison for three muon transport codes : propmu @xcite , music @xcite , and mum @xcite . results for pure water are presented in fig.7 . in case with
mum the difference @xmath72 never exceeds 1% , for music it reaches 2@xmath273% .
inaccuracy of the propmu code is much worse and in a range @xmath13 10 tev is up to 25% .
we try 2 versions of propmu with different values of @xmath40 and different random generators ( fig.8 ) . in some cases
inner accuracy becomes better but , in any case it stays worse than 10% that is unacceptable according to conclusion of sec.2 .
fig.9 reports the results of accuracy test for mum and propmu with standard rock .
qualitatively , these results are the same . in fig.10 differential spectra for atmospheric muons at different depths in pure water are presented as simulated with propmu , music and mum .
muons at the sea level were sampled according to the same spectrum @xcite . also parameterizations @xcite and @xcite are shown . simulations with music and mum give practically the same results because both codes reproduce almost the same energy losses ( see fig.7 ) .
music s and mum s spectra coincide with parameterization @xcite that is based on the same sea level muon spectrum as was used for simulation presented on the plot and on muon propagation with mum .
parameterization @xcite is lower than kbs , mum and music ( up to 18% in terms of integral muon flux at @xmath62 1 km ) at relatively shallow depths and becomes higher starting with @xmath73 5 km .
the reason is that it is based on very hard sea - level muon spectrum with index @xmath74 2.57 @xcite , while kbs parameterization adopts a spectrum with @xmath74 2.72 . as a result , there are less muons with relatively low energies ( which contribute the main bulk to the muon flux at low depths ) and more muons of high energies which contribute to the muon flux at larger depths . the reduced muon energy losses in propmu result in the depth muon spectra which ( a ) are significantly higher ( 31% , 30% , 27% and 17% in terms of integral muon flux at the depths @xmath62 1 km , 3 km , 6 km and 10 km , correspondingly ) and ( b ) are expanded to the low energies .
the spectrum for @xmath62 1 km was simulated also with `` corrupted '' version of mum : all muon cross - sections along with bethe - bloch ionization formula were multiplied by a factor 0.9 , thus energy losses were reduced by 10% and became close to ones that are reproduced by propmu 1 tev at the sea level . for this range propmu
reduces the simulated energy losses by @xmath75 10% as can be seen in upper panel of fig.7 . ] .
the result is given on upper left panel of fig.10 by open circles ( mum@xmath63 ) .
a good agreement between propmu and mum@xmath63 represents a good quantitative cross - check for results on propmu inner accuracy .
the optimum value for inner accuracy of a mc code for the muon transport is close to 1% . in any case we never know the muon energy losses better . to keep such inner accuracy it is quite enough to simulate the muon interactions with a fractions of energy lost that are as large as @xmath76 [email protected] , treating the rest of energy losses ( soft part ) by means of stopping - power formula .
such a test has been applied to three muon transportation codes . according to `` 1% goodness criterium '' codes mum and
music can be accepted as accurate enough .
accuracy of the propmu algorithm is insufficient ( down to 25% ) and depends on medium , version of code and simulation parameters .
it leads , for example , to @xmath1230% overestimation of atmospheric muon fluxes at depths 1@xmath275 km w.e .
99 aarnio , p. a. , _ et al .
_ , preprint cern - tis - rp/106 , 213pp . , 1983 .
abramowicz , h. _ et al .
_ , phys.lett.b 269 , 465 , 1991 ; abramowicz , h. , levy , a. , [ hep - ph/9712415 ] .
agrawal , v. _ et al .
_ , phys.rev.d 53 , 1314 , 1996 [ hep - ph/9509423 ] .
ahrens , j. _ et al .
_ , proc . of 9th intern
symp . on neutrino telescopes , venice , 2001 , v.2 , p.569 .
ambrosio , m. _ et al .
_ , phys.revd 52 , 3793 , 1995 .
andreev , yu . m. _ et al .
_ , phys.at.nucl .
57 , 2066 , 1994 .
antonioli , p. _ et al .
_ , astropart.phys .
7 , 357 , 1997 [ hep - ph/9705408 ] .
aslanides , e. _ et al .
_ , astro - ph/9907432 .
balkanov , v. a. _ et al .
_ , nucl.phys.proc.suppl .
91 , 438 , 2000 [ astro - ph/0011313 ] .
barwick , s. , _ et al .
wisconsin university report no .
mad - ph-629 , 1991 .
bezrukov , l. b. , bugaev , e. v. , sov.j.nucl.phys .
32 , 847 , 1980 ; ibid .
33 , 635 , 1981 .
bezrukov , l. b. , bugaev , e. v. , proceedings of the 17th icrc , paris , 1981 , v.7 , p.102 .
breitweg , j. _ et al .
_ , europ.phys.j.c 7 , 609 , 1999 .
bugaev , e. v. , shlepin , yu . v. , to be published .
butkevich , a. v. , mikheyev , s. p. , hep - ph/0109060 .
capella , a. _ et al .
_ , phys.lett.b 337 , 358 , 1994 [ hep - ph/9405338 ] .
carmona , e. _ et al .
_ , nucl.phys.proc.suppl .
95 , 161 , 2001 .
donnachie , a. , landshoff , p. v. , phys.lett.b 296 , 227 , 1992 [ hep - ph/9209205 ] .
dutta , s. i. _ et al .
_ , phys.rev.d63 , 094020 , 2001 [ hep - ph/0012350 ] .
forshaw , j. r. _ et al _
, phys.rev.d 60 , 074012 , 1999 [ hep - ph/9903341 ] .
gaisser , t. k. , _ cosmic rays and particle physics _
( cambridge university press , cambridge , 1990 ) .
gaisser , t. , these proceedings .
grieder , p. k. f. _ et al .
_ , nuovocim . 24c , 771 , 2001 .
kelner , s. r. _ et al .
_ , moscow engineering physics inst .
preprint 024 - 95 , 1995 ; phys.at.nucl . 60 , 576 , 1997 .
kelner , s. r. , moscow engineering physics inst .
preprint 016 - 97 , 1997 .
kelner , s. r. _ et al .
_ , phys.atom.nucl .
62 , 1894 , 1999 .
klimushin , s. i. _ et al .
_ , phys.rev.d 64 , 014016 , 2001 [ hep - ph/0012032 ] .
kokoulin , r , p. , petrukhin , a. a. , proc . of the 11th icrc , budapest , 1969 [ acta phys.acad.sci.hung .
29 , suppl . 4 , 277 , 1970 ] .
kokoulin , r , p. , petrukhin , a. a. , proc . of the 12th icrc , hobard , 1971 , v.6 , p.a2436 .
kokoulin , r. p. , petrukhin , a. a. , in proc . of the 22nd
icrc , dublin , 1991 , v.4 , p.536 .
kokoulin , r. p. , nucl.phys.b(proc . suppl . )
70 , 475 , 1999 .
kudryavtsev , v. a. _ et al .
_ , phys.lett.b 471 , 251 , 1999 [ hep - ph/9911493 ] .
lagutin , a. a. _ et al .
altayskogo gos .
universiteta , special issue , 93 , 1998 .
lipari , p. , stanev , t. , phys.rev.d 44 , 3543 , 1991 .
lipari , p. , these proceedings .
lohmann , w. _ et al .
_ , cern report no .
85 - 03 , 1985 .
martin , a. d. _ et al .
_ , eur.phys.j.c 14 , 133 , 2000 [ hep - ph/9907231 ] .
miyake , s. , in proc . of the 13th icrc , denver , 1973 , v.5 ,
naumov , v. a. _ et al .
_ , phys.atomnucl .
57 , 412 , 1994 .
naumov , v. a. , these proceedings .
okada , a. , astropart.phys . 2 , 393 , 1994 .
panfilov , a. , private communication .
resvanis , l. k. _ et al .
_ , nucl.phys.proc.suppl .
35 , 294 , 1994 .
rhode , w. , thesis wub - dis 93 - 11 , wuppertal university , 232 pp . , 1993 .
rhode , w. , crloganu , c. , proc . of the workshop on simulation and analysis methods for large neutrino telescopes , zeuthen , 1998 , p.247 .
rhode , w. , chirkin , d. , these proceedings .
sinegovskaya , t. s. , sinegovsky , s. i. , phys.rev.d 63 , 096004 , 2001 [ hep - ph/0007234 ] .
sokalski , i. , spiering , ch .
( eds . ) , _ nt-200 project description _
, baikal collaboration report 92/11 , 1992 .
sokalski , i. a. _ et al .
_ , phys.rev.d64 , 074015 , 2001 [ hep - ph/0010322 ] . | chain of calculations which have to be performed to predict any kind of signal in a deep underwater / ice neutrino detector necessarily includes the lepton propagation through thick layers of matter , as neutrino can be observed only by means of leptons ( muons , first of all , due to their large ranges ) that are generated in @xmath0 interactions .
thus , the muon propagation plays a key role when analyzing data and it is important to understand clearly how transportation part of simulation chain contributes to total inaccuracy of final results .
here we consider sources of uncertainties that appear in monte carlo algorithms for simulation of muon transport .
the trivial but effective test is proposed to measure the propagation algorithm accuracy .
the test is applied to three mc muon transport codes ( propmu , music , mum ) and results are reported . |
current thinking considers two major pathways to the formation of spiral bulges .
simplistically , either the bulge formed before the disk ( `` bulge - first , '' e.g. van den bosch 1998 ) , or formed _ from _ the disk ( `` disk - first , '' e.g. , combes & sanders 1981 ) .
previous studies have shown that bulges of big - bulge spirals ( like m31 ) share at least some stellar population properties with mid - sized elliptical galaxies .
they fall along the @xmath0@xmath1 relation ( dressler 1987 ) and the fundamental plane ( bender , burstein & faber 1992 ) . moreover , jablonka et al . ( 1996 ) and idiart et al . (
1996 ) find that bulges of spirals ( as late as sc ) fall along the mg@xmath1 relation defined by early - type galaxies , suggesting that bulges share a mass - metallicity relation with elliptical galaxies .
together with kinematical evidence , these observations have led to the idea that mid - sized ellipticals accrete disks from some leftover gas reservoir , forming spiral galaxies ( see , e.g. , kauffmann , white & guiderdoni 1993 ) .
therefore , the stars in spiral galaxy bulges should be similar to the stars in ellipticals ( i.e. metal - rich , high [ mg / fe ] ; e.g. , worthey , faber & gonzlez 1992 ; trager et al.1998b ) and unlike the stars in spiral galaxy disks .
kinematical and surface brightness observations of small - bulge spirals indicate that their bulges share many properties with their disks : many bulges in these spirals are better represented by a shallower exponential profile than by the steep de vaucouleurs profile ( e.g. , de jong 1996 ) and many of these bulges exhibit disk - like kinematics ( see kormendy 1993 for an excellent review ) .
moreover , peletier & balcells ( 1996 ) and de jong ( 1996 ) find that the color difference between bulge and disk in an individual galaxy is much smaller than the variation in colors across galaxies of a single hubble type .
these observations have led these authors and others to propose that bulges ( of small - bulge spirals at least ) are formed from the stars already present in their underlying disks .
that is , the stars in spiral galaxy bulges should be similar to the stars in spiral galaxy disks ( e.g. , ages and metallicities of their inner disks , solar [ mg / fe ] ) and unlike stars in ellipticals .
these two scenarios do have testable consequences , and sophisticated techniques now available can provide definitive answers to these questions .
we have embarked on a multi - year survey to probe the stellar content of the bulges and inner disks of spirals using absorption - line strengths .
these measurements can provide the critical tests needed to understand the basic mechanisms driving bulge formation .
we have selected a sample of 91 southern face - on spirals , ranging in type from s0/a to sdm , barred and unbarred , from the eso ( b ) survey .
these spirals are not interacting , have major axes larger than 3 , are within 4000 @xmath2 and are located at @xmath3 .
the observations have been made using the long - slit boller & chivens spectrograph at las campanas observatory in the blue , giving @xmath4 ( @xmath5 ) resolution in the 40005200 region .
this spectral range covers portions of both the lick / ids ( burstein et al .
1984 ; worthey et al .
1994 ; trager et al .
1998a ) and rose ( 1985 , 1994 ) absorption - line strength systems . to date , ten ( primarily unbarred ) spirals have been observed , most along both major and minor axes , and data from four have been processed .
line - strength profiles for these four galaxies have been calibrated onto the lick / ids system using stellar observations in common with jones ( 1996 ) , and velocity dispersion profiles and rotation curves have been derived following the fourier - quotient procedure described by gonzlez ( 1993 ) . as described above
, bulge - first formation mechanisms find support in the observation that bulges in galaxies as late as sc fall along the same mg@xmath6@xmath7 relation as early - type galaxies ( jablonka et al .
1996 ; idiart et al .
1996 ) , suggesting the existence of a mass - metallicity relation in spirals and a narrow spread in ages at fixed @xmath7 .
we confirm these observational results : our bulges fall on the mg@xmath8@xmath7 relation defined by early - type galaxies ( our observations do not cover the red sideband of the broad mg@xmath6 index ; see data in trager et al .
however , the mg@xmath1 relation is inherently _ degenerate to compensating variations in metallicity and age in old stellar populations_large age spreads can exist if there is a complementary age - metallicity relation ( worthey , trager & faber 1996 ; trager 1997 ) .
this intrinsic age - metallicity degeneracy in mg@xmath8 and mg@xmath6 ( and other metal lines and broadband colors ) prevents the mg@xmath1 relation from being an effective stellar population age discriminator .
more sophisticated techniques are obviously necessary to distinguish between bulge formation mechanisms .
there is a way to break the age - metallicity degeneracy
balmer lines are more age - sensitive than metal - line indices ( with the possible exception of the g band ; worthey 1994 ) . in the absence of of nebular emission or large numbers of blue horizontal branch stars or
blue stragglers , balmer - line strengths reflect light - weighted mean turnoff temperature of the composite stellar population i.e . , the mean age of the population .
balmer lines do have some intrinsic metal dependence , however , so a clean separation of age and metallicity effects requires diagnostic diagrams combining balmer line and metal line strengths .
figure [ bmd ] presents such diagrams for the first four galaxies analyzed in our sample . the balmer metal - line diagrams for large bulges ( sab sbc , @xmath9 ) suggest that the bulges of these early - type spirals are consistent with having old , metal - rich stellar populations , as seen in our own galaxy ( @xmath10 gyr , @xmath11}\gtrsim0 $ ] ; e.g. , mcwilliam & rich 1994 ; bruzual et al .
these bulges are also older than their inner disks by several gyr , suggesting that these bulges formed early on in the galaxies histories , and that bulge - first formation is a likely scenario for these galaxies .
on the other hand , small bulges ( sc , @xmath12 ) seem to have younger and more metal - poor populations ( @xmath13 gyr , @xmath14}\lesssim0 $ ] ) , and at least in some galaxies ( ngc 1637 ) , the bulge and inner disk are roughly the same age .
such an observation provides strong support for a common origin of bulge and disk material that is , for a disk - first formation scenario
. however , contamination from emission is difficult to remove , separation of bulge and disk light has not yet been attempted , and small amounts of young or intermediate - age populations can significantly increase balmer - line strengths ( trager et al .
in these late - type , star - forming , small - bulged galaxies , such effects may obscure truly old bulge populations .
( 100,200)(50,30 ) ( 0,0 )
the initial indications are that large bulges are genuinely old , metal - rich systems , as expected from our own galactic bulge .
the stellar populations of small bulges appear to be younger and more metal - poor than the large bulges , but contamination from emission lines and disk light make these conclusions uncertain for the moment . if these results stand with more data and more extended analysis , we may find that bulges may have more than one formation mechanism or even more than one mechanism may occur in a single galaxy ( rix , this meeting ) .
we thank our collaborators dr .
r. o. marzke and dr .
a. mcwilliam for many stimulating converations during the course of this work .
sct would like to thank the organizers for holding this very interesting workshop and for their financial support .
1992 , _ apj _ , * 399 * , 462 1997 , _ aj _ , * 114 * , 1531 1984 , _ apj _ , * 287 * , 586 1981 , _ a&a _ , * 96 * , 164 1996 , _ a&a _ , * 313 * , 377 1987 , _ apj _ , * 317 * , 1 1993 , _ ph.d .
thesis , university of california , santa cruz _ 1996 , _
aj_,*112*,2541 1996 , _ aj _ , * 112 * , 1415 1996 , _ ph.d . thesis ,
north carolina _ 1993 , _ mnras _ , * 264 * , 201 1993 , in _ galactic bulges , iau symposium 153 _ , ed .
h. dejonghe & h. j. habing , p. 209
1994 , _ apjs _ , * 91 * , 749 1996 , _ aj _ , * 111 * , 2238 1985 , _ aj _ , * 90 * , 1927 1994 , _ aj _ , * 107 * , 206 1997 , _ ph.d .
thesis , uc santa cruz _ 1998a , _ apjs _ , * 116 * , 1 1998b , _ in prep .
_ 1998 , preprint ( astro - ph/9805113 ) 1994 , _ apjs _ , * 95 * , 107 1992 , _ apj _ , * 398 * , 69 1994 , _ apjs _ , * 94 * , 687 1996 , in _
fresh views on elliptical galaxies _ , ed .
a. buzzoni , a. renzini , & a. serrano , a. s. p. conf .
, vol . 86 ( san francisco : asp ) , p. 203
1997 , _ apjs _ , * 111 * , 377 | we present first results from an on - going survey of the stellar populations of the bulges and inner disks of spirals at various points along the hubble sequence .
in particular , we are investigating the hypotheses that bulges of early - type spirals are akin to ( and may in fact originally have been ) intermediate - luminosity ellipticals while bulges of late - type spirals are formed from dynamical instabilities in their disks .
absorption - line spectroscopy of the central regions of sa sd spirals is combined with stellar population models to determine integrated mean ages and metallicities .
these ages and metallicities are used to investigate stellar population differences both between the bulges and inner disks of these spirals and between bulges and ellipticals in an attempt to place observational constraints on the formation mechanisms of spiral bulges .
# 1#1 0123456789 abcdef = eurm10 = eurm10 at 7pt = eurm10 at 5pt = = = @ = `` 0@19 = ' ' 0@16 = msam10 = msam10 at 7pt = msam10 at 5pt = = = @ = `` 3@36 = ' ' 3@3e = = = = @mtlplain@loaded = cmssi10 = cmssi10 = cmssi10 = cmmib10 = mtssi10 = mtssbx10 = mtssbi10 = mtmib10 = " 0103 @mtlplain@loaded |
a general feature of high - energy scattering is that a fast particle moves along its straight - line classical trajectory and the only quantum effect is the eikonal phase factor acquired along this propagation path . in qcd , for the fast quark or gluon scattering off some target
, this eikonal phase factor is a wilson line - the infinite gauge link ordered along the straight line collinear to particle s velocity @xmath1 : @xmath2 here @xmath3 is the gluon field of the target , @xmath4 is the transverse position of the particle which remains unchanged throughout the collision , and the index @xmath5 labels the rapidity of the particle . repeating the above argument for the target ( moving fast in the spectator s frame ) we see that particles with very different rapidities perceive each other as wilson lines and therefore these wilson - line operators form the convenient effective degrees of freedom in high - energy qcd ( for a review , see ref .
@xcite ) . let us consider the deep inelastic scattering from a hadron at small @xmath6 .
the virtual photon decomposes into a pair of fast quarks moving along straight lines separated by some transverse distance .
the propagation of this quark - antiquark pair reduces to the `` propagator of the color dipole '' @xmath7 - two wilson lines ordered along the direction collinear to quarks velocity .
the structure function of a hadron is proportional to a matrix element of this color dipole operator @xmath8 switched between the target s states ( @xmath9 for qcd ) .
the gluon parton density is approximately @xmath10 where @xmath11 .
( as usual , we denote operators by `` hat '' ) .
the energy dependence of the structure function is translated then into the dependence of the color dipole on the slope of the wilson lines determined by the rapidity @xmath5 .
thus , the small - x behavior of the structure functions is governed by the rapidity evolution of color dipoles @xcite . at relatively high energies and for sufficiently small dipoles
we can use the leading logarithmic approximation ( lla ) where @xmath12 and get the non - linear bk evolution equation for the color dipoles @xcite : @xmath13 \label{bk}\end{aligned}\ ] ] the first three terms correspond to the linear bfkl evolution @xcite and describe the parton emission while the last term is responsible for the parton annihilation . for sufficiently high
@xmath14 the parton emission balances the parton annihilation so the partons reach the state of saturation@xcite with the characteristic transverse momentum @xmath15 growing with energy @xmath16 ( for a review , see @xcite ) as usual , to get the region of application of the leading - order evolution equation one needs to find the next - to - leading order ( nlo ) corrections . in the case of the small - x evolution equation ( [ bk ] ) there is another reason why nlo corrections are important . unlike the dglap evolution , the argument of the coupling constant in eq .
( [ bk ] ) is left undetermined in the lla , and usually it is set by hand to be @xmath15 .
careful analysis of this argument is very important from both theoretical and experimental points of view . from the theoretical viewpoint , we need to know whether the coupling constant is determined by the size of the original dipole @xmath17 or of the size of the produced dipoles @xmath18 and/or @xmath19 since we may get a very different behavior of the solutions of the equation ( [ bk ] ) . on the experimental side ,
the cross section is proportional to some power of the coupling constant so the argument determines how big ( or how small ) is the cross section .
the typical argument of @xmath20 is the characteristic transverse momenta of the process . for high enough energies
, they are of order of the saturation scale @xmath15 which is @xmath21 gev for the lhc collider , so even the difference between @xmath22 and @xmath23 can make a substantial impact on the cross section . the precise form of the argument of @xmath20 should come from the solution of the bk equation with the running coupling constant , and the starting point of the analysis of the argument of @xmath20 in eq .
( [ bk ] ) is the calculation of the nlo evolution .
let us present our result for the nlo evolution of the color dipole ( hereafter we use notations @xmath24 , @xmath25 , @xmath26 , and @xmath27 ) @xmath28\big\ } ~[{\rm
tr}\{\hat{u}_x \hat{u}^{\dagger}_z\}{\rm tr}\{\hat{u}_z \hat{u}^{\dagger}_y\ } -n_c{\rm tr}\{\hat{u}_x \hat{u}^{\dagger}_y\ } ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2\over 16\pi^4 } \int \!d^2 zd^2 z ' \bigg [ \big(-{4\over ( z - z')^4}+\big\{2{x^2{y'}^2+{x'}^2y^2 - 4(x - y)^2(z - z')^2\over ( z - z')^4[x^2{y'}^2-{x'}^2y^2]}\nonumber\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big ) \nonumber\\ & & \hspace{52 mm } \times~[{\rm tr}\{\hat{u}_x\hat{u}^\dagger_z\}{\rm tr}\{\hat{u}_z\hat{u}^\dagger_{z'}\}{\rm tr}\{\hat{u}_{z'}\hat{u}^\dagger_y\ } -{\rm tr}\{\hat{u}_x\hat{u}^\dagger_z \hat{u}_{z'}u^\dagger_y\hat{u}_z\hat{u}^\dagger_{z'}\}-(z'\rightarrow z ) ] \nonumber\\ & & \hspace{-2 mm } + ~\big\{{(x - y)^2\over ( z - z')^2 } \big[{1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] -{(x - y)^4\over x^2{y'}^2{x'}^2y^2}\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } ~{\rm
tr}\{\hat{u}_x \hat{u}^\dagger_z\}{\rm tr}\{\hat{u}_z\hat{u}^\dagger_{z'}\}{\rm tr}\{\hat{u}_{z'}\hat{u}^\dagger_y\ } \nonumber\\ & & \hspace{-2 mm } + ~4n_f \big\{{4\over(z - z')^4 } -2{{x'}^2y^2+{y'}^2x^2-(x - y)^2(z - z')^2\over ( z - z')^4(x^2{y'}^2-{x'}^2y^2- ) } \ln{x^2{y'}^2\over { x'}^2y^2}\big\}{\rm tr}\{t^a\hat{u}_xt^b\hat{u}^{\dagger}_y\ } [ { \rm tr}\{t^a \hat{u}_zt^b \hat{u}^\dagger_{z'}\}-(z'\rightarrow z)]\bigg ] \nonumber\end{aligned}\ ] ] here @xmath29 is the normalization point in the @xmath30 scheme and @xmath31 is the first coefficient of the @xmath32-function .
the result of this paper is the gluon part of the evolution , the quark part of eq .
( [ nlobk ] ) proportional to @xmath33 was found earlier @xcite .
also , the terms with cubic nonlinearities were previously found in the large-@xmath34 approximation in ref .
the nlo kernel is a sum of the running - coupling part ( proportional to @xmath35 ) , the non - conformal double - log term @xmath36 and the three conformal terms which depend on the two four - point conformal ratios @xmath37 and @xmath38 .
note that the logarithm of the second conformal ratio @xmath39 is absent .
it should be emphasized that the nlo result itself does not lead automatically to the argument of coupling constant @xmath20 in eq .
[ bk ] . in order to get this argument one can use the renormalon - based approach@xcite :
first get the quark part of the running coupling constant coming from the bubble chain of quark loops and then make a conjecture that the gluon part of the @xmath32-function will follow that pattern .
( [ nlobk ] ) proves this conjecture in the first nontrivial order : the quark part of the @xmath32 - function @xmath40 calculated earlier gets promoted to full @xmath35 .
the analysis of the argument of the coupling constant was performed in refs .
@xcite and we briefly review it in sect . 7 for completeness . roughly speaking ,
the argument of @xmath20 is determined by the size of the smallest dipole @xmath41 .
the paper is organized as follows . in sect .
ii we remind the derivation of the bk equation in the leading order in @xmath20 . in sect .
iii and iv , which are central to the paper , we calculate the gluon contribution to the nlo kernel of the small-@xmath0 evolution of color dipoles : in sect .
iii we calculate the part of the nlo kernel corresponding to one - to - three dipoles transition and in sect .
iv the one - to - two dipoles part . in sect .
v we assemble the nlo bk kernel and in sect .
vi we compare the forward nlo bk kernel to the nlo bfkl results @xcite .
the results of the analysis of the argument of coupling constant are briefly reviewed in sect .
appendix a is devoted to the calculation of the uv - divergent part of the one - to - three dipole kernel and in appendix b we discuss the dependence of the nlo kernel on the cutoff in the longitudinal momenta .
before discussing the small - x evolution of color dipole in the next - to - leading approximation it is instructive to recall the derivation of the leading - order ( bk ) evolution equation . as discussed in the introduction , the dependence of the structure functions on @xmath14 comes from the dependence of wilson - line operators @xmath42 on the slope of the supporting line .
the momenta @xmath43 and @xmath44 are the light - like vectors such that @xmath45 and @xmath46 where @xmath47 is the momentum of the target and @xmath48 is the mass . throughout the paper ,
we use the sudakov variables @xmath49 and the notations @xmath50 and @xmath51 related to the light - cone coordinates : @xmath52 . to find the evolution of the color dipole ( [ fla1 ] ) with respect to the slope of the wilson lines in the leading log approximation we consider the matrix element of the color dipole between ( arbitrary ) target states and integrate over the gluons with rapidities @xmath53 leaving the gluons with @xmath54 as a background field ( to be integrated over later ) . in the frame of gluons with @xmath55 the fields with @xmath54
shrink to a pancake and we obtain the four diagrams shown in fig .
[ bkevol ] . technically , to find the kernel in the leading - ordrer approximation we write down the general form of the operator equation for the evolution of the color dipole @xmath56 ( where dots stand for the higher orders of the expansion ) and calculate the l.h.s . of eq .
( [ eveq ] ) in the shock - wave background @xmath57 in what follows we replace @xmath58 by @xmath59 for brevity . evolution of color dipole[bkevol ] .
gauge links are denoted by dotted lines.,scaledwidth=110.0% ] with future nlo computation in view , we will perform the leading - order calculation in the lightcone gauge @xmath60 .
the gluon propagator in a shock - wave external field has the form@xcite @xmath61u^{ab } \big[g^{\perp\xi}_\nu -{2\over \alpha s}(p_2^\xi p^\perp_\nu+p_{2\nu}p_\perp^\xi ) \big ] e^{i{p_\perp^2\over\alpha s}y_\ast}|y_\perp ) \nonumber \\ & & \hspace{-2mm}-~\theta(-x_\ast)\theta(y_\ast ) \!\int_0^\infty\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}\alpha~ { e^{i\alpha(x - y)_\bullet}\over 2\alpha } ( x_\perp|e^{i{p_\perp^2\over\alpha s}x_\ast } \big[g^\perp_{\mu\xi}-{2\over \alpha s}(p^\perp_\mu p_{2\xi}+p_{2\mu}p^\perp_\xi)\big ] u^{\dagger ab } \big [ g^{\perp\xi}_{~\nu}-{2\over \alpha s}(p_2^\xi p^\perp_\nu+p_{2\nu}p_\perp^\xi)\big ] e^{-i{p_\perp^2\over\alpha s}y_\ast}|y_\perp ) \nonumber\end{aligned}\ ] ] where @xmath62 hereafter we use schwinger s notations @xmath63 ( the scalar product of the four - dimensional vectors in our notations is @xmath64 ) .
note that the interaction with the shock wave does not change the @xmath65-component of the gluon momentum .
we obtain @xmath66 formally , the integral over @xmath65 diverges at the lower limit , but since we integrate over the rapidities @xmath67 we get ( in the lla ) @xmath68 and therefore @xmath69 the contribution of the diagram in fig .
[ bkevol]b is obtained from eq .
( [ bk3 ] ) by the replacement @xmath70 , @xmath71 and the two remaining diagrams are obtained from eq .
[ bk2 ] by taking @xmath72 ( fig .
[ bkevol]c ) and @xmath73 ( fig .
[ bkevol]d ) . finally , one obtains @xmath74 so @xmath75^{\eta_2 } \label{bk5}\end{aligned}\ ] ] there are also contributions coming from the diagrams shown in fig .
[ virdiagrams ] ( plus graphs obtained by reflection with respect to the shock wave ) . ]
these diagrams are proportional to the original dipole @xmath76 and therefore the corresponding term can be derived from the contribution of fig . 1 graphs using the requirement that the r.h.s . of
the evolution equation should vanish for @xmath73 since @xmath77 ) .
it is easy to see that this requirement leads to @xmath78^{\eta_2 } \label{bk6}\end{aligned}\ ] ] which is equivalent to the bk equation for the evolution of the color dipole ( [ bk ] ) .
in the next - to - leading order there are three types of diagrams . diagrams of the first type have two intersections of the emitted gluons with the shock wave , diagrams of the second type have one intersection , and finally there are diagrams of the third type without intersections . in principle , there could have been contributions coming from the gluon loop which lies entirely in the shock wave , but we will demonstrate below that such terms are absent ( see the discussion at the end of sect . [
sect : compare ] ) . for the nlo calculation we use the lightcone gauge @xmath60 .
also , we find it convenient to change the prescription for the cutoff in the longitudinal direction .
we consider the light - like dipoles ( in the @xmath43 direction ) and impose the cutoff on the maximal @xmath65 emitted by any gluon from the wilson lines so @xmath79 \nonumber\\ & & \hspace{-2 mm } a^\eta_\mu(x)~=~\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^4 k \theta(e^\eta-|\alpha_k|)e^{-ik\cdot x } a_\mu(k ) \label{cutoff}\end{aligned}\ ] ] as we will see below , the ( almost ) conformal result ( [ nlobk ] ) comes from the regularization ( [ cutoff ] ) . in appendix b we will present the nlo kernel for the cutoff with the slope ( [ defy ] ) . . ]
we start with the calculation of the fig .
[ nlobk2]a diagram . multiplying two propagators ( [ gluprop ] ) , two 3-gluon vertices and two bare propagators we obtain @xmath80 where @xmath81 in this formula @xmath82 comes from the integration over @xmath83 parameter in the l.h.s . and
@xmath84 from the integration of the left three - gluon vertex over the half - space @xmath85 .
similarly , we get @xmath86 from the integration over @xmath87 parameter and @xmath88 from the integration of the right three - gluon vertex over the half - space @xmath89 .
the factor @xmath90 in the r.h.s .
is combinatorial .
note that in the light - cone gauge one can always neglect the @xmath91 components of the momenta in the three - gluon vertex since they are always multiplied by some @xmath92 .
taking residues at @xmath93 and @xmath94 , @xmath95 we obtain @xmath96 we have omitted terms @xmath97 in the arguments of @xmath92 since they do not contribute to @xmath98 , see eq .
( [ demunu ] ) . introducing the variable @xmath99 and taking residues at @xmath100 and @xmath101
we obtain @xmath102 where we have imposed a cutoff @xmath103 in accordance with eq .
( [ cutoff ] ) .
using formulas @xmath104 we can represent the contribution of diagram in fig .
[ nlobk2]a in the form @xmath105 \big[(k_1 ^ 2-k_2 ^ 2)\delta_{ij}-{2\over u}k_{1i}(k_1+k_2)_j+ { 2\over \bar{u}}(k_1+k_2)_ik_{2j}\big ] \nonumber\end{aligned}\ ] ] throughout the paper we use greek letters for indices @xmath106 ( with @xmath107=(1,-1,-1,-1 ) ) and latin letters for transverse indices @xmath108 .
the diagram shown in fig .
[ nlobk2]b is obtained by the substitution @xmath109 ( the different sign comes from the replacement @xmath110_y$ ] by @xmath111_x$ ] ) .
we get @xmath112 \over ( q_1+q_2)^2(k_1+k_2)^2(q_1 ^ 2\bar{u}+q_2 ^ 2u)(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \nonumber\\ & & \hspace{-2mm}\times~ \big[(q_1 ^ 2-q_2 ^ 2)\delta_{ij}-{2\over u}q_{1i}(q_1+q_2)_j+ { 2\over \bar{u}}(q_1+q_2)_iq_{2j}\big ] \big[(k_1 ^ 2-k_2 ^ 2)\delta_{ij}-{2\over u}k_{1i}(k_1+k_2)_j+ { 2\over \bar{u}}(k_1+k_2)_ik_{2j}\big ] \nonumber\end{aligned}\ ] ] . ]
next , consider the `` cut vertex '' diagram in fig .
[ nlobk3]a .
the analog of eq .
( [ selfenergy2 ] ) has the form : @xmath113 } \nonumber\\ & & \hspace{42 mm } \times \int\!d^2z d^2z ' { e^{-i(k_1+k_2,y)_{\perp}}u_z^{mb } u_{z'}^{nc}e^{-i(q_1-k_1,z)_\perp - i(q_2-k_2,z')_\perp } \over k_1 ^ 2(k_1 ^ 2\alpha_2+k_2 ^ 2\alpha_1 ) ( \alpha_1\beta_1 s - q_{1\perp}^2+i\epsilon ) ( \alpha_2\beta_2 s - q_{2\perp}^2+i\epsilon ) } \label{vertex1}\end{aligned}\ ] ] going to variables
@xmath114 , @xmath99 and taking residues at @xmath115 and @xmath116 we get @xmath117 and therefore @xmath118 where we have used the formula @xmath119 \label{vertex4}\end{aligned}\ ] ] following from eq .
( [ nice ] ) . the contribution of the diagram shown in fig .
[ nlobk3]b differs from eq .
( [ vertex3 ] ) by the substituion @xmath120 and changing the order of @xmath121 , @xmath122 matrices .
( similarly to the case of the fig .
( [ nlobk2])b diagram , this prescription follows from the replacement @xmath110_y$ ] by @xmath123_x$ ] but now we consider the second term of the expansion in the gauge field ) .
we get @xmath124u^\dagger_y\ } \nonumber \end{aligned}\ ] ] there is another type of diagrams with two gluon - shockwave intersections shown in fig .
[ nlobk4 ] . ]
@xmath125_\lambda \over ( \alpha_1+\alpha_2)(\beta_1+\beta_2-i\epsilon ) [ ( \alpha_1+\alpha_2)(\beta_1+\beta_2 ) s-(q_1+q_2)^2_{\perp}+i\epsilon]}~ { k_{1\mu}+{2(k_1,q_1)_{\perp}\over\alpha_1 s}p_{2\mu } \over \alpha_1\beta_1 s - q_{1\perp}^2+i\epsilon}~ { k_{2\nu}+{2(k_2,q_2)_{\perp}\over\alpha_2 s}p_{2\nu } \over \alpha_2\beta_2 s - q_{2\perp}^2+i\epsilon } \bigg\ } \nonumber\end{aligned}\ ] ] taking residues at @xmath115 and at @xmath126 and going to variables @xmath114 and @xmath99 we get @xmath127k_{1i}k_{2j } \nonumber\end{aligned}\ ] ] and therefore @xmath128 { k_{1i}k_{2j}\over k_1 ^ 2k_2 ^ 2}{\rm tr}\{t^au_xt^{m'}t^{n ' } u^\dagger_y\ } \nonumber\end{aligned}\ ] ] where again we have used formula ( [ vertex4 ] ) .
the sum of the contributions ( [ vkladnlobk2 ] ) , ( [ vkladnlobk3 ] ) and ( [ vertex7 ] ) can be represented as follows @xmath129u_x \nonumber\\ & & \hspace{-12 mm } \times~ \bigg[t^bf^{bm'n ' } { ( e^{-i(k_1+k_2,x)_\perp}-e^{-i(k_1+k_2,y)_\perp } ) \over ( k_1+k_2)^2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \big[(k_1 ^ 2-k_2 ^ 2)\delta_{ij}-{2\over u } k_{1i}(k_1+k_2)_j+ { 2\over\bar{u}}(k_1+k_2)_ik_{2j}\big ] \nonumber\\ & & \hspace{-12 mm } -~4ik_{1i}k_{2j}\bigg ( { t^{n'}t^{m'}e^{-i(k_1+k_2,x)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } -{e^{-i(k_1,x)_\perp - i(k_2,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{m'}t^{n ' } + ~{t^{m'}t^{n'}e^{-i(k_1+k_2,y)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \bigg)\bigg ] u^\dagger_y\big\ } \nonumber\end{aligned}\ ] ] if we add contribution of the diagrams with the gluon on the right side of the shock wave attached to the wilson line at the point @xmath130 instead of @xmath0 ( which differs from eq .
( [ partialsum ] ) by the substitution @xmath131 ) we obtain @xmath132u_x \nonumber\\ & & \hspace{-12 mm } \times~ \bigg[t^bf^{bm'n ' } { ( e^{-i(k_1+k_2,x)_\perp}-e^{-i(k_1+k_2,y)_\perp } ) \over ( k_1+k_2)^2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \big[(k_1 ^ 2-k_2 ^ 2)\delta_{ij}-{2\over u } k_{1i}(k_1+k_2)_j+ { 2\over\bar{u}}(k_1+k_2)_ik_{2j}\big ] \nonumber\\ & & \hspace{-12 mm } -~4ik_{1i}k_{2j}\bigg ( { t^{n'}t^{m'}e^{-i(k_1+k_2,x)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } -{e^{-i(k_1,x)_\perp - i(k_2,y)_\perp}\over \bar{u}uk_1 ^ 2k_2
^ 2}t^{m'}t^{n ' } + ~{t^{m'}t^{n'}e^{-i(k_1+k_2,y)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \bigg)\bigg ] u^\dagger_y\big\ } \nonumber\end{aligned}\ ] ] the result ( [ partialsum1 ] ) can be obtained from the self - energy contribution ( [ vkladnlobk2 ] ) by the replacement of the term corresponding to the emission of the two gluons via the three - gluon vertex @xmath133 \nonumber\end{aligned}\ ] ] with similar contribution containing the `` effective vertex '' @xmath134 \nonumber\\ & & \hspace{-12 mm } -~4ik_{1i}k_{2j}\bigg ( { t^{n'}t^{m'}e^{-i(k_1+k_2,x)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } -{e^{-i(k_1,x)_\perp - i(k_2,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{m'}t^{n ' } + ~{t^{m'}t^{n'}e^{-i(k_1+k_2,y)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \bigg)\label{effective1}\end{aligned}\ ] ] , scaledwidth=110.0% ] , scaledwidth=110.0% ] , scaledwidth=110.0% ] it can be demonstrated that the sum of the contributions of the diagrams shown in fig .
[ 2cutdms ] i , ... , iv , xi , ... , xvi can be obtained from the self - energy contribution ( [ vkladnlobk2 ] ) by replacing the gluon vertex @xmath135\nonumber\\\end{aligned}\ ] ] with similar `` effective vertex '' @xmath135 \nonumber\\ & & \hspace{-12 mm } + ~4iq_{1i}q_{2j}\bigg ( { t^mt^ne^{i(q_1+q_2,x)_\perp}\over\bar{u } q_1 ^ 2(q_1 ^ 2\bar{u}+q_2 ^ 2u ) } -{e^{i(q_1,x)_\perp+i(q_2,y)_\perp}\over \bar{u}uq_1 ^ 2q_2 ^ 2}t^nt^m + { t^nt^me^{i(q_1+q_2,y)_\perp}\over \bar{u}q_1 ^ 2(q_1 ^ 2\bar{u}+q_2 ^ 2u ) } \bigg)\label{effective2}\end{aligned}\ ] ] note that ( [ effective2 ] ) is equal to @xmath136 .
let us consider now the box diagrams topology shown in fig .
[ 2cutdms ] xvii - xxxiv .
the calculation of these diagrams is similar to the above calculation of `` cut self - energy '' and `` cut vertex '' diagrams
so we present here only the final result @xmath137u_x \nonumber\\ & & \hspace{-12 mm } \times~ \big[k_{1i}k_{2j}\bigg ( { t^{n'}t^{m'}e^{-i(k_1+k_2,x)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u)}+ { t^{m'}t^{n'}e^{-i(k_1+k_2,x)_\perp}\over uk_2 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } -{e^{-i(k_1,x)_\perp - i(k_2,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{m'}t^{n ' } -{e^{-i(k_2,x)_\perp - i(k_1,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{n'}t^{m ' } \nonumber\\ & & \hspace{52 mm } + ~{t^{m'}t^{n'}e^{-i(k_1+k_2,y)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } + { t^{n'}t^{m'}e^{-i(k_1+k_2,y)_\perp}\over uk_2 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u)}\bigg)\bigg ] u^\dagger_y\big\ } \nonumber\end{aligned}\ ] ] this expression agrees with the sum of `` box topology '' diagrams in ref . @xcite .
now we observe that each three - gluon vertex diagram is equal to its own cross diagram ( the same can not be said for box diagrams ) .
thus we may redefine the `` effective vertex '' ( [ effective1 ] ) in the following way @xmath138 \nonumber\\ & & \hspace{-12 mm } -~i2k_{1i}k_{2j}\bigg ( { t^{n'}t^{m'}e^{-i(k_1+k_2,x)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u)}+ { t^{m'}t^{n'}e^{-i(k_1+k_2,x)_\perp}\over uk_2 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } -{e^{-i(k_1,x)_\perp - i(k_2,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{m'}t^{n ' } -{e^{-i(k_2,x)_\perp - i(k_1,y)_\perp}\over \bar{u}uk_1 ^ 2k_2 ^ 2}t^{n'}t^{m ' }
\nonumber\\ & & \hspace{52 mm } + ~{t^{m'}t^{n'}e^{-i(k_1+k_2,y)_\perp}\over \bar{u}k_1 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } + { t^{n'}t^{m'}e^{-i(k_1+k_2,y)_\perp}\over uk_2 ^ 2(k_1 ^ 2\bar{u}+k_2 ^ 2u)}\bigg)\end{aligned}\ ] ] which corresponds to writing each contribution of the three - gluon vertex diagrams as a sum of two equal terms .
a similar expression can be written for the `` effective vertex '' ( [ effective2 ] ) and therefore the sum of all diagrams with two gluon - shockwave intersections can be written as @xmath139 \big [ { \cal s}^ { m'n'}(k_1,k_2;x , y)u^\dagger_y\big]\big\ } \label{twocutsum}\end{aligned}\ ] ] separating the contributions of different color structures one obtains @xmath140 \nonumber\\ & & \hspace{-2 mm } -~t^af^{amn}{q_{1i}q_{2j}\over q_1 ^
2q_2 ^ 2}(e^{i(q_1,x)+i(q_2,y')}-x\leftrightarrow y ) + i\{t^m , t^n\}{q_{1i}q_{2j}\over q_1 ^ 2q_2 ^ 2 } ( e^{i(q_1,x)}-e^{i(q_1,y)})(e^{i(q_2,x')}-e^{i(q_2,y')})\big\}u_x \nonumber\\ & & \hspace{-2 mm } \times~ \big\{t^bf^{bm'n'}{(e^{-i(k_1,x)-i(k_2,x')}-x\leftrightarrow y ) \over ( k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \big[{(k_1 ^ 2-k_2 ^ 2)\bar{u }
u\delta_{ij}-2\bar{u } k_{1i}(k_1+k_2)_j+ 2u(k_1+k_2)_ik_{2j}\over ( k_1+k_2)^2 } -u{k_{1i}k_{2j}\over k_1 ^ 2}+\bar{u}{k_{1i}k_{2j}\over k_2 ^ 2}\big ] \nonumber\\ & & \hspace{-2 mm } -~t^bf^{bm'n'}{k_{1i}k_{2j}\over k_1
^ 2k_2 ^ 2}(e^{-i(k_1,x)-i(k_2,y ' ) } -x\leftrightarrow y ) -i\{t^{m'},t^{n'}\}{k_{1i}k_{2j}\over
k_1 ^ 2k_2 ^ 2 } ( e^{-i(k_1,x)}-e^{-i(k_1,y)})(e^{-i(k_2,x')}-e^{-i(k_2,y')})\big\}u_y \nonumber\end{aligned}\ ] ] this result agrees with ref . .
performing the fourier transformation @xmath141 we get @xmath142 + i{\{t^b , t^c\}\over\bar{u } u}\big({x_i\over x^2}-{y_i\over y^2}\big ) \big({x'_j\over { x'}^2}-{y'_j\over { y'}^2}\big ) \big\}u_x\nonumber\\ & & \hspace{-4mm}\times~ \big\{t^{a'}f^{a'b'c'}\bigg[{{x_{ij}\over ( z - z')^2 } + { x_ix'_j\over ux^2 } -{x_ix'_j\over \bar{u}{x'}^2}\over ux^2+\bar{u}{x'}^2 } -{x_iy'_j\over \bar{u } u x^2{y'}^2}-(x\leftrightarrow y)\bigg ] -i{\{t^{b'},t^{c'}\}\over\bar{u } u}\big({x_i\over x^2}-{y_i\over y^2}\big ) \big({x'_j\over { x'}^2}-{y'_j\over { y'}^2}\big ) \big\}u^{\dagger}_y \label{2cutsum}\end{aligned}\ ] ] where we introduced the notations @xmath143 it is convenient to return back to the notation @xmath165 and @xmath166 ( after @xmath167 the value of @xmath65 is set equal to @xmath168 ) .
@xmath169 + ~{2i\sigma\over x^2(\alpha_1y^2+\alpha_2{y'}^2)}{\rm tr}\{t^ct^bu_xf^{a'b'c'}t^{a'}u^\dagger_y -t^a f^{abc}u_xt^{b'}t^{c'}u^\dagger_y\ } \bigg ] \nonumber\end{aligned}\ ] ] the corresponding term in @xmath170 is ( see eq .
( [ bk ] ) ) @xmath171 the relevant term in the `` matrix element '' @xmath172 in the external shock - wave background comes from @xmath173 taken in the leading order in @xmath20 ( so that @xmath174 ) and @xmath175 taken in the first order in @xmath20 @xmath176 or vice versa : @xmath177 and @xmath178 here we have used the leading - order equations for wilson lines with arbitrary color indices @xcite . substituting eqs .
( [ subtr4 ] ) and ( [ subtr5 ] ) in eq .
( [ subtr3 ] ) we obtain @xmath179 from eq .
( [ subtraction ] ) we get @xmath180 \nonumber\\ & & \hspace{57 mm } + ~{2i\sigma\over x^2(\alpha_1y^2+\alpha_2{y'}^2)}{\rm tr}\{t^ct^bu_xf^{a'b'c'}t^{a'}u^\dagger_y -t^a f^{abc}u_xt^{b'}t^{c'}u^\dagger_y\ } \bigg ] \nonumber\\ & & \hspace{-4 mm } -~{2\alpha_s^2\over\pi^4}\!\int_0^\sigma\ !
{ d\alpha_2\over \alpha_2}\!\int\ ! d^2z~ { ( x , y)(y',z - z')\over x^2y^2{y'}^2(z - z')^2 } { \rm tr}\ { if^{a'b'c'}t^ct^bu_xt^{b'}u^\dagger_y -if^{abc}t^au_x t^{b'}t^{c ' } u^{\dagger}_y \}u_z^{bb'}u_{z'}^{cc ' } \nonumber\\ & & \hspace{-4 mm } = ~ { \alpha_s^2\over 2\pi^4 } \!\int_0 ^
1{du\over \bar{u}}\!\int \!d^2 zd^2 z'~u_z^{bb'}u_{z'}^{cc ' } { ( x , y)(y',z - z')\over ( z - z')^2{y'}^2 } \bigg[f^{abc}f^{a'b'c'}{\rm tr}\{t^au_xt^{a'}u^\dagger_y\ } \big[{2u\over ( ux^2+\bar{u}{x'}^2)(uy^2+\bar{u}{y'}^2)}-{2\over x^2y^2}\big ] \nonumber\\ & & \hspace{-4 mm } -~{i\over y^2}\big[{1\over ux^2+\bar{u}{x'}^2}-{1\over x^2}\big]f^{abc } { \rm tr}\{t^au_x\{t^{b'},t^{c'}\}u^\dagger_y\ } + { i\over y^2}\big[{1\over ux^2+\bar{u}{x'}^2}- { 1\over x^2}\big]f^{a'b'c'}{\rm tr}\{\{t^b , t^c\}u_xt^{a'}u^\dagger_y\ } \nonumber\\ & & \hspace{51 mm } + ~{2i\over x^2}\big[{1\over uy^2+\bar{u}{y'}^2}-{1\over y^2}\big]{\rm tr}\{t^ct^bu_xf^{a'b'c'}t^{a'}u^\dagger_y -t^a f^{abc}u_xt^{b'}t^{c'}u^\dagger_y\ } \bigg ] \nonumber\end{aligned}\ ] ] which corresponds to the @xmath181_+$ ] prescription ( [ pluscription ] ) ( the same prescription was used in ref .
note that the `` plus '' prescription ( [ pluscription ] ) is a consequence of the `` rigid '' cutoff @xmath182 ( [ cutoff ] ) ; with the `` smooth '' cutoff ( [ defy ] ) we would get different results - see appendix b. there are four color structures in the r.h.s . of eq .
( [ 2cutsum ] ) .
three of them reduce to @xmath184 we will not need the explicit form of the fourth color structure @xmath185 since it is multiplied by pure lo@xmath186 integral @xmath187 and does not contribute to the nlo kernel .
performing integration over @xmath83 using the prescription ( [ pluscription ] ) after some algebra we get @xmath188 ^ 2 \nonumber\\ & & \hspace{-12 mm } = ~ { 1\over ( z - z')^4 } \big[-4 + 2{x^2{y'}^2+{x'}^2y^2 - 4\delta^2(z - z')^2\over x^2{y'}^2-{x'}^2y^2}\ln{x^2{y'}^2\over { x'}^2y^2 } \big ] \nonumber\\ & & \hspace{-12 mm } + ~ \big({(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big ) \ln{x^2{y'}^2\over { x'}^2y^2 } \label{integralsym}\end{aligned}\ ] ] and @xmath189 \big({x_i\over x^2}-{y_i\over y^2}\big ) \big({x'_j\over { x'}^2}-{y'_j\over { y'}^2}\big ) \nonumber\\ & & \hspace{-2 mm } = ~-{(x - y)^4\over 2x^2y^2{x'}^2{y'}^2}\ln { x^2{y'}^2\over { x'}^2y^2 } + { ( x - y)^2\over 2(z - z')^2}\big[{1\over x^2{y'}^2}+{1\over{x'}^2y^2}\big ] \ln
{ x^2{y'}^2\over { x'}^2y^2 } \label{integralantisym}\end{aligned}\ ] ] so the two - cut contribution ( [ 2cutsum ] ) reduces to @xmath190 } \nonumber\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big ) \ln{x^2{y'}^2\over { x'}^2y^2}\big\ } \nonumber\\ & & \hspace{42 mm } \times~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\ } ] \nonumber\\ & & \hspace{-2 mm } + ~\big\ { -{(x - y)^4\over x^2{y'}^2{x'}^2y^2 } + { ( x - y)^2\over ( z - z')^2 } \big({1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big)\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\}\bigg ] \label{2cut}\end{aligned}\ ] ] this result agrees with the 1@xmath1833 dipoles kernel calculated in ref . @xcite .
the integral in the r.h.s . of eq .
( [ 2cut ] ) diverges as @xmath191 .
it is convenient to separate the divergent term by subtracting and adding the contribution at @xmath192 : @xmath193 \\ & & \hspace{-2 mm } = ~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\ } -(z'\rightarrow z ) ] + [ { \rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_y\}-{\rm tr}\{u_xu^\dagger_y\ } ] \nonumber\end{aligned}\ ] ] for the last line in the r.h.s . of eq . (
[ 2cut ] ) the subtraction is redundant since @xmath194 the easiest way to prove this is to set @xmath195 and make an inversion @xmath196 so the integral ( [ pochemyzero1 ] ) reduces to @xmath197 thus , we obtain @xmath198}\nonumber\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big ) \nonumber\\ & & \hspace{42 mm } \times~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\}-(z'\rightarrow z ) ] \nonumber\\ & & \hspace{-2 mm } + ~\big\ { -{(x - y)^4\over x^2{y'}^2{x'}^2y^2 } + { ( x - y)^2\over ( z - z')^2 } \big({1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big)\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\}\bigg ] \nonumber\\ & & \hspace{-2 mm } + ~ { \alpha_s^2\over 16\pi^4 } \int \!d^2 z[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_y\}-{\rm tr}\{u_xu^\dagger_y\ } ] \!\int\!d^2 z ' \big [ -{4\over ( z - z')^4}+\big\{2{x^2{y'}^2+{x'}^2y^2 - 4(x - y)^2(z - z')^2\over ( z - z')^4[x^2{y'}^2-{x'}^2y^2]}\nonumber\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big]\end{aligned}\ ] ] the first term is now finite while the second term contains the uv divergent contribution which reflects the usual uv divergency of the one - loop diagrams . to find the second term we use the dimensional regularization in the transverse space and set @xmath199 . because the fourier transforms ( [ furie ] ) are more complicated at @xmath200 it is convenient to return back to eq .
( [ twocutsum ] ) and calculate the subtracted term in the the momentum representation . the calculation is performed in appendix a and here we only quote the final result ( [ appendixresult ] ) @xmath201}\label{2cutfinal}\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big ) \nonumber\\ & & \hspace{42 mm } \times~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\}-(z'\rightarrow z ) ] \nonumber\\ & & \hspace{-2 mm } + ~\big\ { -{(x - y)^4\over x^2{y'}^2{x'}^2y^2 } + { ( x - y)^2\over ( z - z')^2 } \big({1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big)\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } { \rm
tr}\{u_x u^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\}\bigg ] \nonumber\\ & & \hspace{-2 mm } -~ { \alpha_s^2n_c\over 8\pi^3}\!\int\ ! d^2z { ( x - y)^2\over x^2y^2}\big[{11\over 3}\ln{x^2y^2\over ( x - y)^2}\mu^2 + { 67\over 9}-{\pi^2\over 3}\big ] [ { \rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_y\ } -{1\over n_c}{\rm tr}\{u_xu^\dagger_y\ } ] \nonumber\end{aligned}\ ] ] where @xmath29 is the normalization scale in the @xmath30 scheme .
the relevant diagrams are shown in fig .
[ run ] ( plus permutations ) .
let us start from the sum of diagrams fig . [ run ]
a and b. it has the form : @xmath202 \nonumber\end{aligned}\ ] ] where the first term in the square brackets comes from fig .
[ run]a and the second from fig .
we use the principal - value prescription for the @xmath203 terms in @xmath204 in loop integrals .
.,scaledwidth=113.0% ] .,scaledwidth=113.0% ] to regularize the uv divergence we change the dimension of the transverse space to 2-@xmath205 . after some algebra one obtains @xmath206 } \nonumber\\ & & \hspace{-8 mm } \times~\bigg\{-\varepsilon [ ( \alpha-2\alpha')\beta s+{k'}_{\perp}^2-(k - k')_{\perp}^2 ] [ { \alpha-2\alpha'\over\alpha}(k , q)_{\perp}+(2k'-k , q)_{\perp}]+ 2{(\alpha-2\alpha')^2\over\alpha}(k , q)_{\perp}\beta s \nonumber\\ & & \hspace{-8 mm } + ~ { ( \alpha-2\alpha')\over\alpha}(k , q)_{\perp}(2k'-k , k)_{\perp}+ ( \alpha-2\alpha')\beta s(2k'-k , q)_{\perp}+(q , k)_{\perp}(k-2k',k - k')_{\perp } + ( q , k')_{\perp}(k,2k'-k)_{\perp } \nonumber\\ & & \hspace{-8 mm } + ~{\alpha+\alpha'\over\alpha-\alpha ' } \bigg[{\alpha-2\alpha'\over\alpha}(q , k)_{\perp}(k , k - k')_{\perp}+ ( q,2k'-k)_{\perp}(k , k - k')_{\perp}+ ( q , k)_{\perp}({k'}^2_{\perp}-k^2_{\perp})+(q , k - k')_{\perp}(k,2k - k')_{\perp } \nonumber\\ & & \hspace{-8 mm } + ~(\alpha-2\alpha')(q , k - k')_{\perp}\beta s+ ( q , k - k')_{\perp}({k'}^2_{\perp}-(k - k')^2_{\perp})- ( q , k')_{\perp}(k',k - k')_{\perp}+(q , k - k')_{\perp}(k - k')^2_{\perp}\bigg ] \nonumber\\ & & \hspace{-8 mm } + ~{\alpha'-2\alpha\over\alpha ' } \bigg[{\alpha-2\alpha'\over\alpha}(q , k)_{\perp}(k , k')_{\perp}+ ( q,2k'-k)_{\perp}(k , k')_{\perp}-(q , k')_{\perp}(k , k+k')_{\perp } + ( q , k)_{\perp}(k',2k - k')_{\perp } \nonumber\\ & & \hspace{-8 mm } + ~(\alpha-2\alpha')(q , k')_{\perp}\beta s+ ( q , k')_{\perp}(k,2k'-k)_{\perp}-(q , k')_{\perp}{k'}^2_{\perp}+ ( q , k - k')_{\perp}(k',k - k')_{\perp}\bigg ] \nonumber\\ & & \hspace{-8 mm } + ~{\alpha+\alpha'\over\alpha-\alpha'}~{\alpha'-2\alpha\over\alpha ' } \big[(k , k')_{\perp}(q , k - k')_{\perp}+(q , k')_{\perp}(k , k - k')_{\perp}\big ] + [ ( k - k')_{\perp}^2+{k'}^2](q , k)_{\perp}+ \nonumber\\ & & \hspace{-8 mm } + ~4\alpha(q , k)_{\perp}\big[{\alpha-\alpha'\over\alpha'}+ { \alpha\over\alpha-\alpha'}\big ] \beta s- 4\alpha(q , k)_{\perp } \big[{{k'}_{\perp}^2\over\alpha-\alpha'}+{(k - k')_{\perp}^2\over\alpha'}\big ] \bigg\ } \nonumber\end{aligned}\ ] ] where we have omitted the contribution @xmath207 taking residues at @xmath208 and @xmath209 and changing to variable @xmath210 we obtain @xmath211 }
\nonumber\\ & & \hspace{-2 mm } \times~\bigg\ { -2\varepsilon(2k'-k , k)(q , k'-ku)+(1 - 2u)(k , q)(2k'-k , k)+(q , k)(k-2k',k - k')+(q , k')(k,2k'-k ) \nonumber\\ & & \hspace{-2 mm } + ~{1+u\over \bar{u}}\big[(1 - 2u)(q , k)(k , k - k')+2(q , k){k'}^2 - 2(k , k')_\perp(q , k ' ) \big ] \nonumber\\ & & \hspace{-2 mm } -~{2-u\over u } \big[(1 - 2u)(q , k)(k , k')+2(q , k)(k - k',k')-2(q , k')(k - k',k)\big ] \nonumber\\ & & \hspace{-2 mm } -~{(1+u)(2-u)\over u\bar{u}}[(k , k')(q , k - k')+(q , k')(k , k - k')]+(q , k ) \big[(k - k')^2+{k'}^2 - 4{{k'}^2\over\bar{u}}-4{(k - k')^2\over u}\big]\bigg\ } \nonumber\end{aligned}\ ] ] using the `` plus '' - prescription ( [ pluscription ] ) to subtract the ( lo)@xmath186 contribution we get @xmath212\ !
\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q ~{e^{i(q , x)-i(k , y)}\over k^2q^2 } \nonumber\\ & & \hspace{-2 mm } \times~ \bigg [ \big({(q,2k - k')\over { k'}^2}-{(q , k+k')\over(k - k')^2}\big)\ln{(k - k')^2\over{k'}^2 } + \!\int_0 ^ 1\ !
du~{(2-\varepsilon)(q , ku - k')(k , k-2k')+2(q , k)k^2 \over k^2[{k'}^2\bar{u}+(k - k')^2u]}\bigg ] \nonumber\end{aligned}\ ] ] next we calculate diagram shown in fig .
[ run]c .
@xmath213 there are 2 regions of integration over @xmath65 s : @xmath214 and @xmath215 . taking relevant residues ,
we obtain @xmath216 \nonumber\\ & & \hspace{-2 mm } + ~{{k'}^2\over u}(q,2k - k')-2(q , k)(k',k - k')\big ] + { 1\over ( k - k')^2\bar{u}+ k^2u}\big[{1\over u}[(q , k){k'}^2+(q , k')k^2]-(k',2k - k')(q , k ) \big]\big\ } \label{vert2}\end{aligned}\ ] ] where we have introduced the variable @xmath217 as usual .
after integration over @xmath83 with help of eq .
( [ pluscription ] ) this reduces to @xmath218+(q , k'-2k)\big]\ln{(k - k')^2\over{k'}^2 } \nonumber\\ & & \hspace{-2 mm } + ~\big[{1\over ( k - k')^2}[(q , k){k'}^2+(q , k')k^2]+(q , k)\big]\ln{(k - k')^2\over k^2 } -\int_0 ^ 1\ ! du~{2(q , k)(k',k - k')\over { k'}^2\bar{u}+(k - k')^2u } \big\}~{e^{i(q , x)-i(k , y)}\over { k'}^2q^2k^2 } \label{vert3}\end{aligned}\ ] ] and therefore @xmath219\ !
\nonumber\\ & & \hspace{-22 mm } \times\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q ~{e^{i(q , x)-i(k , y)}\over k^2q^2 } ~\big\{\big[{(q , k')\over ( k - k')^2}+{(q , k')k^2\over { k'}^2(k - k')^2}+ { ( q , k'-2k)\over { k'}^2}\big]\ln{(k - k')^2\over { k'}^2}\nonumber\\ & & \hspace{-22mm}-~ \big[{(q , k)\over { k'}^2}+{(q , k)\over ( k - k')^2}+{(q , k - k')k^2\over { k'}^2(k - k')^2 } \big]\ln{k^2\over { k'}^2 } -2\!\int_0 ^ 1\!du~{(q , k)(k',k - k')\over { k'}^2[{k'}^2\bar{u}+(k - k')^2u]}\big\ } \label{vert4}\end{aligned}\ ] ] next we calculate the sum of diagrams in fig .
[ run ] d , e , and f. the contribution of the diagram shown in fig .
[ run]d is @xmath220 where we took residues at @xmath221 , @xmath222 and introduced the variable @xmath223 .
it should be noted that the cutoff @xmath103 in the r.h.s . of this equation
translates into @xmath224 while our cutoff ( [ cutoff ] ) corresponds to @xmath225 .
fortunately , the difference @xmath226 ~=~{1\over { k'}^2}\!\int_0 ^ 1\
! { dv\over v}\ln{{k'}^2+k^2v\over { k'}^2\bar{v}+k^2v}\ ] ] does not contain @xmath227 and hence does not contribute to the nlo kernel .
similarly , one can impose the cutoff @xmath228 instead of the cutoff @xmath229 in other diagrams whenever convenient . before calculating the diagrams in fig .
[ run]e and fig .
[ run]f it is convenient to make the replacement @xmath230 which can be performed since the color indices @xmath35 and @xmath231 in @xmath232 are contracted . for the diagram in fig .
[ run]e we get @xmath233 where @xmath234 . taking residues at @xmath235 , @xmath221 at @xmath236 and @xmath237 , @xmath221 at @xmath238 we obtain @xmath239 the diagram in fig .
[ run]f yields @xmath240 adding eqs .
( [ vert5 ] ) , ( [ vert7 ] ) , ( [ vert8 ] ) and integrating over @xmath83 using eq . ( [ pluscription ] )
we get @xmath241 \nonumber\end{aligned}\ ] ] note that the diagram in fig .
[ run]f does not contribute to the nlo kernel . the contribution of the last `` running coupling '' diagram shown in fig .
[ run]g has the form @xmath242 where we have again replaced @xmath243 by @xmath244 . using the eq .
( [ bk1 ] ) we get @xmath245 which is obviously a ( lo)@xmath186 term which does not contribute to the nlo kernel . combining expressions ( [ vkladotself ] ) , ( [ vert4 ] ) , and ( [ vert9 ] ) we get @xmath246\ !
\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q ~{e^{i(q , x)-i(k , y)}\over k^2q^2 } \nonumber\\ & & \hspace{-2 mm } \times~ \bigg\ { \big[{k^2\over { k'}^2}(q,2k - k')-{k^2\over(k - k')^2}(q , k+k')\big]\ln{(k - k')^2\over{k'}^2 } + \!\int_0 ^ 1\ !
du~{(2-\varepsilon)(q , ku - k')(k , k-2k')+2(q , k)k^2 \over { k'}^2\bar{u}+(k - k')^2u } \nonumber\\ & & \hspace{-2 mm } + ~\big[{(q , k')\over ( k - k')^2}+{(q , k')k^2\over { k'}^2(k - k')^2}+ { ( q , k'-2k)\over { k'}^2}\big]\ln{(k - k')^2\over { k'}^2}\nonumber\\ & & \hspace{-2mm}-~ \big[{(q , k)\over { k'}^2}+{(q , k)\over ( k - k')^2}+{(q , k - k')k^2\over { k'}^2(k - k')^2 } \big]\ln{k^2\over { k'}^2 } -2\!\int_0 ^ 1\!du~{(q , k)(k',k - k')\over { k'}^2[{k'}^2\bar{u}+(k - k')^2u ] } + { 2(q , k)\over { k'}^2}\ln{k^2\over{k'}^2}\bigg\ } \nonumber\\ & & \hspace{-2 mm } = ~-2\alpha_s^2n_c\mu^{2\varepsilon}\!\int\ ! d^{2-\varepsilon}z [ { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_z u^\dagger_y\ } -{1\over n_c}{\rm tr}\{u_x u^\dagger_y\}]\
! \int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q { e^{i(q , x)-i(k , y)}\over k^2q^2 } \bigg\{\ln{k^2\over { k'}^2 } \nonumber\\ & & \hspace{-2 mm } \times~{3(q , k')k^2 - 4(q , k)(k , k')\over { k'}^2(k - k')^2 } + \!\int_0 ^ 1\ !
du~{(2-\varepsilon)(q , ku - k')(k , k-2k')+2(q , k)k^2\over k^2[{k'}^2\bar{u}+(k - k')^2u]}~ -\int_0 ^ 1\!du~{2(q , k)(k',k - k')\over { k'}^2[{k'}^2\bar{u}+(k - k')^2u ] } \bigg\ } \nonumber\end{aligned}\ ] ] using the integral over @xmath247 @xmath248}~ \nonumber\\ & & \hspace{-12 mm } -\int_0 ^ 1\!du~{2(q , k)(k',k - k')\over { k'}^2[{k'}^2\bar{u}+(k - k')^2u]}\bigg\}~=~{(q , k)\over 4\pi}\bigg\{{\gamma(\varepsilon/ 2)\over ( k^2)^{\varepsilon/ 2 } } { \gamma^2(1-{\varepsilon\over 2})\over\gamma(2-\varepsilon ) } \big[{11\over 3}-\varepsilon{\pi^2\over 6}\big]+{1\over 9}\bigg\ } \nonumber\end{aligned}\ ] ] one reduces the r.h.s . of eq .
( [ vklad1 ] ) to @xmath249\ !
\nonumber\\ & & \hspace{-2 mm } \times\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q ~e^{i(q , x)-i(k , y)}~{(q , k)\over k^2q^2 } \big\{{\gamma(\varepsilon/ 2)\over ( k^2)^{\varepsilon/ 2 } } { \gamma^2(1-{\varepsilon\over 2})\over\gamma(2-\varepsilon ) } \big[{11\over 3}-\varepsilon{\pi^2\over 6}\big]+{1\over 9}\big\ } \label{vklad1otve}\end{aligned}\ ] ] next we subtract the counterterm @xmath250\ ! \!\int\ ! { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q~{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k~e^{i(q , x)-i(k , y ) } { ( q , k)\over q^2k^2 } \label{ct}\end{aligned}\ ] ] corresponding to the poles @xmath251 in the loop diagrams in fig .
[ run ] ( we use the @xmath30 scheme ) .
we obtain @xmath252\ !
\nonumber\\ & & \hspace{-2 mm } \times\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}q ~e^{i(q , x)-i(k , y)}~{(q , k)\over k^2q^2 } \big\{{11\over 3}\ln{\mu^2\over k^2}+{67\over 9}-{\pi^2\over 3}\big\ } \label{vklad1otvet}\end{aligned}\ ] ] the complete set of running - coupling diagrams is presented in fig .
[ 1cutdms ] . ,
scaledwidth=110.0% ] , scaledwidth=110.0% ] , scaledwidth=110.0% ] the contribution of diagrams in fig .
[ 1cutdms ] vii - xii differs from eq .
( [ vklad1otvet ] ) by the replacement @xmath253 and sign change .
there is also a symmetric set of diagrams xiii - xxiv obtained by the reflection of diagrams i - xii with respect to @xmath254 axis .
the result is obtained by the substitution @xmath255 and therefore the contribution of all diagrams in fig .
( [ 1cutdms ] ) takes the form @xmath256\ !
\nonumber\\ & & \hspace{-2 mm } \times\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q ~[e^{i(q , x)}-e^{i(q , y)}][e^{-i(k , x)}-e^{-i(k , y)}]{(q , k)\over k^2q^2 } \big\{{11\over 3}\ln{\mu^2\over k^2}+{67\over 9}-{\pi^2\over 3}\big\ } \label{vklad1fina}\end{aligned}\ ] ] the remaining diagrams xxv - xxviii contribute only to the ( lo)@xmath186 .
we have shown this for diagram xxvii ( fig .
[ run]g ) , see eq .
( [ xxviib ] ) .
the diagram in fig . [ 1cutdms ]
xxv is obtained from the equation ( [ xxviib ] ) by the replacement @xmath71 , and the diagrams xxvi and xxviii by the replacements @xmath257 by @xmath258 and @xmath259 , respectively .
thus , the diagrams xxv - xxvii do not contribute to the nlo kernel .
there is another set of diagrams obtained by the reflection of diagrams shown in fig .
( [ 1cutdms ] ) with respect to the shock - wave line .
their contribution is obtained from eq .
( [ vklad1fina ] ) by the replacement @xmath260 in the logarithm so the final result for the sum of all `` running coupling '' diagrams of fig . [ 1cutdms ]
type has the form @xmath261\ ! \label{vklad1final}\\ & & \hspace{44 mm } \times\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k'{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q ~[e^{i(q , x)}-e^{i(q , y)}][e^{-i(k , x)}-e^{i(k , y)}]{(q , k)\over k^2q^2 } \big\{{11\over 3}\ln{\mu^4\over q^2k^2}+{134\over 9}+{2\pi^2\over 3}\big\ } \nonumber\\ & & \hspace{-4 mm } = ~ { \alpha_s^2n_c\over 8\pi^3}\!\int\ ! d^2z [ { \rm
tr}\{u_x u^\dagger_z\}{\rm tr}\{u_z u^\dagger_y\ } -{1\over n_c}{\rm tr}\{u_x u^\dagger_y\}]\ ! ~\big\{{(x - y)^2\over x^2y^2 } \big[{11\over 3}\ln { x^2y^2\over \mu^{-4}}+{134\over 9}-{2\pi^2\over 3}\big ] + { 11\over 3}\big[{1\over x^2}-{1\over y^2}\big]\ln{x^2\over y^2}\big\ } \nonumber \end{aligned}\ ] ] there is one more class of diagrams with one gluon - shockwave intersection shown in fig .
[ 1cutadd ] .
these diagrams are uv - convergent so we do not need to change the dimension of the transverse space to @xmath262 . first we calculate diagrams shown in fig .
[ 1cutadd]a , b .
@xmath263 \theta(\alpha_1)\int { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q e^{i(q , x - z)_\perp+i(k_1,z)_\perp}{(q_1,k_1)_\perp\over \alpha_1(\alpha_1\beta_1 s - k_{1\perp}^2+i\epsilon ) q_\perp^2 } { e^{i(k_2,x)_\perp}\over\alpha_2(\alpha_2\beta_2 s - k_{2\perp}^2+i\epsilon ) } \nonumber\\ & & \hspace{-1 mm } = ~-{g^4\over \pi^2}{\rm tr}\{t^au_xt^bt^ct^du^\dagger_y\}\!\int\!d^{2-\varepsilon}z \!\int_0^\sigma\!{d\alpha\over\alpha}\!\int_0 ^ 1\ ! du\!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon}k { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^{2-\varepsilon } k'~ e^{-i(k , y)_\perp - i(k',y)_\perp } \nonumber\\ & & \hspace{41 mm } \times~\big[{\delta^{bd}u^{ac}_z\over k^2\bar{u}(k^2\bar{u}+{k'}^2u)}+{\delta^{ab}u^{dc}_z\over { k'}^2u(k^2\bar{u}+{k'}^2u ) } \big ] \!\int { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q e^{i(q , x - z)_\perp+i(k , z)_\perp+i(k',x)_\perp}{(q , k)_\perp\over q_\perp^2 } \nonumber\\ & & \hspace{-1 mm } = ~-{g^4\over \pi^2}{\rm tr}\{t^au_xt^bt^ct^du^\dagger_y\}\!\int\!d^2z \!\int_0^\infty\!\ !
{ d\alpha\over\alpha}\!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 k ' { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q ~{(q , k)\over q^2}~ e^{i(q , x)-i(k , y)_\perp+i(k',x - y)_\perp } { \delta^{bc}u^{ad}_z-\delta^{bd}u^{ac}_z\over k^2{k'}^2}\ln{k^2\over{k'}^2 } \nonumber\end{aligned}\ ] ] and therefore @xmath264 \!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 k ' { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q ~{(q , k)\over q^2k^2{k'}^2}~ e^{i(q , x - z)-i(k , y - z)_\perp+i(k',x - y)_\perp } \ln{k^2\over{k'}^2 } \nonumber\end{aligned}\ ] ] the contribution of diagrams shown in fig . [ 1cutadd]c , d is obtained from eq .
( [ add2 ] ) by the replacement @xmath71 in the left part of the graph and the sign change so that @xmath265 .
the sum of the diagrams in fig .
[ 1cutadd]a - d takes the form @xmath266 \label{sumad}\\ & & \hspace{33 mm } \times~ \!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k ' { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q ~{(q , k)\over q^2k^2{k'}^2}~ e^{i(q , x - z)}\big(e^{-i(k , y - z)_\perp - i(k',x - y)_\perp}-x\leftrightarrow y\big ) \ln{k^2\over{k'}^2 } \nonumber\end{aligned}\ ] ] 2 dipoles transition diagrams .
[ 1cutadd],scaledwidth=115.0% ] next relevant diagram is shown in fig .
[ 1cutadd]e @xmath267_\lambda \over q^2[(\alpha_1+\alpha_2)(\beta_1+\beta_2)s-(k_1+k_2)^2_{\perp}+i\epsilon ] } \gamma^{\mu\nu\lambda}(k_1,k_2,-k_1-k_2)~ \nonumber\\ \nonumber\end{aligned}\ ] ] there are three regions of integration over @xmath65 s : @xmath268 , @xmath269 and @xmath270 .
going to the variables @xmath271 in the first region , @xmath272 in the second and @xmath273 in the third , we obtain @xmath274 } \nonumber\\ & & \hspace{-1mm}-~ \bar{u } \big(k^\perp_1 + 2{(k_1+k_2)_\perp^2\over \alpha\bar{u } s}p_2\big)_\mu k^\perp_{2\nu}\big(q+{2(k_1+k_2,q)_{\perp}\over \alpha\bar{u } s}p_2\big)_{\lambda } { \gamma^{\mu\nu\lambda}(-u\alpha p_1+k^\perp_1,\alpha p_1+k^\perp_2 , -\alpha\bar{u}-k^\perp_1-k^\perp_2 ) \over uk_{2\perp}^2(k_1+k_2)_\perp^2[u(k_1+k_2)_\perp^2+\bar{u } k_{1\perp}^2 ] } \bigg ] \nonumber\end{aligned}\ ] ] using the formula @xmath275 \nonumber\end{aligned}\ ] ] we get @xmath276 \bigg\{{(q , k_1)(k_2,k_1+k_2)-(q , k_2)(k_1,k_1+k_2)\over \bar{u}u k_1 ^ 2k_2 ^ 2(k_1+k_2)^2 } \nonumber\\ & & \hspace{-1 mm } + ~ { ( 1+\bar{u})(k_1+k_2)^2(q , k_1)-(1+u)(q , k_1+k_2)(k_1,k_2)- \bar{u}[(q , k_1)(k_2,k_1+k_2)-(q , k_2)(k_1,k_1+k_2 ) ] \over uk_1 ^ 2(k_1+k_2)^2(k_2 ^ 2\bar{u}+(k_1+k_2)^2u ) } \nonumber\\ & & \hspace{-1mm}+~ { -(1+\bar{u})(k_1+k_2)^2(q , k_2)+(1+u)(q , k_1+k_2)(k_1,k_2)- \bar{u}[(q , k_1)(k_2,k_1+k_2)-(q , k_2)(k_1,k_1+k_2 ) ] \over uk_2
^ 2(k_1+k_2)^2(k_1 ^ 2\bar{u}+(k_1+k_2)^2u)}\bigg\ } \nonumber\end{aligned}\ ] ] performing the integration over u ( with prescription ( [ pluscription ] ) ) we obtain @xmath277 \!\int\ ! { { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_1{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_2\over k_1 ^ 2k_2 ^ 2(k_1+k_2)^2 } { { { \textstyle d }
\lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 q\over q^2 } ~e^{i(q - k_1,x)-i(k_2,y ) } u_z^{cl } \nonumber\\ & & \hspace{-1mm}\times~ \big\{(k_1+k_2)^2 \big[(q , k_2)\ln{(k_1+k_2)^2\over k_1 ^ 2}-(q , k_1)\ln{(k_1+k_2)^2\over k_2 ^ 2 } \big]-(q , k_1+k_2)(k^2_1+k_2 ^ 2)\ln{k_1 ^ 2\over k_2 ^ 2}\big\ } \nonumber\\ & & \hspace{-1 mm } = ~{g^4n_c\over 8\pi^2}\!\int\ !
d^2z~ [ { \rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_y\ } ] \!\int\ ! { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k ' { { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 q\over q^2 } ~e^{i(q - k',x - z)-i(k - k',y - z ) } u_z^{cl } \nonumber\\ & & \hspace{-1mm}\times~ \big\ { { ( q , k - k')\over { k'}^2(k - k')^2}\ln{k^2\over { k'}^2}-{(q , k')\over { k'}^2(k - k')^2}\ln{k^2\over ( k - k')^2 } + { ( q , k)\over k^2{k'}^2}\ln{(k - k')^2\over { k'}^2}+{(q , k)\over k^2(k - k')^2}\ln{(k - k')^2\over { k'}^2}\big\ } \nonumber\end{aligned}\ ] ] where we made the change of variables @xmath278 and @xmath279 . the sum of diagrams shown in fig .
[ 1cutadd]a - e can be represented as @xmath280 \!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k ' { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q \nonumber\\ & & \hspace{-1 mm } \times~e^{i(q , x - z ) } \big(e^{-i(k , y - z)_\perp - i(k',x - y)_\perp}-x\leftrightarrow y\big ) \big[{(q , k - k')\over { k'}^2(k - k')^2}+{(q , k)\over k^2(k - k')^2}-{(q , k)\over k^2{k'}^2}\big]\ln{k^2\over{k'}^2 } \label{sum1cutadd}\end{aligned}\ ] ] note that the expressions ( [ sumad ] ) and ( [ fig1cutaddc ] ) are ir divergent as @xmath281 but their sum ( [ sum1cutadd ] ) is ir stable . once again , the contribution of the diagrams in fig .
[ 1cutadd]f - k are obtained by replacement @xmath282 so the contribution of diagrams of fig . [
1cutadd ] a - k has the form @xmath283 \!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k ' { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q \nonumber\\ & & \hspace{-1 mm } \times~(e^{i(q , x - z ) } -e^{i(q , y - z)})\big(e^{-i(k , y - z)_\perp - i(k',x - y)_\perp}-x\leftrightarrow y\big ) \big[{(q , k - k')\over { k'}^2(k - k')^2}+{(q , k)\over k^2(k - k')^2}-{(q , k)\over k^2{k'}^2}\big]\ln{k^2\over{k'}^2 } \label{sum1cutadda}\end{aligned}\ ] ] performing the fourier transformation with the help of the formula @xmath284 + { 1\over 2}\ln{x^2\over y^2}\ln{(x - y , y)+i\kappa\over ( x - y , y)-i\kappa}\bigg\ } \nonumber\\ & & \hspace{-12 mm } + ~{i\over 8\pi^2}\big({(x , y)\over x^2}x_i - y_i\big){1\over i\kappa } \bigg\{\int_0 ^ 1\ !
du~ \bigg[{\ln u\over u-{(x , x - y)-i\kappa\over ( x - y)^2 } } -{\ln u\over u-{(x , x - y)+i\kappa\over ( x - y)^2}}\bigg ] -{1\over 2}\ln{(x - y)^2\over x^2}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa}\bigg\ } \nonumber\end{aligned}\ ] ] ( here @xmath285 ) one obtains @xmath286 { ( x - y)^2\over x^2y^2}\ln{x^2\over ( x - y)^2}\ln{y^2\over ( x - y)^2 } \label{1cutaddcoord}\ ] ] note that the two last terms in the r.h.s .
( [ fypie ] ) do not contribute . the contribution of the diagram obtained by reflection of fig .
[ 1cutadd ] with respect to the shock wave differs from eq .
( [ sum1cutadd ] ) by replacement @xmath260 which doubles result ( [ 1cutaddcoord ] ) .
the final expression for the contribution of all `` dipole recombination diagrams '' of fig .
[ 1cutadd ] type has the form @xmath287 { ( x - y)^2\over x^2y^2}\ln{x^2\over ( x - y)^2}\ln{y^2\over ( x - y)^2 } \label{1cutaddfinal}\ ] ]
adding results ( [ 2cut ] ) , ( [ vklad1final ] ) and ( [ 1cutaddfinal ] ) one obtains the contribution of the diagrams with one and two gluon intersections with the shock wave in the form : @xmath288 \label{12cuts}\\ & & \hspace{-2 mm } + ~{11\over 3}\big[{1\over x^2}-{1\over y^2}\big]\ln{x^2\over y^2 } -2{(x - y)^2\over x^2y^2}\ln{x^2\over ( x - y)^2}\ln{y^2\over ( x - y)^2}\big\ } [ { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_z u^\dagger_y\ } -{1\over n_c}{\rm tr}\{u_x u^\dagger_y\ } ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2\over 16\pi^4 } \int \!d^2 zd^2 z ' \bigg [ \big(-{4\over ( z - z')^4}+\big\{2{x^2{y'}^2+{x'}^2y^2 - 4(x - y)^2(z - z')^2\over ( z - z')^4[x^2{y'}^2-{x'}^2y^2]}\nonumber\\ & & \hspace{-2 mm } + ~\big({(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big ) \nonumber\\ & & \hspace{42 mm } \times~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\}-(z'\rightarrow z ) ] \nonumber\\ & & \hspace{-2 mm } + ~\big\ { -{(x - y)^4\over x^2{y'}^2{x'}^2y^2 } + { ( x - y)^2\over ( z - z')^2 } \big({1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big)\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\}\bigg ] \nonumber\end{aligned}\ ] ] .,scaledwidth=113.0% ] there are also diagrams without gluon - shockwave intersection like the graph shown in fig .
[ 0cuts ] .
they are proportional to the parent dipole @xmath289 and their contribution can be found from eq .
( [ 12cuts ] ) using the requirement that the r.h.s . of the evolution equation must vanish at @xmath73 ( since @xmath290=1 ) .
it is easy to see that the replacement @xmath291 by @xmath292 fulfills the above requirement so one obtains the final gluon contribution to the nlo kernel in the form @xmath293
\label{nlogluon}\\ & & \hspace{-2 mm } + ~{11\over 3}\big[{1\over x^2}-{1\over y^2}\big]\ln{x^2\over y^2 } -2{(x - y)^2\over x^2y^2}\ln{x^2\over ( x - y)^2}\ln{y^2\over ( x - y)^2}\big\ } [ { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_z u^\dagger_y\ } -n_c{\rm tr}\{u_x u^\dagger_y\ } ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2\over 16\pi^4 } \int \!d^2 zd^2 z ' \bigg [ \big(-{4\over ( z - z')^4}+\big\{2{x^2{y'}^2+{x'}^2y^2 - 4(x - y)^2(z - z')^2\over ( z - z')^4[x^2{y'}^2-{x'}^2y^2]}\nonumber\\ & & \hspace{-2 mm } + ~\big({(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2}\big ) \nonumber\\ & & \hspace{42 mm } \times~[{\rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\ } -{\rm tr}\{u_xu^\dagger_z u_{z'}u^\dagger_yu_zu^\dagger_{z'}\}-(z'\rightarrow z ) ] \nonumber\\ & & \hspace{-2 mm } + ~\big\ { -{(x - y)^4\over x^2{y'}^2{x'}^2y^2 } + { ( x - y)^2\over ( z - z')^2 } \big({1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big)\big\}\ln{x^2{y'}^2\over { x'}^2y^2 } { \rm tr}\{u_x u^\dagger_z\}{\rm tr}\{u_zu^\dagger_{z'}\}{\rm tr}\{u_{z'}u^\dagger_y\}\bigg ] \nonumber\end{aligned}\ ] ] promoting wilson lines in the r.h.s of this equation to operators and adding the quark contribution from ref .
@xcite @xmath294 \nonumber\\ & & \hspace{6 mm } \times~ \big [ -{\alpha_s n_f\over 6\pi}{(x - y)^2\over x^2 y^2}\big(\ln ( x - y)^2\mu^2+{5\over 3}\big)+ { \alpha_s n_f\over 6\pi}{x^2-y^2\over x^2y^2}\ln{x^2\over y^2}\big ] \nonumber\\ & & \hspace{-6 mm } ~+{\alpha^2_s\over \pi^4}n_f { \rm tr}\{t^au_xt^bu^{\dagger}_y\ } \!\int\!d^2z d^2z'~ { \rm tr}\{t^a u_zt^b u^\dagger_{z'}-t^au_zt^bu^\dagger_z\}~{1\over ( z - z')^4 } \nonumber\\ & & \hspace{6 mm } \times~ \big\{1 -{{x'}^2y^2+{y'}^2x^2-(x - y)^2(z - z')^2\over 2({x'}^2y^2-{y'}^2x^2 ) } \ln{{x'}^2y^2\over { y'}^2x^2}\big\ } \label{nlobkw}\end{aligned}\ ] ] we obtain the full nlo kernel cited in eq .
( [ nlobk ] ) .
in this section we compare our kernel to the forward nlo bfkl results@xcite . the linearized equation ( [ nlobk ] )
has the form @xmath295\big\ } [ \hat{\cal u}(x , z)+\hat{\cal u}(z , y)-\hat{\cal u}(x , y ) ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2n_c^2\over 16\pi^4 } \int \!d^2 zd^2 z ' \bigg [ -{4\over ( z - z')^4}+\big\{2{x^2{y'}^2+{x'}^2y^2 - 4(x - y)^2(z - z')^2\over ( z - z')^4[x^2{y'}^2-{x'}^2y^2]}\nonumber\\ & & \hspace{-2 mm } + ~{(x - y)^4\over x^2{y'}^2-{x'}^2y^2}\big [ { 1\over x^2{y'}^2}+{1\over y^2{x'}^2}\big ] + { ( x - y)^2\over ( z - z')^2}\big[{1\over x^2{y'}^2}-{1\over { x'}^2y^2}\big]\big\ } \ln{x^2{y'}^2\over { x'}^2y^2 } \nonumber\\ & & \hspace{-2 mm } -~{n_f\over n_c^3 } \big\{{4\over(z - z')^4 } -2{{x'}^2y^2+{y'}^2x^2-(x - y)^2(z - z')^2\over ( z - z')^4({x'}^2y^2-{y'}^2x^2 ) } \ln{{x'}^2y^2\over { y'}^2x^2}\big\ } \bigg]\hat{\cal u}(z , z ' ) \label{nlolin}\end{aligned}\ ] ] for the case of forward scattering @xmath296 and the linearized equation ( [ nlolin ] ) reduces to @xmath297\big\ } [ \langle\hat{\cal u}(x - z)\rangle+\langle\hat{\cal u}(z)\rangle-\langle\hat{\cal u}(x)\rangle ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2n_c^2\over 16\pi^4 } \int \!d^2 zd^2 z ' \bigg[-{4\over z^4}+\big\{2{(x - z - z')^2{z'}^2+(x - z')^2(z+z')^2 - 4x^2z^2\over z^4[(x - z - z')^2{z'}^2-(x - z')^2(z+z')^2]}+{x^4\over ( x - z - z')^2{z'}^2-(x - z')^2(z+z')^2 } \nonumber\\ & & \hspace{-2 mm } \times~\big [ { 1\over ( x - z - z')^2{z'}^2}+{1\over ( x - z')^2(z+z')^2}\big ] + { x^2\over z^2}\big[{1\over ( x - z - z')^2{z'}^2}-{1\over ( x - z')^2(z+z')^2}\big]\big\ } \ln{(x - z - z')^2{z'}^2\over ( x - z')^2(z+z')^2 } \nonumber\\ & & \hspace{-2 mm } -~{n_f\over n_c^3 } \big\{{4\over z^4 } -2{(x - z - z')^2{z'}^2+(x - z')^2(z+z')^2-x^2z^2\over z^4[(x - z - z')^2{z'}^2-(x - z')^2(z+z')^2 ] } \ln{(x - z - z')^2{z'}^2\over ( x - z')^2(z+z')^2}\big\ } \bigg]\hat{\cal u}(z)\rangle \label{nloforward}\end{aligned}\ ] ] using the integral @xmath298 from hep - ph/9704267 @xcite we get @xmath299 { 1\over { z'}^2(x - z - z')^2}\ln{(x - z - z')^2{z'}^2\over ( x - z')^2(z+z')^2}~ \nonumber\\ & & \hspace{-1 mm } = ~ ~{2x^2\over z^2}\bigg\ { { ( x^2-z^2)\over ( x - z)^2(x+z)^2 } \big[\ln{x^2\over z^2}\ln{x^2z^2(x - z)^4\over ( x^2+z^2)^4 } + 2li_2\big(-{z^2\over x^2}\big)-2li_2\big(-{x^2\over z^2}\big)\big ] \nonumber\\ & & \hspace{-1 mm } -~\big(1-{(x^2-z^2)^2\over ( x - z)^2(x+z)^2}\big)\big[\!\int_0 ^ 1-\int_1^\infty\big ] { du\over ( x - zu)^2}\ln{u^2z^2\over x^2}\bigg\ } \label{integral1}\end{aligned}\ ] ] @xmath297\big\ } [ \langle\hat{\cal u}(x - z)\rangle+\langle\hat{\cal u}(z)\rangle-\langle\hat{\cal u}(x)\rangle ] \nonumber\\ & & \hspace{-2 mm } + ~{\alpha_s^2n_c^2\over 4\pi^3}\!\int\ ! d^2z~ { x^2\over z^2 } \bigg\{\big(1+{n_f\over n_c^3}\big){3(x , z)^2 - 2x^2z^2\over16x^2z^2 } \big({2\over x^2}+{2\over z^2}+{x^2-z^2\over x^2z^2}\big)\ln{x^2\over z^2 } \nonumber\\ & & \hspace{-1 mm } -~\big[3+\big(1+{n_f\over n_c^3}\big)\big(1-{(x^2+z^2)^2\over 8x^2z^2 } + { 3x^4 + 3z^4 - 2x^2z^2\over 16x^4z^4}(x , z)^2\big)\big]\!\int_0^\infty\!dt{1\over x^2+t^2z^2}\ln{1+t\over |1-t| } \nonumber\\ & & \hspace{-1 mm } + ~ { ( x^2-z^2)\over ( x - z)^2(x+z)^2 } \big[\ln{x^2\over z^2}\ln{x^2z^2(x - z)^4\over ( x^2+z^2)^4 } + 2li_2\big(-{z^2\over x^2}\big)-2li_2\big(-{x^2\over z^2}\big)\big ] \nonumber\\ & & \hspace{-1 mm } -~\big(1-{(x^2-z^2)^2\over ( x - z)^2(x+z)^2}\big)\big[\!\int_0 ^ 1-\int_1^\infty\big ] { du\over ( x - zu)^2}\ln{u^2z^2\over x^2}\bigg\}~{\cal u}(z ) \label{nlobfkernel}\end{aligned}\ ] ] to compare the eigenvalues of the eq . ( [ nlobfkernel ] ) with nlo bfkl we expand @xmath300 in eigenfunctions @xmath301 compute the evolution of @xmath302 from eq .
( [ nlobfkernel ] ) and compare it to the calculation based on the nlo bfkl results from @xcite .
( for the quark part of the nlo bk kernel the agreement with nlo bfkl was proved in ref .
@xcite ) .
the relevant integrals have the form @xmath303 { x^2\over(x - z)^2z^2 } ~=~\chi(n,\gamma ) \nonumber\\ & & \hspace{-1 mm } { 1\over \pi } \!\int\ ! d^2z~ [ 2 ( z^2/x^2)^\gamma e^{in\phi}-1 ] \big({1\over(x - z)^2}- { 1\over z^2}\big)\ln{(x - z)^2\over z^2 } ~=~\chi^2(n,\gamma)-\chi'(n,\gamma)- { 4\gamma\chi(\gamma)\over\gamma^2-{n^2\over 4 } } \nonumber\\ & & \hspace{-1 mm } { 1\over \pi } \!\int\ ! d^2z ~ ( z^2/x^2)^\gamma { x^2\over(x - z)^2z^2 } e^{in\phi}\ln{(x - z)^2\over x^2}\ln{z^2\over x^2 } ~=~{1\over 2}\chi''(n,\gamma)+\chi'(n,\gamma)\chi(n,\gamma ) \label{relintegral1}\end{aligned}\ ] ] where @xmath304 , and @xmath305\!\int_0^\infty\!dt{1\over x^2+t^2z^2}\ln{1+t\over |1-t| } \bigg\ } \nonumber\\ & & \hspace{-1 mm } = ~\big\{-\big[3+\big(1+{n_f\over n_c^3}\big ) { 2 + 3\gamma\bar{\gamma}\over ( 3 - 2\gamma)(1 + 2\gamma)}\big]\delta_{0n } + \big(1+{n_f\over n_c^3}\big ) { \gamma\bar{\gamma}\over 2(3 - 2\gamma)(1 + 2\gamma)}\delta_{2n}\big\ } { \pi^2\cos\pi\gamma\over(1 - 2\gamma)\sin^2\pi\gamma } ~\equiv~f(n,\gamma ) \nonumber\end{aligned}\ ] ] @xmath306 \nonumber\\ & & \hspace{-1 mm } -~\big(1-{(x^2-z^2)^2\over ( x - z)^2(x+z)^2}\big)\big[\!\int_0 ^ 1-\int_1^\infty\big ] { du\over ( x - zu)^2}\ln{u^2z^2\over x^2}\bigg\}~=~-\phi(n,\gamma)-\phi(n,1-\gamma ) \label{relintegral3}\end{aligned}\ ] ] where @xcite @xmath307\big\ } \label{fi}\end{aligned}\ ] ] the convenient way to calculate the integrals over angle @xmath308 is to represent @xmath309 as @xmath310 and use formulas for the integration of chebyshev polynomials from ref . @xcite .
using integrals ( [ relintegral1 ] ) - ( [ relintegral3 ] ) one easily obtains the evolution equation for @xmath311 in the form @xmath312\chi(n,\gamma ) + { \alpha_s b\over 4\pi } \big[{1\over 2}\chi^2(n,\gamma)-{1\over 2}\chi'(n,\gamma ) -{2\gamma\chi(n,\gamma)\over\gamma^2-{n^2\over 4 } } \big ] \nonumber\\ & & \hspace{-1 mm } + ~{\alpha_s n_c\over 4\pi}\big[-\chi"(n,\gamma)-2\chi(n,\gamma)\chi'(n,\gamma)+4\zeta(3)+f(n,\gamma ) -2\phi(n,\gamma)-2\phi(n,1-\gamma)\big]\big\}\langle\hat{\cal u}(n,\gamma)\rangle \label{eigenvalue}\end{aligned}\ ] ] where @xmath313 etc .
next we calculate the same thing using nlo bfkl results @xcite .
the impact factor @xmath314 for the color dipole @xmath315 is proportional to @xmath316 so one obtains the cross section of the scattering of color dipole in the form @xmath317 where @xmath318 is the partial wave of the forward reggeized gluon scattering amplitude satisfying the equation @xmath319 and @xmath320 is the target impact factor .
the kernel @xmath321 is symmetric with respect to @xmath322 and the eigenvalues are @xmath323 , \label{lip3}\\ & & \hspace{-1 mm } \delta(n,\gamma)~=~-{b\over 2}[\chi'(n,\gamma)+\chi^2(n,\gamma ) ] + \big({67\over 9}-{\pi^2\over 3}-{10\over 9}{n_f\over n_c^3}\big)\chi(n,\gamma ) + 6\zeta(3 ) \nonumber\\ & & \hspace{35mm}-\chi"(n,\gamma)+f(n,\gamma ) -2\phi(n,\gamma)-2\phi(n,1-\gamma)\big\ } \nonumber\end{aligned}\ ] ] the corresponding expression for @xmath324 takes the form @xmath325 where @xmath326 is the angle between @xmath327 and @xmath0 axis . using eq .
( [ wgw ] ) we obtain @xmath328 the integration over q can be performed using @xmath329 \label{lipxz}\\ & & \hspace{-1 mm } \nonumber\end{aligned}\ ] ] ( recall that @xmath330 and @xmath331 with our accuracy ) .
the result is @xmath332 \nonumber\end{aligned}\ ] ] where the angle @xmath333 corresponds to @xmath334 . since @xmath335
we can neglect terms @xmath336 in the argument of @xmath337 and expand @xmath338 . using again eq .
( [ wgw ] ) in the leading order we can replace extra @xmath339 by @xmath340 and obtain @xmath341 \big ] \label{lip7}\end{aligned}\ ] ] finally , expanding @xmath342 we obtain @xmath343\big\ } \!\int\!{d^2 p\over p^2 } { d^2 q'\over { q'}^2}e^{-in\varphi}\alpha_s(p)\big({p^2\over 4\mu^2}\big)^\gamma \phi_b(q')\!\int_{a - i\infty}^{a+i\infty}\!{d\omega\over 2\pi i}\big({s\over pq'}\big)^\omega g_\omega(p , q ' ) \label{lip8}\end{aligned}\ ] ] which can be rewritten as an evolution equation @xmath344\big)\chi(n,\gamma ) + { \alpha_sn_c\over 4\pi}[\delta(n,\gamma)-2\chi(n,\gamma)\chi'(n,\gamma ) ] \big\ } \langle { \cal u}(n,\gamma)\rangle \nonumber\\ & & \hspace{-1 mm } ~=~{\alpha_s n_c\over \pi } \big\{\big[1-{b\alpha_s\over 4\pi}{d\over d\gamma } + \big({67\over 9}-{\pi^2\over 3}\big)n_c-{10\over 9}{n_f\over n_c^2}\big]\chi(n,\gamma ) + { \alpha_s b\over 4\pi } \big[{1\over 2}\chi^2(n,\gamma)-{1\over 2}\chi'(n,\gamma)-{2\gamma\over\gamma^2-{n^2\over 4 } } \chi(\gamma ) \big ] \nonumber\\ & & \hspace{-1 mm } + ~{\alpha_s n_c\over 4\pi}\big[-\chi"(n,\gamma)-2\chi(n,\gamma)\chi'(n,\gamma)+6\zeta(3)+f(n,\gamma ) -2\phi(n,\gamma)-2\phi(n,1-\gamma)\big]\big\}\langle \hat{\cal u}(n,\gamma)\rangle \label{leigenvalue}\end{aligned}\ ] ] this eigenvalue coincides with eq .
( [ eigenvalue ] ) up to the extra term @xmath345 .
it would correspond to the additional contribution to the r.h.s .
( [ nlobk ] ) in the form of @xmath346 which contradicts the requirement @xmath347 at @xmath73 .
a possible reason for the disagreement is the connection between the matrix element of the color dipole with a rigid cutoff @xmath103 and the cutoff by energy @xmath348 in eq .
( [ lip1 ] ) .
it is worth noting that the coefficient @xmath349 in eq .
( [ leigenvalue ] ) agrees with the @xmath350 asymptotics of the three - loop anomalous dimensions of leading - twist gluon operators @xcite .
.,scaledwidth=80.0% ] it should be emphasized that the coincidence of terms with the nontrivial @xmath351 dependence proves that there is no additional @xmath352 correction to the vertex of the gluon - shock wave interaction coming from the small loop inside the shock wave , see fig .
[ loopinside ] ( in other words , all the effects coming from the small loop in the shock wave are absorbed in the renormalization of coupling constant in the definition of the u operator ( [ defy ] ) ) . in the case of quark loop
, we proved that by the comparison of our results for @xmath289 in the shock - wave background with explicit light - cone calculation of the behavior of @xmath289 as @xmath353 @xcite .
for the gluon loop , we can use the nlo bfkl results as an independent calculation . let us repeat the arguments of ref .
@xcite for this case .
the characteristic transverse scale inside the shock wave is small ( see the discussion in ref .
@xcite ) and therefore the contribution of the diagram in fig .
[ loopinside ] reduces to the contribution of some operator _ local _ in the transverse space .
this would bring the additional terms with the nontrivial @xmath354 dependence to the kernel which translates into the nontrivial additional @xmath351-dependent term in the eigenvalues .
such terms do not exist and therefore the gluon interaction with the shock wave does not get an extra @xmath352 correction .
in this section we briefly summarize the results of the renormalon - based analysis of the argument of the coupling constant carried in refs .
@xcite to get an argument of coupling constant we can trace the quark part of the @xmath32-function ( proportional to @xmath33 ) . in the leading log approximation
@xmath355 the quark part of the @xmath32-function comes from the bubble chain of quark loops in the shock - wave background .
we can either have no intersection of quark loop with the shock wave ( see fig .
[ quarkloops]a ) or we may have one of the loops in the shock - wave background ( see fig . [ quarkloops]b ) . .
] the sum of these diagrams yields @xmath356 [ e^{-i(p - l , x)_\perp}-e^{-i(p - l , y)_\perp } ] \nonumber\\ & & \hspace{-16 mm } \times~ { 1\over p^2(1+{\alpha_s\over 6\pi}\ln{\mu^2\over p^2 } ) } \big(1-{\alpha_sn_f\over 6\pi}\ln{l^2\over \mu^2}\big)\partial_\perp^2u^{ab}(l ) { 1\over ( p - l)^2(1+{\alpha_s\over 6\pi}\ln{\mu^2\over ( p - l)^2 } ) } \label{bubblechain1}\end{aligned}\ ] ] where we have left only the @xmath32-function part of the quark loop . replacing the quark part of the @xmath32-function @xmath357 by the total contribution
@xmath358 we get @xmath359 [ e^{-i(p - l ,
x)_\perp}-e^{-i(p - l , y)_\perp } ] { \alpha_s(p^2)\over p^2 } \alpha_s^{-1}(l^2)\partial_\perp^2u^{ab}(q ) { \alpha_s((p - l)^2)\over ( p - l)^2 } \label{bubblechain2}\end{aligned}\ ] ] in principle , one should also include the `` renormalon dressing '' of the double - log and conformal terms in eq .
( [ nlobk ] ) .
we think , however , that they form a separate contribution which has nothing to do with the argument of the bk equation . to go to the coordinate space , we expand the coupling constants in eq .
( [ bubblechain2 ] ) in powers of @xmath360 , i.e. return back to eq .
( [ bubblechain1 ] ) with @xmath361 .
unfortunately , the fourier transformation to the coordinate space can be performed explicitly only for a couple of first terms of the expansion @xmath362 . in the first order
we get the running - coupling part of the nlo bk equation ( [ nlobk ] ) @xmath363 \big[{(x - y)^2\over x^2 y^2 } \big(1+b{\alpha_s \over 4\pi}\ln ( x - y)^2\mu^2\big)-b { \alpha_s \over 4\pi}{x^2-y^2\over x^2y^2}\ln{x^2\over y^2}\big ] \nonumber\end{aligned}\ ] ] the result of the fourier transformation up to the second order has the form @xcite @xmath364\big\{{(x - y)^2\over x^2 y^2 } \big[1+{b\alpha_s\over 4\pi}\big(\ln ( x - y)^2\mu^2+{5\over 3}\big ) \label{chtoposchitano}\\ & & \hspace{-6 mm } + ~\big({b\alpha_s\over 4\pi}\big)^2\ln^2 ( x - y)^2\mu^2\big ] + { b\alpha_s\over 4\pi}{1\over x^2}\ln{x^2\over y^2 } \big[1+{b\alpha_s\over 4\pi}\ln ( x - y)^2\mu^2 + { b\alpha_s\over 4\pi}\ln x^2\mu^2\big]\big ] \nonumber\\ & & \hspace{-6 mm } -~{b\alpha_s\over 4\pi}{1\over y^2}\ln{x^2\over y^2}\big[1+{b\alpha_s\over 4\pi}\ln ( x - y)^2\mu^2 + { b\alpha_s\over 4\pi}\ln y^2\mu^2\big]\big\}+ ... \nonumber\end{aligned}\ ] ] we extrapolate the @xmath365 terms in the above equation as follows : @xmath366 \label{runningcoupling}\\ & & \hspace{-6 mm } \times~\big[{(x - y)^2\over x^2 y^2 } + { 1\over x^2}\big({\alpha_s(x^2)\over\alpha_s(y^2)}-1\big ) + { 1\over y^2}\big({\alpha_s(y^2)\over\alpha_s(x^2)}-1\big)\big]+ ... \end{aligned}\ ] ] where dots stand for the remaining conformal terms and @xmath367 term .
( here we promoted wilson lines in the r.h.s . to operators ) .
when the sizes of the dipoles are very different the kernel of the above equation reduces to @xmath368 in the earlier paper@xcite the eq .
( [ runningcoupling ] ) was interpreted as an indication that the argument of the coupling constant is the size of the parent dipole @xmath369 .
we are grateful to g. salam for pointing out that the proper interpretation is the size of the smallest dipole as follows from eq .
( [ limits ] ) .
it is instructive to compare our result to the paper @xcite where the nlo bk equation is rewritten in terms of three effective coupling constants .
the authors of ref .
@xcite extrapolate eq .
( [ chtoposchitano ] ) in a different way @xmath370 \label{ixinterpretation}\\ & & \hspace{-6 mm } \times~\big [ { 1\over x^2}\alpha_s(x^2 ) + { 1\over y^2 } \alpha_s(y^2 ) - { 2(x - z , y - z)\over x^2 y^2 } { \alpha_s(x^2 ) \alpha_s(y^2)\over \alpha_s(r^2 ) } \big ] \nonumber\end{aligned}\ ] ] where @xmath371 is some scale interpolating between @xmath372 and @xmath373 ( the explicit form can be found in ref .
@xcite ) . theoretically , until the fourier transformations in all orders in @xmath374 are performed , both of these interpretations are models of the high - order behavior of running coupling constant
. the convenience of these models can be checked by the numerical estimates of the size of the neglected term(s ) in comparison to terms taken into account by the model , see the discussion in refs .
we have calculated the nlo kernel for the evolution of the color dipole .
it consists of three parts : the running - coupling part proportional to @xmath32-function ( see diagrams shown in fig .
[ 1cutdms ] ) , the conformal part describing 1 @xmath183 3 dipoles transition ( diagrams in fig .
[ 2cutdms ] ) and the non - conformal term coming from the diagrams in fig .
( [ 1cutadd ] ) .
the result agrees with the forward nlo bfkl kernel @xcite up to a term proportional @xmath375 times the original dipole .
we think that the difference could be due to different definitions of the cutoff in the longitudinal momenta ( see the discussion in previous section ) .
it would be instructive to get the @xmath350 asymptotics of the anomalous dimensions of gluon operators directly from eq .
( [ nlobk ] ) , without a fourier transformation of our result to the momentum space and comparing to nlo bfkl as it is done in sect .
[ sect : compare ] .
the study is in progress .
there is a recent paper @xcite where the dipole form of the non - forward nlo bfkl kernel is calculated using the non - forward nlo bfkl kernel@xcite .
the kernel obtained in @xcite is different from our result ( and not conformally invariant ) .
we think that at least part of the difference is coming from the fact that the evolution kernel ( [ nlobk ] ) should be compared to the non - symmetric `` evolution '' nlo bfkl kernel @xmath376 rather that to the symmetric kernel @xmath321 defined by eq .
( [ lip1 ] ) .
the kernel @xmath377 corresponds to the green function @xmath378 defined by eq .
( [ lip1 ] ) with different lower cutoff for the longitudinal integration @xmath379 the @xmath380 satisfies the equation ( [ wgw ] ) with the kernel @xmath381 @xmath382 and the relation between @xmath383 and @xmath321 has the form ( cf .
@xcite ) @xmath384 it is easy to see that the structure ( [ getilde ] ) repeats itself after differentiation with respect to @xmath348 so it can be rewritten as an evolution equation for @xmath385 ( whereas the derivative of the original formula ( [ lip1 ] ) does not have the structure of the evolution equation due to an extra @xmath386 ) . in terms of eigenvalues ,
the modified kernel ( [ ktilde ] ) lead to the shifts of the type @xmath387 which we saw in sect .
[ sect : compare]b .
it should be emphasized that the conformally invariant nlo kernel describes the evolution of the light - like wilson lines with the `` rigid '' cutoff in the longitudinal momenta ( [ cutoff ] ) . on the contrary , for dipoles with the non - light - like slope the sum of the diagrams in fig .
[ 2cutdms ] is not conformally invariant ( see appendix ) .
the reason is that a general wilson line is a non - local operator which is not conformally invariant to begin with - for example , the non - light - like wilson line turns into a circle under the inversion @xmath388 . with the light - like wilson lines
, the situation is different .
formally , a wilson line @xmath389~=~{\rm pexp}~\big\{ig\!\int_{-\infty}^\infty\!dx^+~a_+(x^+,x_\perp)\big\}\ ] ] is invariant under the inversion @xmath388 ( with respect to the point with zero ( - ) component ) .
indeed , @xmath390 so after the inversion @xmath391 and @xmath392 and therefore @xmath389~\rightarrow~{\rm pexp}~\big\{ig\!\int_{-\infty}^\infty\!d{x^+\over x_\perp^2}~a_+({x^+\over x_\perp^2},x_\perp)\big\ } ~=~[\infty p_1+x_\perp,-\infty p_1+x_\perp]\ ] ] thus , it is not surprising that the bulk of our nlo kernel for the light - like dipoles is conformally invariant in the transverse space .
the part proportional to the @xmath32-function is not conformally invariant and should not be , but there is another term @xmath393 which is not invariant .
the reason for that is probably the cutoff @xmath182 which can be expressed as a cutoff in longitudinal coordinate @xmath394 , and therefore under the inversion @xmath392 the cutoff can pick up some logs of transverse separations .
it is worth noting that conformal and non - conformal terms come from graphs with different topology : the conformal terms come from 1@xmath1833 dipoles diagrams in fig .
( [ 2cutdms ] ) which describe the dipole creation while the non - conformal double - log term comes from the1@xmath1832 dipole transitions ( see fig .
[ 1cutadd ] ) which can be regarded as a combination of dipole creation and dipole recombination .
it is possible that in the effective action language , symmetric with respect to the projectile and the target @xcite , the evolution kernel is conformally invariant .
we hope to study this problem in a separate publication .
the authors are indebted to yu.v .
kovchegov , l.n .
lipatov , g. salam and h. weigert for valuable discussions .
i.b . would like to thank e. iancu and other members of theory group at cea saclay for for valuable discussions and kind hospitality .
this work was supported by contract de - ac05 - 06or23177 under which the jefferson science associates , llc operate the thomas jefferson national accelerator facility .
as we mentioned above , it is convenient to separate the uv - divrgent and uv - finite parts of the eq .
( [ 2cutsum ] ) by writing down @xmath395 .
the contribution of the first part leads to eq .
( [ 2cut ] ) while the second uv - divergent term have the same color structure as the leading - order bk equation . after replacing @xmath396 by @xmath397 , integrating over @xmath83 with the prescription ( [ pluscription ] ) and changing variables to @xmath398 , @xmath399 , @xmath400 ( so that @xmath401 , @xmath402 and @xmath403 ) the eq .
( [ 2cutsum ] ) turns into @xmath404 \nonumber\\ & & \hspace{-2 mm } + ~\!\int\ !
{ { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 p{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2l{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k'\big\{(e^{i(p , x)}-e^{i(p , y ) } ) ( e^{-i(p - l - k',x)-i(k',y)}-e^{-i(p - l - k',y)-i(k',x ) } ) \nonumber\\ & & \hspace{32 mm } \times~ { ( k',p - l - k')(p - k')^2 - 2(p - k',p - l - k')(k',p - k')\over p^2(p - k')^2{k'}^2(p - l - k')^2}\ln{(p - k')^2\over { k'}^2 } \label{a1}\end{aligned}\ ] ] where @xmath405 and @xmath406 } \label{f2}\\ & & \hspace{-12 mm } + ~\big({(p - l)_i\over ( p - l)^2}-{(p - l - k')_i\over ( p - k'-l)^2}\big ) { 1/{k'}^2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2 } \bigg ] \nonumber\\ & & \hspace{-12 mm } + ~{2\over p^2}\bigg [ { 2(l , p - k'-l)(p , k')/{k'}^2\over ( p - k'-l)^2[(p - k')^2-(p - k'-l)^2 ] } \ln { ( p - k')^2\over ( p - k'-l)^2 } -{(p - k',p - l - k')(p , k')\over { k'}^2(p - k')^2(p - k'-l)^2 } \ln { ( p - k')^2\over { k'}^2 } \nonumber\\ & & \hspace{-12 mm } + ~{(l , k')/{k'}^2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2 } -{(p - k'-l , k')\over 2{k'}^2(p - l - k')^2}\ln { ( p - k')^2\over { k'}^2 } + { ( p - l , k')/{k'}^2\over ( p - l - k')^2-{k'}^2}\ln { ( p - k'-l)^2\over { k'}^2 } \nonumber\\ & & \hspace{-12 mm } + ~{2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2}- { 2\ln ( p - l - k')^2/ { k'}^2\over ( p - l - k')^2-{k'}^2 } + { ( p , p - l - k')\over ( p - l - k')^2{k'}^2}\ln { ( p - k')^2\over p^2 } \bigg ] \nonumber\\ & & \hspace{-12 mm } + ~{2\over ( p - l)^2}\bigg [ { -2(l , p - k')(p - l , k')/{k'}^2\over ( p - k')^2[(p - k')^2-(p - k'-l)^2 ] } \ln { ( p - k')^2\over ( p - k'-l)^2 } -{(p - k',p - l - k')(p - l , k')\over { k'}^2(p - k')^2(p - k'-l)^2 } \ln { ( p - k'-l)^2\over { k'}^2 } \nonumber\\ & & \hspace{-12 mm } -~{(l , k')/{k'}^2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2 } + { ( p , k')/{k'}^2\over ( p - k')^2-{k'}^2}\ln { ( p - k')^2\over { k'}^2 } -{(p - k',k')\over 2{k'}^2(p - k')^2}\ln { ( p - l - k')^2\over { k'}^2 } \nonumber\\ & & \hspace{-12 mm } + ~{2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2}- { 2\ln ( p - k')^2/ { k'}^2\over ( p - k')^2-{k'}^2 } + { ( p - l , p - k')\over ( p - k')^2{k'}^2}\ln { ( p - l - k')^2\over ( p - l)^2 } \bigg]\bigg)\end{aligned}\ ] ] we need to perform the integration over @xmath247 .
let us start with the uv - divergent term @xmath407 .
using the integrals @xmath408 \nonumber\\ & & \hspace{-2 mm } 4\pi\!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^dk'{(k',k'-l)\over { k'}^2(l - k')^2}\ln{(p - k')^2\over { k'}^2}~=~{1\over 2}\ln{p^2\over l^2}\ln{(p - l)^2\over l^2 } \nonumber\\ & & \hspace{-2 mm } 4\pi\!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^dk'{(p - k',p - k'-l)\over ( p - k')^2 ( p - k'-l)^2 } \ln{(p - k')^2(p - k'-l)^2\over { k'}^4}~ = ~-\ln{p^2\over l^2}\ln{(p - l)^2\over l^2 } \nonumber\\ & & \hspace{-2 mm } 4\pi\!\int\!{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^dk'{(p , p - l - k')\over ( p - l - k')^2{k'}^2}\ln { p^2\over { k'}^2 } \nonumber\\ & & \hspace{-2mm}=~ ( p , p - l){\gamma({d\over 2})\gamma({d\over 2}-1)\over\gamma(d-1 ) } { \gamma(2-{d\over 2})\over |p - l|^{4-d}}\big[\ln{p^2\over ( p - l)^2}+{2\over d-2 } + \psi(2-{d\over 2})+\psi(d-1)-\psi({d\over 2})-\psi(1)\big ] \label{kintegrals1}\end{aligned}\ ] ] one obtains @xmath409 let us at first consider the uv - divergent contribution @xmath410 \nonumber\\ & & \hspace{-12 mm } \label{uvpart}\end{aligned}\ ] ] to this contribution we should add the counterterm corresponding to quark and gluon loops lying inside the shock wave . the rigorous calculation of the counterterm was performed in ref . @xcite and the result is @xmath411 where we need the gluon part of @xmath35 ( @xmath412 ) . after subtraction of the counterterm ( [ ct1 ] )
the uv - divergent contribution ( [ uvpart ] ) reduces to @xmath413 \nonumber\\ & & \hspace{-12 mm } \label{uvpart1}\end{aligned}\ ] ] so one obtains the regularized @xmath414 in the form @xmath415 \label{f1reg}\ ] ] it is convenient to calculate first the fourier transform with @xmath416 . using the integrals @xmath417
we get @xmath418 -{1\over 48\pi^3}\big[{1\over x^2}\ln{y^2\over \delta^2 } + { 1\over y^2}\ln{x^2\over \delta^2}\big ] \nonumber\end{aligned}\ ] ] hereafter we use the notation @xmath419 .
next we calculate the @xmath420 contribution .
we need the following fourier integrals : @xmath421 } \nonumber\\ & & \hspace{-12 mm } + ~\big({(p - l)_i\over ( p - l)^2}-{(p - l - k')_i\over ( p - k'-l)^2}\big ) { 1/{k'}^2\over ( p - k')^2-(p - k'-l)^2 } \ln { ( p - k')^2\over ( p - l - k')^2 } \bigg ] ~ = ~{(x , y)\over 16\pi^3x^2y^2 } \big[\ln^2{x^2\over y^2}+{2\pi^2\over 3}\big ] \label{term2}\end{aligned}\ ] ] @xmath422 \nonumber\\ & & \hspace{-2mm}=~ { 1\over 16\pi^3 } \bigg\{-{\pi^2\over 3}{(x+y)^2\over x^2y^2}+~{2\over y^2}\!\int_0 ^ 1\!du{\ln u\over u-{x^2\over x^2-y^2 } } + ~{2\over
x^2}\!\int_0 ^ 1\!du{\ln u\over u+{y^2\over x^2-y^2 } } \nonumber\\ & & \hspace{-2 mm } + ~{i\kappa\over x^2y^2 } \bigg(2\!\int_0 ^ 1\!du\big[{\ln u\over u-{(x,\delta)-i\kappa\over \delta^2 } } + { \ln u\over u+{(y,\delta)+i\kappa\over \delta^2}}-c.c.\big ] + \ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa \over ( x , y)-i\kappa}\bigg)\bigg\ } \label{term3 } \end{aligned}\ ] ] where @xmath423 , and @xmath424 \nonumber\\ & & \hspace{-2 mm } = ~{1\over 32\pi^3x^2y^2}\big[x^2\ln^2{x^2\over \delta^2}+y^2\ln^2{y^2\over \delta^2 } + 2(x , y)\ln{x^2\over\delta^2}\ln{y^2\over\delta^2}\big ] \nonumber\\ & & \hspace{22 mm } + ~{i\kappa\over 16\pi^3x^2y^2 } \!\int_0 ^ 1\!du\big[{\ln u\over u-{(\delta , x)+i\kappa\over \delta^2 } } + { \ln u\over u+{(\delta , y)-i\kappa\over \delta^2}}-c.c .
-{1\over 2}\ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa}\big ] \nonumber\end{aligned}\ ] ] @xmath425 } \ln { ( p - k')^2\over ( p - l - k')^2 } + { 2(p - l , k')p^{-2}\over { k'}^2[(p - k'-l)^2-{k'}^2 ] } \nonumber\\ & & \hspace{-6 mm } \times~\ln { ( p - l - k')^2\over { k'}^2}-{2(l , k')(p - l)^{-2}\over { k'}^2[(p - k')^2-(p - k'-l)^2 ] } \ln { ( p - k')^2\over ( p - l - k')^2 } + { 2(p , k')(p - l)^{-2}\over { k'}^2[(p - k')^2-{k'}^2]}\ln { ( p - k')^2\over { k'}^2}\big ) \nonumber\\ & & \hspace{-6 mm } = ~-{1\over 16\pi^3}\big({1\over y^2}\ln{x^2\over\delta^2 } -{1\over x^2}\ln{y^2\over\delta^2}\big)\ln{x^2\over y^2 } + { 1\over 8\pi^3}\bigg[{\pi^2\over 6x^2}+{\pi^2\over 6y^2 } -{1\over x^2}\!\int_0 ^ 1\!{du\ln u \over u+{y^2\over x^2-y^2 } } -{1\over y^2}\!\int_0 ^ 1\!{du\ln u \over u-{x^2\over x^2-y^2}}\bigg ] \label{term5}\end{aligned}\ ] ] @xmath426 \nonumber\\ & & \hspace{-2 mm } = ~{1\over 32\pi^3}\ln{x^2\over y^2}\big[{1\over y^2}\ln{x^2\over \delta^2 } -{1\over x^2}\ln{y^2\over \delta^2}\big ] \label{term6}\end{aligned}\ ] ] @xmath427 } \label{term7}\\ & & \hspace{-2 mm } + ~{4\over ( p - l)^2[(p - k')^2-(p - k'-l)^2 ] } \ln { ( p - k')^2\over ( p - l - k')^2}- { 4\ln ( p - k')^2/ { k'}^2\over ( p - l)^2[(p - k')^2-{k'}^2]}~ = ~{1\over 4\pi^3y^2}\ln{x^2\over\delta^2}+ { 1\over 4\pi^3x^2}\ln{y^2\over\delta^2 } \nonumber\end{aligned}\ ] ] @xmath428 \label{term8}\\ & & \hspace{-2 mm } = ~ { i\kappa\over 16\pi^3x^2y^2}\bigg\ { 2\!\int_0 ^ 1\ !
du\big [ { \ln u\over u-{(x,\delta)+i\kappa\over \delta^2}}+{\ln u\over u+{(y,\delta)-i\kappa\over \delta^2 } } -~c.c.\big ] -\ln{x^2y^2\over \delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa } \bigg\ } + { ( x , y)\over 16\pi^3x^2y^2}\ln^2{x^2\over y^2 } \nonumber\end{aligned}\ ] ] adding the integrals ( [ term2 ] ) - ( [ term6 ] ) we obtain @xmath429 + { ( x+y)^2\over 32\pi^3x^2y^2}\ln{x^2\over\delta^2}\ln{y^2\over\delta^2 } + { ( x , y)\over 8\pi^3x^2y^2}\ln^2{x^2\over y^2 } \nonumber\\ & & \hspace{-12 mm } + ~{i\kappa\over 16\pi^3x^2y^2}\bigg\{\!\int_0 ^ 1\!du\big[{\ln u\over u-{(\delta , x)+i\kappa\over \delta^2 } } + { \ln u\over u+{(\delta , y)-i\kappa\over \delta^2}}-c.c.\big ] -{1\over 2}\ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa } \bigg\ } \nonumber\end{aligned}\ ] ] and therefore @xmath430 ~ \nonumber\\ & & \hspace{-2mm}= ~-{1\over 8\pi^3}{(x , y)\over x^2y^2 } \big[{11\over 3}\ln{x^2y^2\over \delta^2}\mu^2+{67\over 9}-{\pi^2\over 3}\big ] + { 1\over 32\pi^3}{\delta^2\over x^2y^2}\ln{x^2\over\delta^2}\ln{y^2\over\delta^2 } + { 11\over 48\pi^3}\big[{1\over x^2}\ln{y^2\over\delta^2 } + { 1\over y^2}\ln{x^2\over \delta^2}\big ] \nonumber\\ & & \hspace{-2 mm } + ~{i\kappa\over 16\pi^3x^2y^2}\bigg\{\!\int_0 ^ 1\!du\big[{\ln u\over u-{(\delta , x)+i\kappa\over \delta^2 } } + { \ln u\over u+{(\delta , y)-i\kappa\over \delta^2}}-c.c.\big ] -{1\over 2}\ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa } \bigg\ } \label{f12}\end{aligned}\ ] ] note that the r.h.s . of this equation
is finite as @xmath431 ( taken separately , the contributions of @xmath414 and @xmath420 are singular in this limit ) : @xmath432 ~=~-{1\over 8\pi^3}{1\over x^2 } \big[{11\over 3}\ln x^2\mu^2+{67\over 9}-{\pi^2\over 3}\big ] \label{f12xy}\end{aligned}\ ] ] using eqs .
( [ f12 ] ) and ( [ f12xy ] ) we obtain @xmath430 ~(e^{i(p , x)}-e^{i(p , y)})(e^{-i(p - l , x)}-e^{-i(p - l , y ) } ) \nonumber\\ & & \hspace{-2mm}= ~-{1\over 8\pi^3}{\delta^2\over x^2y^2 } \big[{11\over 3}\ln{x^2y^2\over \delta^2}\mu^2+{67\over 9}-{\pi^2\over 3}\big ] -{1\over 16\pi^3}{\delta^2\over x^2y^2}\ln{x^2\over\delta^2}\ln{y^2\over\delta^2 } \nonumber\\ & & \hspace{-2 mm } -~{i\kappa\over 8\pi^3x^2y^2}\bigg\{\!\int_0 ^ 1\!du\big[{\ln u\over u-{(\delta , x)+i\kappa\over \delta^2 } } + { \ln u\over u+{(\delta , y)-i\kappa\over \delta^2}}-c.c.\big ] -{1\over 2}\ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa } \bigg\ } \label{f12contribution}\end{aligned}\ ] ] now we turn our attention to last two terms in eq .
( [ a1 ] ) . using fourier transformation
@xmath433 \nonumber\\ & & \hspace{72mm}+~{1\over 2}\ln{x_1 ^ 2\over x_{12}^2 } \ln{(x_2,x_{12})+i\kappa_{12}\over ( x_2,x_{12})-i\kappa_{12}}\bigg\ } + { ix_{12i}\over 16\pi^2x_{12}^2}\ln{x_1 ^ 2\over x_{12}^2}\ln{x_1 ^ 2\over x_2 ^ 2 } \nonumber\end{aligned}\ ] ] ( where @xmath434 and @xmath435 ) one easily obtains @xmath436[(\delta , y)+i\kappa]\over[(\delta , x)-i\kappa][(\delta , y)-i\kappa ] } \bigg ] \nonumber\\ & & \hspace{-2 mm } = ~{i\kappa\over 16\pi^3x^2y^2}\bigg[-\!\int_0 ^ 1\!du\big[{\ln u\over u-{(\delta , x)+i\kappa\over \delta^2 } } + { \ln u\over u+{(\delta , y)-i\kappa\over \delta^2}}-c.c.\big ] -{1\over 2}\ln{x^2y^2\over\delta^4}\ln{(x , y)+i\kappa\over ( x , y)-i\kappa}\bigg ]
-{(x , y)\over 32\pi^3 x^2y^2}\ln^2{x^2\over y^2}\nonumber\\ \label{1kycok}\end{aligned}\ ] ] similarly , @xmath437 + { 1\over 2}\ln{x_1 ^ 2\over x_{12}^2 } \ln{(x_2,x_{12})+i\kappa\over ( x_2,x_{12})-i\kappa}\bigg ] \nonumber\\ & & \hspace{-6 mm } -~{i\over 16\pi^2 } \big(x_{2i}-{(x_2,x_{12})\over x_{12}^2}x_{12i}\big){1\over i\kappa}\bigg [ \int_0 ^ 1\ ! du \big[{\ln
u\over u+{(x_2,x_{12})-i\kappa\over x_2 ^ 2}}-c.c.\big ] + { 1\over 2}\ln{x_2 ^ 2\over x_{12}^2 } \ln{(x_1,x_{12})+i\kappa\over ( x_1,x_{12})-i\kappa}\bigg ] + { ix_{12i}\over 32\pi^2x_{12}^2}\ln{x_1 ^ 2x_2 ^ 2\over x_{12}^4}\ln{x_1 ^ 2\over x_2 ^ 2 } \nonumber\\ & & \hspace{-6 mm } + ~{i\over 16\pi^2 } \big(x_{12i}-{(x_1,x_{12})\over x_1 ^ 2}x_{1i}\big){1\over i\kappa}\bigg [ \int_0 ^ 1\ ! du \big[{\ln u\over u-{(x_1,x_{12})-i\kappa\over x_{12}^2}}-c.c.\big ] -{1\over 2}\ln{x_1 ^ 2\over x_{12}^2 } \ln{(x_1,x_2)+i\kappa\over ( x_1,x_2)-i\kappa}\bigg ] -{ix_{1i}\over 32\pi^2x_1 ^ 2}\ln{x_1 ^ 2\over x_2 ^ 2}\ln{x_2 ^ 2\over x_{12}^2 } \nonumber\end{aligned}\ ] ] and therefore @xmath438 \nonumber\\ & & \hspace{-2 mm } -~{(x , y)\over 32\pi^3 x^2y^2}\ln{x^2\delta^2\over y^4}\ln{x^2\over\delta^2 } -{(x , y)\over 32\pi^3 x^2y^2}\ln{y^2\delta^2\over x^4}\ln{y^2\over\delta^2 } -{1\over 32\pi^3 } \big({1\over x^2}+{1\over y^2}\big ) \ln{x^2\over y^2}\ln{y^2\over\delta^2 } \label{2kycok}\end{aligned}\ ] ] adding the equations ( [ 1kycok ] ) and ( [ 2kycok ] ) we obtain @xmath439 + { \delta^2\over 32\pi^3 x^2y^2}\ln{x^2\over \delta^2}\ln{y^2\over\delta^2 } \nonumber\end{aligned}\ ] ] it is easy to see that the contribution of the last term in eq . ( [ a1 ] ) is equal to ( [ summa12 ] ) so we get @xmath440 \nonumber\\ & & \hspace{-2 mm } \times~\bigg [ \int\ ! { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 p{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2l~[f_1^{\rm reg}(p , l)+f_2(p , l ) ] ~(e^{i(p , x)}-e^{i(p , y)})(e^{-i(p - l , x)}-e^{-i(p - l , y ) } ) \nonumber\\ & & \hspace{-2 mm } + ~2\!\int\ ! { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 p{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2l{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k'(e^{-i(p - l , x)}-e^{-i(p - l , y)})(e^{i(p - k',x)+i(k',y)}-e^{i(p - k',y)+i(k',x ) } ) \nonumber\\ & & \hspace{32 mm } \times~{(k',p - k')(p - k')^2 - 2(p - k',p - l - k')(k',p - l - k')\over ( p - l)^2(p - k')^2{k'}^2(p - l - k')^2}\ln{(p - l - k')^2\over { k'}^2}\bigg ] \nonumber\\ & & \hspace{-2 mm } = ~~-{\alpha_s^2n_c\over 8\pi^3 } \!\int d^2z~ [ { \rm tr}\{u_xu^\dagger_z\}{\rm tr}\{u_zu^\dagger_y\ } -{1\over n_c}{\rm tr}\{u_xu^\dagger_y\ } ] { \delta^2\over x^2y^2 } \big[{11\over 3}\ln{x^2y^2\over \delta^2}\mu^2+{67\over 9}-{\pi^2\over 3}\big ] \label{appendixresult}\end{aligned}\ ] ] note that the dilogarithms and products of logarithms have canceled .
the simplicity of the final result indicates that there should be a less tedious derivation but we were not able to find it .
we will repeat the procedure from sect .
( [ sect : losubtraction ] ) , this time using the cutoff by the slope . @xmath441 instead of eq .
( [ selfenergy1 ] ) we get @xmath442[\alpha\beta ' s-(k_1+k_2)_\perp^2+i\epsilon ] } \nonumber\\ & & \hspace{-6 mm } { d_{\mu\xi}(\alpha_1 p_1+\beta_1 p_2+q_{1\perp } ) \over \alpha_1\beta_1 s - q_{1\perp}^2+i\epsilon } { d^\xi_{~\mu'}(\alpha_1 p_1+\beta'_1 p_2+k_{1\perp } ) \over \alpha_1\beta'_1 s - k_{1\perp}^2+i\epsilon } { d_{\nu\eta}((\alpha-\alpha_1 ) p_1+\beta_2 p_2+q_{2\perp } ) \over ( \alpha-\alpha_1)\beta_2 s - q_{2\perp}^2+i\epsilon } ~{d^\eta_{~\nu'}((\alpha-\alpha_1)p_1+\beta'_2 p_2+k_{2\perp } ) \over ( \alpha-\alpha_1)\beta'_2 s - k_{2\perp}^2+i\epsilon } \nonumber\\ & & \hspace{-6 mm } \gamma^{\mu\nu\lambda } ( \alpha p_1+q_{1\perp},(\alpha-\alpha_1)p_1+q_{2\perp},-\alpha p_1-q_{1\perp } -q_{2\perp})~ \gamma^{\mu'\nu'\lambda ' } ( \alpha p_1+k_{1\perp},(\alpha-\alpha_1)p_1 + k_{2\perp},-\alpha p_1-k_{1\perp}-k_{2\perp } ) \nonumber\end{aligned}\ ] ] where @xmath443 . in this formula
@xmath444 comes from the integration over @xmath83 parameter in the l.h.s . and
@xmath445 from the integration over @xmath87 parameter . taking residues at @xmath446 and @xmath447 and @xmath94 , @xmath95 we obtain @xmath448 which leads to ( cf .
( [ selfenergy4 ] ) @xmath449[(k_1+k_2)^2+\xi\alpha^2 ] ( q_1 ^ 2\bar{u}+q_2 ^ 2u)(k_1 ^ 2\bar{u}+k_2 ^ 2u ) } \nonumber\\ & & \hspace{-2mm}\times~ \big[(q_1 ^ 2-q_2 ^ 2)\delta_{ij}-{2\over u}q_{1i}(q_1+q_2)_j+ { 2\over \bar{u}}(q_1+q_2)_iq_{2j}\big ]
\big[(k_1 ^ 2-k_2 ^ 2)\delta_{ij}-{2\over u}k_{1i}(k_1+k_2)_j+ { 2\over \bar{u}}(k_1+k_2)_ik_{2j}\big ] \nonumber\\ & & \hspace{-2 mm } ~=~{g^4\over 4\pi^2}{\rm tr}\{t^au_xt^bu^\dagger_y\}f^{anl}f^{bn'l ' } \!\int\ !
d^2z d^2z'u_z^{nn'}u_{z'}^{ll ' } \nonumber\\ & & \hspace{-2 mm } \times~\xi{d\over d\xi}\ !
d\alpha\!\int_0^\alpha\ ! \!\!d\alpha '
\!\int\ ! { { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 q_1{{\textstyle d }
\lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q_2{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_1{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_2~ { ( \alpha-\alpha')\alpha'e^{i(q_1,x - z)_\perp+i(q_2,x - z')_\perp -i(k_1,y - z)_\perp - i(k_2,y - z')_\perp } \over [ ( q_1+q_2)^2+\xi\alpha^2][(k_1+k_2)^2+\xi\alpha^2 ] ( q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha')(k_1 ^ 2(\alpha-\alpha')+k_2 ^ 2\alpha ' ) } \nonumber\\ & & \hspace{-2mm}\times~ \big[{\delta_{ij}\over\alpha}(q_1 ^ 2-q_2 ^ 2)-{2\over \alpha'}q_{1i}(q_1+q_2)_j+ { 2\over \alpha-\alpha'}(q_1+q_2)_iq_{2j}\big ] \big[{\delta_{ij}\over\alpha}(k_1 ^ 2-k_2 ^ 2)-{2\over \alpha'}k_{1i}(k_1+k_2)_j+ { 2\over \alpha-\alpha'}(k_1+k_2)_ik_{2j}\big ] \nonumber\end{aligned}\ ] ] ( recall that @xmath450 ) . the contribution which is sensitive to the subtraction of ( lo)@xmath186 is @xmath451[(k_1+k_2)^2+\xi\alpha^2 ] ( q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha')(k_1 ^ 2(\alpha-\alpha')+k_2 ^ 2\alpha ' ) } \nonumber\\ & & \hspace{-4mm}\times~ \big[{\alpha\over\alpha'}(q_1,k_1)+{\alpha\over\alpha-\alpha'}(q_2,k_2)\big](q_1+q_2,k_1+k_2 ) \label{sensi}\end{aligned}\ ] ] the `` + ' ' -prescription ( [ pluscription ] ) leads to the subtraction @xmath452[(k_1+k_2)^2+\xi\alpha^2]}\bigg\{\!\int_0^\alpha\ ! { d\alpha'\over\alpha ' } \big [ { \alpha^2(q_1,k_1)\over ( q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha ' ] [ k_1 ^ 2(\alpha-\alpha')+k_2 ^ 2\alpha']}- { ( q_1,k_1)\over q_1 ^ 2k_1 ^ 2}\big ] \nonumber\\ & & \hspace{42 mm } + ~\!\int_0^\alpha\ ! { d\alpha'\over\alpha-\alpha ' } \big[{\alpha^2(q_2,k_2)\over [ q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha ' ] [ k_1 ^ 2(\alpha-\alpha')+k_2 ^
2\alpha']}- { ( q_2,k_2)\over q_2 ^ 2k_2 ^ 2}\big]\bigg\ } \label{b7}\end{aligned}\ ] ] the details of the upper cutoff in @xmath65 do not matter since they correspond to changes in the impact factor which do not affect the evolution .
for example , @xmath453[(k_1+k_2)^2+\xi\alpha^2 ] } \!\int_0^\alpha\ ! { d\alpha'\over\alpha'}\big [ { \alpha^2(q_1,k_1)\over ( q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha ' ] [ k_1 ^ 2(\alpha-\alpha')+k_2 ^ 2\alpha']}- { ( q_1,k_1)\over q_1
^ 2k_1 ^ 2}\big ] \nonumber\\ & & \hspace{-2 mm } = ~ { 1\over ( q_1+q_2)^2(k_1+k_2)^2 } \!\int_0 ^ 1 { du\over u}\big [ { ( q_1,k_1)\over ( q_1 ^ 2\bar{u}+q_2 ^ 2u ] [ k_1 ^ 2\bar{u}+k_2 ^ 2u]}- { ( q_1,k_1)\over q_1
^ 2k_1 ^ 2}\big ] \nonumber\\ & & \hspace{-2 mm } = ~{d\over d\sigma}\!\int_0^\sigma\ ! { d\alpha\over\alpha } { 1\over ( q_1+q_2)^2(k_1+k_2)^2 } \!\int_0^\alpha\ ! { d\alpha'\over\alpha'}\big [ { \alpha^2(q_1,k_1)\over ( q_1 ^ 2(\alpha-\alpha')+q_2 ^ 2\alpha ' ] [ k_1 ^ 2(\alpha-\alpha')+k_2 ^ 2\alpha']}- { ( q_1,k_1)\over q_1
^ 2k_1 ^ 2}\big ] \label{b8}\end{aligned}\ ] ] where the last line is exactly our `` rigid cutoff '' with `` + '' subtraction ( [ pluscription ] ) .
on the contrary , the details of the upper cutoff in @xmath454 are essential for the evolution equation ( [ subtractionapp ] ) .
the contribution to @xmath455 corresponding to the `` slope '' cutoff ( [ defy ] ) has the form @xmath456 \nonumber\end{aligned}\ ] ] and therefore the difference between the subtractions in `` rigid cutoff '' ( [ b7 ] ) and `` slope cutoff '' ( [ b8 ] ) prescriptions can be written as @xmath457 \nonumber\\ & & \hspace{-2 mm } + ~2\xi{d\over d\xi}\!\int_0^\infty\ !
{ d\alpha\over\alpha}~{(q_1+q_2,k_1+k_2)\over [ ( q_1+q_2)^2+\xi\alpha^2][(k_1+k_2)^2+\xi\alpha^2 ] } \big[{(q_1,k_1)\over q_1 ^ 2k_1 ^ 2}+{(q_2,k_2)\over q_2 ^ 2k_2 ^ 2}\big ] \!\int_0^\alpha\ ! { d\alpha'\over\alpha ' } \bigg\ } \nonumber\\ & & \hspace{-2 mm } = ~-{g^4\over 2\pi^2}{\rm tr}\{t^au_xt^bu^\dagger_y\}f^{anl}f^{bn'l ' } \!\int\ ! d^2z d^2z'u_z^{nn'}u_{z'}^{ll ' } \!\int\ !
{ { \textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2 q_1{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2q_2{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_1{{\textstyle d } \lower.03ex\hbox{\kern-0.40em$^{\scriptstyle-}$}\kern-0.08em{}}^2k_2~e^{i(q_1,x)_\perp+i(q_2,x')_\perp -i(k_1,y)_\perp - i(k_2,y')_\perp } \nonumber\\ & & \hspace{-2 mm } \times~\!\int_0^\infty\ ! { d\alpha'\over\alpha'}~ \bigg\{{(q_1+q_2,k_1+k_2)\over ( q_1+q_2)^2 ( k_1+k_2)^2 } \big[{(q_1,k_1)\over ( q_1 ^ 2+\xi{\alpha'}^2)(k_1 ^ 2+\xi{\alpha'}^2 ) } + { ( q_2,k_2)\over ( q_2 ^ 2+\xi{\alpha'}^2)(k_2 ^ 2+\xi{\alpha'}^2 ) } \big ] \nonumber\\ & & \hspace{-2 mm } -~{(q_1+q_2,k_1+k_2)\over [ ( q_1+q_2)^2+\xi{\alpha'}^2][(k_1+k_2)^2+\xi{\alpha'}^2 ] } \big[{(q_1,k_1)\over q_1 ^ 2k_1 ^ 2}+{(q_2,k_2)\over q_2 ^ 2k_2 ^ 2}\big ] \bigg\ } \nonumber\end{aligned}\ ] ] it is instructive to rewrite this result in schwinger s notations @xmath458 \nonumber\end{aligned}\ ] ] we see now that the difference between the two regularizations of the longitudinal divergence is given by the difference of ( lo)@xmath186 contributions with cutoffs in @xmath65 determined by the momenta on the first and on the second step of ( lo)@xmath186 evolution .
it is easy to see that for the sum of all diagrams this yields ( see eq .
( [ colors ] ) ) @xmath459\!\int_0^\infty\!{dt\over t } \nonumber\\ & & \hspace{-3 mm } \times~\big\{\big[(x|{p_i\over p^2+t}|z)-(y|{p_i\over p^2+t}|z)\big]^2 \big[(z|{p_i\over p^2}|z')-(y|{p_i\over p^2}|z')\big]^2 -\big[(x|{p_i\over p^2}|z)-(y|{p_i\over p^2}|z)\big]^2 \big[(z|{p_i\over p^2+t}|z')-(y|{p_i\over p^2+t}|z')\big]^2 \nonumber\\ & & \hspace{-3 mm } + ~\big[(x|{p_i\over p^2+t}|z')-(y|{p_i\over p^2+t}|z')\big]^2 \big[(z'|{p_i\over p^2}|z)-(x|{p_i\over p^2}|z)\big]^2 -\big[(x|{p_i\over p^2}|z')-(y|{p_i\over p^2}|z')\big]^2 \big[(z'|{p_i\over p^2+t}|z)-(x|{p_i\over p^2+t}|z)\big]^2\big\ } \nonumber\end{aligned}\ ] ] using the integral @xmath460 @xmath461 \nonumber\\ & & \hspace{-12 mm } \times~\bigg ( { y^2\over { y'}^2}\big\{(x , y)\big[{x^2+y^2\over x^2-y^2}\ln { x^2\over y^2}+2\big]+(\delta , y)\ln x^2-(\delta , x)\ln y^2\big\ } \nonumber\\ & & \hspace{-12 mm } -~{\delta^2\over { y'}^2 } \big\{(z - z',y')\big[{(z - z')^2+{y'}^2\over ( z - z')^2-{y'}^2}\ln { ( z - z')^2\over { y'}^2 } + 2\big]-(y , y')\ln ( z - z')^2+(y , z - z')\ln { y'}^2\big\}+x\leftrightarrow y\bigg ) \nonumber\end{aligned}\ ] ] the nlo kernel for the evolution of color dipoles with respect to the slope is the sum of eq .
( [ nlobk ] ) and the correction ( [ slopedif ] ) .
note that the correction term ( [ slopedif ] ) is not conformally invariant ( cf .
this is hardly surprising since the non - light - like wilson line turns into a circle under the inversion @xmath462 .
99 i. balitsky , _ `` high - energy qcd and wilson lines '' _ , in * shifman , m. ( ed . ) : at the frontier of particle physics , vol . 2 * , p. 1237 - 1342
( world scientific , singapore,2001 ) [ hep - ph/0101042 ] a.h .
mueller , _ nucl .
phys . _ * b415 * , 373 ( 1994 ) ; a.h .
mueller and bimal patel , _ nucl .
phys . _ * b425 * , 471 ( 1994 ) .
n.n . nikolaev and b.g .
zakharov , _ phys .
* b 332 * , 184 ( 1994 ) ; _ z. phys .
_ * c64 * , 631 ( 1994 ) ; n.n .
nikolaev b.g .
zakharov , and v.r .
zoller , _ jetp letters _ * 59 * , 6 ( 1994 ) .
kovchegov , _ phys .
rev . _ * d60 * , 034008 ( 1999 ) ; _ phys .
* d61*,074018 ( 2000 ) .
fadin , e.a .
kuraev , and l.n .
lipatov , _ phys .
* b 60 * , 50 ( 1975 ) ; i. balitsky and l.n .
lipatov , _ sov .
_ * 28 * , 822 ( 1978 )
gribov , e.m .
levin , and m.g .
ryskin , _ phys .
_ * 100 * , 1 ( 1983 ) , a.h .
mueller and j.w .
qiu , _ nucl .
_ * b268 * , 427 ( 1986 ) ; a.h .
mueller , _ nucl .
phys . _ * b335 * , 115 ( 1990 ) .
e. iancu and r. venugopalan , in * hwa , r.c .
et al . : quark gluon plasma * 249 - 3363 , [ e - print : hep - ph/0303204 ] ; + h. weigert , _
prog.part.nucl.phys._*55 * , 461(2005 ) ; + j. jalilian - marian and yu.v .
kovchegov , _
prog.part.nucl.phys._*56 * , 104(2006 ) .
m. beneke , _
phys.rept._*317*,1(1999 ) ; m. beneke and v.m .
braun , _ `` renormalons and power corrections . ''
_ , in * shifman , m. ( ed .
) : at the frontier of particle physics , vol . 3 * , p. 1719 - 1773
( world scientific , singapore,2001 ) [ hep - ph/0010208 ] v.s .
fadin and l.n .
lipatov , _ phys . lett._*b429 * , 127 ( 1998 ) ; g. camici and m. ciafaloni , _ phys .
lett._*b430 * , 349 ( 1998 ) .
i. balitsky , _ phys .
rev . _ * d60 * , 014020 ( 1999 ) .
fadin , m.i .
kotsky , and l.n .
`` gluon pair production in the quasimulti - regge kinematics '' _ , [ hep - ph/9704267 ] .
yu.v . kovchegov and h. weigert _
nucl.phys._*a789 * , 260(2007 ) .
a.v . kotikov and l.n .
lipatov , _ nucl .
_ * b582 * , 19 ( 2000 ) .
a. vogt , s. moch , and j.a.m .
vermaseren , _
nucl.phys._*b691 * , 129 ( 2004 ) v. s. fadin , r. fiore , a.v .
grabovsky , and a. papa , _
nucl.phys._*b784 * , 49(2007 ) .
v.s . fadin and r. fiore , _ phys .
_ * d72 * , 014018 ( 2005 ) .
a. kovner and m. lublinsky , _
phys.rev._*d71 * , 085004(2005 ) ; _ phys.rev.lett._*94 * , 181603(2005 ) ; _ jhep _ , 0503:001(2005 ) ; y. hatta , e. iancu , l. mclerran , a. stasto and d.n .
triantafyllopoulos , _
nucl.phys._*a764 * , 423 ( 2006 ) ; i. balitsky , _ phys .
_ * d72 * , 074027 ( 2005 ) : a.h .
mueller , a.i .
shoshi , and s.m.h .
nucl.phys._*b715 * , 440(2005 ) .
a. babansky and i. balitsky , _ phys .
rev . _ * d67 * , 054026 ( 2003 ) . | the small-@xmath0 deep inelastic scattering in the saturation region is governed by the non - linear evolution of wilson - line operators . in the leading logarithmic approximation
it is given by the bk equation for the evolution of color dipoles .
in the next - to - leading order the bk equation gets contributions from quark and gluon loops as well as from the tree gluon diagrams with quadratic and cubic nonlinearities .
we calculate the gluon contribution to small - x evolution of wilson lines ( the quark part was obtained earlier ) .
= 1 |
interactions between the octet baryons ( @xmath20 and @xmath17 ) and the @xmath5 cluster are important ingredients to consider the possible existence for various kinds of light hypernuclei through detailed microscopic cluster - model calculations .
if reliable effective @xmath21 interactions were known , one could easily calculate the @xmath22 potentials , using the standard cluster - model techniques .
unfortunately , this is not the case except for @xmath23 , since the bare @xmath21 interaction itself is not well known especially for the @xmath16 and @xmath17 hyperons , because of the technical difficulties of strangeness experiments .
even if these bare interactions were eventually known , we further need to develop the procedure to link the bare interactions and effective interactions through some effective interaction theory such as the @xmath3-matrix formalism .
we have recently developed the qcd - inspired spin - flavor @xmath4 quark model for the baryon - baryon interaction @xcite , which is a unified model for the full octet - baryons @xcite , and have achieved accurate descriptions of the @xmath24 and @xmath19 interactions @xcite . in particular , the @xmath24 interaction of the most recent model fss2 is accurate enough to compare with modern realistic meson - exchange models .
these quark - model interactions were used for the detailed study of few - baryon systems such as @xmath25 @xcite and @xmath26 @xcite , and also of some typical @xmath8-hypernuclei , @xmath27 @xcite and @xmath28 @xcite , through a newly developed three - cluster faddeev formalism @xcite and @xmath3-matrix calculations @xcite .
we can now use these baryon - baryon interactions to calculate not only the @xmath0 interaction , but also @xmath1 and @xmath2 interactions , assuming the harmonic - oscillator ( h.o . ) shell - model wave function for the @xmath5-cluster .
there are , in fact , a couple of advanced procedure to derive the interactions between a single baryon and finite nuclei based on the density - dependent hartree - fock theory of @xmath3-matrix interactions @xcite .
these approaches , however , use the localized @xmath3-matrix interaction in the configuration space and the center - of - mass ( c.m . ) coordinate system connected to the target nucleus . for the @xmath22 interactions , the c.m .
correction is quite important . in this paper
, we will derive @xmath22 born kernels , directly starting from the @xmath3-matrix calculation of the quark - model baryon - baryon interactions .
we deal with the nonlocality of the @xmath3-matrix and the c.m .
motion exactly , using the cluster - model approach , at the expense of the self - consistency between the @xmath5-cluster formation and the @xmath3-matrix interaction .
the @xmath3-matrix is pre - determined by solving the bethe - goldstone equation in symmetric nuclear matter , and the momentum - dependent single - particle ( s.p . )
potentials as well @xcite .
the fermi - momentum is assumed to be the standard value @xmath29 , but the obtained @xmath22 interactions depend on this choice rather monotonously , except for some special cases such as the @xmath0 @xmath30 interaction .
the treatment of the starting - energy dependence and the non - galilean invariant momentum dependence of the @xmath3-matrix is explicitly discussed .
we believe that these approximations are good enough to understand the unknown @xmath1 and @xmath2 interactions qualitatively , starting from the quark - model predictions of various baryon - baryon interactions . for the @xmath0 interaction ,
we compare the predictions by the present approach with some available phenomenological @xmath0 potentials @xcite , obtained by various methods .
another application of the present approach to the @xmath31 interaction will be published in a separate paper , since this system involves an extra nucleon - exchange term .
since the obtained @xmath22 interactions are all nonlocal , we calculate the wigner transform from the @xmath22 born kernels .
we find that the momentum dependence of the wigner transform is very strong , and the procedure to find effective local potential by solving the transcendental equation is necessary to obtain a local - potential image for the @xmath0 interaction .
the @xmath1 and @xmath2 interactions are repulsive , although the isospin @xmath12 ( for @xmath1 ) and @xmath13 ( for @xmath2 ) contributions of the @xmath10 and @xmath11 interactions are attractive . as the first step to study realistic @xmath1 and @xmath2 interactions
, we will only discuss the zero - momentum wigner transform in this paper .
although these @xmath22 interactions are complex due to the imaginary part of the underlying @xmath3-matrices , we discuss only the real part .
the spin - orbit @xmath22 potentials are naturally obtained from the @xmath30 and @xmath32 components of the @xmath3-matrix invariant interaction .
the central @xmath33 , @xmath34 and @xmath35 potentials were calculated by many authors with other ( usually more crude or purely phenomenological ) approaches .
the @xmath33 potential was calculated numerously , and different shapes were considered .
for instance , see refs.@xcite .
some phenomenological @xmath1 and @xmath2 potentials are found in ref.@xcite .
the comparison of our results with these potentials are , however , not easy , since our wigner transform is momentum - dependent and the solutions of the transcendental equations are strongly energy - dependent
. we will only point out if some resemblance between them are found .
the importance of the isospin dependence of the @xmath10 and @xmath11 interactions to the @xmath16-nucleus and @xmath17-nucleus potentials has been discussed by many authors from the phenomenological aspects , for example , in refs.@xcite , and by rijken and yamamoto from the viewpoint of @xmath3-matrix calculations in ref.@xcite .
the organization of this paper is as follows . in the next formulation section ,
we first give in section 2.1 the basic folding formula for the @xmath5-cluster based on the separation of the @xmath22 born kernel to the spin - isospin factors and the spatial part .
the treatment of @xmath3-matrix variables , the starting energy and the c.m .
momentum , will be carefully discussed .
a convenient transformation formula for the rearrangement of relative momenta is given in section 2.2 , by which the partial wave component of the @xmath22 born kernel is explicitly given both for the central component and the @xmath30 component .
the folding formula in the partial - wave expansion and the partial - wave components of the @xmath22 born kernel are explicitly given in section 2.3 .
a procedure to calculate the wigner transform is given in section 2.4 with a couple of approximations .
one of the approximations for the @xmath30 component yields a simple factor for the strength of the @xmath30 potential , which corresponds to the well - known scheerbaum factor @xmath36 @xcite in nuclear matter .
the section 3 is devoted to the results and discussion ; first in section 3.1 the @xmath0 central and @xmath30 potentials both in the @xmath37-matrix approach and in the wigner transform approach .
the isospin dependence of the @xmath1 and @xmath2 potentials is discussed in section 3.2 .
section 4 is devoted to a summary .
the invariant @xmath3-matrix for the most general @xmath38 interaction is discussed in appendix a.
the @xmath22 born kernel for the @xmath39 h.o . @xmath5-cluster wave function is calculated from @xmath40 where @xmath41 is the spin - isospin wave function of @xmath42 , @xmath43 the internal wave function of @xmath5 , and @xmath44 the @xmath45 @xmath3-matrix acting on the particle @xmath46 ( @xmath42 ) and @xmath47 - 5 ( nucleons ) .
we use the short - hand notation @xmath42 to specify one of the octet baryons , @xmath48 or @xmath17 . in , it is important to calculate the born kernel in the total c.m .
system by inserting @xmath49 and 1 in the bra and ket sides , respectively @xcite , since the @xmath3-matrix interaction @xmath44 is non - galilean invariant .
namely , the two - particle @xmath3-matrix which satisfies the translational invariance is parametrized by @xmath50 here , @xmath51 is the starting energy , @xmath52 is the magnitude of the c.m .
momentum , @xmath53 is the fermi momentum of nuclear matter , and the relative momentum @xmath54 ( and also @xmath55 etc . with primes ) , is given by @xmath56 with @xmath57 being the mass ratio between the nucleon and the baryon @xmath42 .
used in appendix b of ref.@xcite is the inverse of @xmath58 . ]
we assume a constant @xmath29 in this paper unless otherwise specified , so that we will omit this index in the following .
in fact , the @xmath3-matrix in contains the exchange term .
it is , therefore , convenient to use the isospin sum of the invariant @xmath3-matrix , @xmath59 with @xmath60 and @xmath61 ( @xmath62 is the isospin of @xmath42 ) , defined through @xmath63_{ii_z}\,|\,g(\bp , \bp^\prime ; k , \omega ) - g(\bp , -\bp^\prime ; k , \omega)\,p_\sigma\,p_f\,| \,[bn]_{ii_z}\,\rangle\ \ \nonumber \\ & & = \left(1+\delta_{b , n}\right)\left [ g^i_0+g^i_{ss } ( \bfsigma_1 \cdot \bfsigma_2 ) + h^i_0\,i \widehat{\bn } \cdot ( \bfsigma_1 + \bfsigma_2 ) + h^i_-\,i \widehat{\bn } \cdot ( \bfsigma_1 - \bfsigma_2 ) \right .
\nonumber \\ & & \hspace{29 mm } \left .
+ \cdots \right ] \label{fm1 - 4}\end{aligned}\ ] ] here @xmath64/(p^\prime p \sin \theta)$ ] , and the invariant functions @xmath65 ( central ) , @xmath66 ( spin - spin ) , @xmath67 ( @xmath30 ) , @xmath68 ( @xmath32 ) , etc .
are functions of @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath51 , and @xmath53 .
these are expressed by the partial - wave components of the @xmath45 @xmath3-matrix as ( see appendix d of ref.@xcite ) @xmath73
\end{array } \right\ } g^{ij}_{s \ell , s \ell}~p_l(\cos\theta)\ \
, \nonumber \\ h^{i}_0 & = & - \frac{1}{4 } \sum_j \frac{(2j+1)}{j(j+1 ) } \left [ g^{ij}_{1 j , 1 j}~p^1_j(\cos\theta ) \right . \nonumber\\ & & \left .
+ j~g^{ij}_{1 j+1 , 1 j+1 } ~p^1_{j+1}(\cos\theta ) - ( j+1 ) g^{ij}_{1 j-1 , 1 j-1 } ~p^1_{j-1}(\cos\theta ) \right]\ \ , \nonumber \\ h^{i}_- & = & \frac{1}{4 } \sum_j \frac{(2j+1)}{\sqrt{j(j+1 ) } } \left [ g^{ij}_{1 j , 0 j } + g^{ij}_{0 j , 1 j } \right]~p^1_j(\cos\theta)\ \ , \label{fm1 - 5 } \end{aligned}\ ] ] where the argument @xmath54 , @xmath55 , @xmath72 , @xmath51 , @xmath53 and subscripts for the baryon channels are omitted for the typographical reason . in the @xmath30 term , @xmath74 with @xmath75 is the associated legendre function of the first rank .
the invariant @xmath3-matrix for the most general @xmath38 interaction is discussed in appendix a. in order to calculate the spin - isospin factors , it is convenient to introduce an isospin @xmath76-exchange operator of the @xmath77 system by @xmath78 with the isospin matrix elements @xmath79 and write the invariant @xmath3-matrix as @xmath80 we need to calculate @xmath81 with the spin factors @xmath82 , @xmath83 , @xmath84 , and @xmath85 . in , @xmath86 is the spin - isospin wave function of @xmath5 ; i.e. , @xmath87 .
then we find that non - zero matrix elements are @xmath88 on the other hand , the spatial part is calculated in appendix ( b6 ) of ref.@xcite .
it is convenient to express this formula as @xmath89 where @xmath90 are the momentum transfer and the local momentum of the @xmath22 system . by assuming @xmath65 , @xmath91 and @xmath92 for @xmath93
, we finally obtain @xmath94}\cdot \bfsigma_1 \\ \end{array } \right\}\ \ , \label{fm1 - 13}\end{aligned}\ ] ] where @xmath95 and @xmath96 } = [ \bp \times \bk]/|\,[\bp \times \bk]\,|$ ] .
it should be noted that the c.m .
momentum of the two interacting particles , @xmath97 , in eqs.([fm1 - 11 ] ) and ( [ fm1 - 13 ] ) implies that the local momentum @xmath98 in plays the role of the incident momentum of the first baryon @xmath42 from , and @xmath54 the local momentum of the two - particle system , which is now an integral variable .
this is a consequence of fixing the c.m .
motion of the @xmath22 system as in , and is a special situation of the direct and knock - on ( exchange ) terms .
the @xmath3-matrix depends only on the magnitude @xmath52 , since we make an angular average in the @xmath3-matrix calculation @xcite .
the @xmath3-matrix value is , therefore , specified by @xmath72 and @xmath51 , or alternatively by the incident momentum @xmath99 and the relative momentum @xmath100 between @xmath42 and @xmath101 ; i.e. , @xmath102 with @xmath103 and @xmath104 .
if @xmath105 and @xmath106 are specified , @xmath72 is determined by the angular averaging , and @xmath107 is determined from @xmath105 and @xmath72 .
then the starting energy @xmath51 is determined as the sum of the relative kinetic energy of two particles and the s.p .
potentials for @xmath108 and @xmath101 at @xmath105 and @xmath107 , respectively .
we therefore choose @xmath109 and @xmath110 in the @xmath3-matrix in eqs.([fm1 - 11 ] ) and ( [ fm1 - 13 ] ) . in order to carry out this rather involved calculation
, we will develop in the next subsection some kind of transformation formula for the rearrangement of relative momenta in the partial - wave components of nonlocal kernels . the folding formula in implies that expressing the @xmath3-matrix interaction @xmath111 and @xmath112 } \cdot \bs$ ] with the subsidiary momentum variables @xmath113 is convenient for the @xmath5-cluster folding .
we express these kernels using the calligraphic letters ; i.e. , @xmath114 } \cdot \bs = g^{ls}(\bp , \bp^\prime ) \,i \widehat{[\bp^\prime \times \bp]}\cdot \bs\ \ .
\label{fm2 - 2}\end{aligned}\ ] ] by using this notation and the @xmath105 , @xmath106 notation for @xmath72 , @xmath51 , discussed in the last subsection , @xmath93 in , for example , becomes @xmath115 with @xmath109 . in the following ,
we omit the argument @xmath72 , @xmath51 or @xmath105 , @xmath106 for simplicity , unless the dependence becomes crucial for the relationship under consideration . after making partial wave decomposition in
, we can easily carry out the integral over @xmath54 .
the resultant expression is denoted by @xmath116 and by @xmath117}\cdot \bs$ ] .
the desired born kernels , @xmath118 and @xmath119}\cdot \bs$ ] , are obtained from another transformation related to .
our task is , therefore , to relate the partial - wave components @xmath120 and @xmath121 , and also @xmath122 and @xmath123 , which are defined through @xmath124 we generalize the transformation as @xmath125 with @xmath126 .
a simple calculation gives @xmath127 where @xmath128 with @xmath129 the transformation of the @xmath30 components is similarly carried out by using the orthogonality relations of @xmath130 and @xmath131 for the numerical integration over @xmath132 in etc . , it is convenient to use the spline interpolation @xmath133 since the g - matrix calculation itself is very much time consuming . after all , we obtain the following formula for the transformation of the partial - wave components : @xmath134 \,g_{\ell^\prime}(k , q;x)\ \ , \label{fm2 - 8}\end{aligned}\ ] ] where @xmath135 the summation over @xmath136 in is from @xmath137 for @xmath138 and from @xmath139 for @xmath140 .
it is important to note some symmetries possessed by @xmath141 . from the time - reversal symmetry ,
the @xmath3-matrix @xmath142 is symmetric for the interchange of @xmath54 and @xmath55 .
since this interchange corresponds to @xmath143 , the transformed partial - wave components , @xmath120 and @xmath122 , are no - zero only for @xmath144 and @xmath145 , respectively . for the coefficients in
, we can show @xmath146 in .
this property implies that , if @xmath147 is transformed to @xmath148 , then @xmath149 is transformed to @xmath150 . in the later application of the present formalism to @xmath151 resonating - group method ( rgm )
, we will find that the knock - on term is obtained by simply replacing @xmath148 to @xmath150 in the direct term .
in fact , the knock - on term is already included even in the hyperon @xmath5 system as in , which is the strangeness exchange term of the hyperon - nucleon interaction . in other words ,
the direct and knock - on terms are treated on the equal footing in the present formalism to deal with the invariant @xmath3-matrix interaction . for the transformation from @xmath152 to @xmath153 , we find that the latter born kernel is symmetric with respect to the interchange of @xmath154 and @xmath155 , which is the consequence of eqs.([fm2 - 5 ] ) and ( [ fm2 - 8 ] ) for the transformation coefficients ( see ) @xmath156 if we use partial - wave decomposition in and the similar expansions for @xmath157 , the folding formula becomes very simple .
for both of the @xmath138 and @xmath30 terms , it is given by @xmath158 where @xmath159 is the spherical bessel function of the imaginary argument .
for the proof of the @xmath30-term folding , we use a simple formula @xmath160}_\nu = ( -1)^\ell \frac{4\pi}{\sqrt{3 } } \sqrt{\frac{\ell ( \ell+1 ) } { ~2\ell+1~}}~\left[y_\ell(\widehat{\bk } ) y_\ell(\widehat{\bq } ) \right]_{1\nu}\ \ , \label{fm3 - 2}\ ] ] which is derived by using @xmath161 and @xmath162_\nu = ( -1)^\ell \frac{4\pi}{\sqrt{3}}\frac{kq}{(2\ell+1 ) } \left\{\sqrt{\frac{(\ell+1)(\ell+2)}{2\ell+3 } } \right.\nonumber \\ & & \left .
\times \left[y_{\ell+1}(\widehat{\bq } ) y_{\ell+1}(\widehat{\bk})\right]_{1\nu } -\sqrt{\frac{(\ell-1)\ell}{2\ell-1 } } \left[y_{\ell-1}(\widehat{\bq } ) y_{\ell-1}(\widehat{\bk})\right]_{1\nu } \right\}\ \ .
\label{fm3 - 4}\end{aligned}\ ] ] the final step to derive the partial - wave components @xmath163 in @xmath164}\cdot \bs\ \ , \nonumber \\
v^c(\bq_f , \bq_i ) & = & \sum^\infty_{\ell=0 } ( 2\ell+1)\,v^c_\ell(q_f , q_i ) \,p_\ell ( \widehat{\bq}_f\cdot \widehat{\bq}_i)\ \ ,
v^{ls}(\bq_f , \bq_i ) & = & \sum^\infty_{\ell=1 } ( 2\ell+1)\,v^{ls}_\ell(q_f , q_i ) \,p^1_\ell ( \widehat{\bq}_f\cdot \widehat{\bq}_i)\ \ , \label{fm3 - 5}\end{aligned}\ ] ] is carried out by using the transformation formula in the preceding subsection with the coefficients in . in the @xmath165-coupling scheme ,
we also need @xmath166 in the partial - wave expansion @xmath167 which is given by @xmath168 = \left\{\begin{array}{c } \frac{\ell}{2 } \\
-\frac{\ell+1}{2 } \\
\end{array } \right . \quad \hbox{for } \quad j=\left\{\begin{array}{c } \ell+\frac{1}{2 } \\ \ell-\frac{1}{2 } \\
\end{array } \right .
.\nonumber \\ \label{fm3 - 7}\end{aligned}\ ] ] in , @xmath169_{jm}$ ] is the angular - spin wave function of the @xmath22 system . for the proof
, we again use . in summary ,
the born kernel of the @xmath22 system is obtained by using eqs .
( [ fm2 - 8 ] ) and ( [ fm3 - 1 ] ) , starting from @xmath170 \right\ } \nonumber \\ & & \hspace{80 mm } ( \ell=1,~2,~3,\cdots)\ \ .
\label{fm3 - 8 } \end{aligned}\ ] ] for @xmath23 , we should multiply the factor 2 to include the knock - on term .
we first use with the coefficients and transform the above @xmath147 to @xmath148 .
then , the @xmath5-cluster folding by yields @xmath171 from @xmath148 . the second transformation from @xmath171 to @xmath172
is carried out with the coefficient .
the selection of @xmath105 and @xmath106 in @xmath173 is now almost apparent .
we choose @xmath105 as @xmath106 in the folding formula .
the relative momentum @xmath106 is actually @xmath174 in .
therefore , the nest structure @xmath175 and @xmath176 is incorporated in the computer code .
namely , we first specify @xmath177 and @xmath106 in .
then @xmath178 is generated from @xmath179 for a general @xmath106 , which is obtained from the transformation formula , , by using the complete off - shell @xmath3-matrix @xmath180 .
the present formalism is convenient to calculate the wigner transform @xmath181 , since they are essentially the fourier transform of @xmath182 .
we define these through @xmath183}\cdot \bs\right\ } \nonumber \\ & = & v^c_w(\br , \bq)+v^{ls}_w(\br , \bq)\,[\br \times \bq]\cdot \bs\ \ .
\label{fm4 - 1}\end{aligned}\ ] ] note that the @xmath30 term is defined by @xmath184 \cdot \bs$ ] , instead of @xmath185}\cdot \bs$ ] .
if we apply the partial - wave expansion @xmath186 with @xmath187 , we can easily derive the partial - wave components as @xmath188 here we use again for the derivation of the @xmath30 term .
since @xmath189 for @xmath190 and @xmath191 for @xmath192 ( see subsec .
2.2 ) , the @xmath193 and @xmath194 terms become the leading terms for the central and @xmath30 wigner transform , respectively .
in fact , in the @xmath195 case , we find that only these leading terms survive for the zero - momentum wigner transform .
it is , therefore , a good approximation to retain only @xmath193 and @xmath194 terms in : @xmath196 which we use throughout this paper .
the zero - momentum wigner transform is a convenient tool to estimate the strength of the interaction . from the folding formula
, we can calculate @xmath197 and @xmath198 .
this process yields @xmath199 for the @xmath30 component , it is easy to introduce another approximation , which gives a simple factor for the @xmath30 strength , similar to the scheerbaum s factor @xcite . for this purpose
, we express @xmath200 in the original form @xmath201\,\frac{g^{ls}_\ell(p , p^\prime)}{p p^\prime } \nonumber \\ & & \times \left(\frac{1}{\sin \theta}\right ) p^1_\ell(\cos \theta ) \
\ , \label{fm4 - 6}\end{aligned}\ ] ] where @xmath202 with @xmath203 , @xmath204 and @xmath205 .
we neglect the @xmath177 dependence except for the front factor @xmath206 and set @xmath207 . then , @xmath208 , @xmath209 , and the @xmath132 integral can be performed . by further using @xmath210
, we obtain @xmath211 if we set @xmath212 \right\ } \nonumber \\ \ \hspace{-10 mm } & = & 2 \sum_{ij } \left(\frac{2i+1}{2i_b+1}\right ) \left [ h^i_0(\bp , \bp^\prime)+h^i_-(\bp , \bp^\prime)\right ] \frac{1}{\sin \theta } \bigr|_{\bp=\bp^\prime=\bq}\ \ , \label{fm4 - 8}\end{aligned}\ ] ] we can write @xmath213 . using this and the integration formula @xmath214 we find @xmath215 we can write in the way similar to the scheerbaum s formula : @xmath216 for the @xmath22 system , the density @xmath217 of the @xmath5-cluster is calculated as @xmath218 where @xmath219 is the coordinate of the nucleon @xmath220 , measured from the c.m . of the @xmath5-cluster .
then , we find that the integral is nothing but @xmath221 . thus becomes @xmath222 similarly , we can write as @xmath223 and obtain @xmath224 this expression corresponds to eq.(50 ) of ref.@xcite , which is the scheerbaum factor of finite nuclei derived from @xmath3-matrix calculations : @xmath225 the difference between eqs.([fm4 - 14 ] ) and ( [ fm4 - 15 ] ) is the weight function , @xmath226 and an extra front factor @xmath227 in .
this enhancement factor appears , since in the standard @xmath3-matrix calculation of single - particle potentials the c.m .
motion of the total system is not correctly treated . in the @xmath22 system
, this approximation affects the result appreciably , which is discussed in ref.@xcite .
the two weight functions in may seem to be fairly different , since @xmath228 with @xmath229 is given by @xmath230 .
however , this is not the case , since in the small @xmath106 region ( the low - energy region ) @xmath231 is anyway very small .
we can further approximate eqs.([fm4 - 14 ] ) and ( [ fm4 - 15 ] ) , by calculating @xmath232 with each weight function , and by replacing @xmath106 in @xmath231 with this @xmath233 .
the average momentum @xmath233 is given by @xmath234 for @xmath235 with @xmath236 , @xmath237 , and for @xmath238 with @xmath239 , @xmath240 .
these values are very close to @xmath241 , calculated from the scheerbaum s estimation for an average momentum transfer @xmath242 . from this prescription ,
@xmath243 and @xmath244 are given by @xmath245 , but for the @xmath30 component @xmath246 . ] , but for the @xmath30 component @xmath246 . ]
in order to gain the overview of the zero - momentum wigner transform given in , we will show in fig.[fig1a ] the central ( @xmath247 ) and @xmath30 ( @xmath248 ) components for the @xmath22 interaction with @xmath20 and @xmath17 , when the quark - model @xmath3-matrix @xmath38 interaction by fss2 are employed with the continuous choice for intermediate spectra . in this case
, we choose @xmath229 and the relative momentum @xmath106 is assigned to the integral mesh point @xmath106 in .
the fermi - momentum used for the @xmath3-matrix calculation is @xmath29 , and the @xmath249 shell - model wave function with the h.o .
size parameter @xmath250 is used for the @xmath5-cluster folding .
we note that the correct treatment of the total c.m .
motion for the @xmath22 system is very important , since the c.m .
correction in the standard approach is of the order of @xmath251 . as the result
, the zero - momentum wigner transform becomes very much short - ranged and deep with the interaction range about @xmath252 .
the @xmath5-particle density in is more compact than the density including the c.m .
motion , @xmath253 , and the central density is about @xmath254 times larger .
the extremely large @xmath255 wigner transform in fig.[fig1a ] is because of the factor 2 in , and also because the other nucleon - exchange term than the knock - on term and the effect of the exchange normalization kernel are neglected .
we will deal with this case in a separate paper .
[ table1 ] .values of the zero - momentum wigner transform at the origin @xmath256 , @xmath257 and @xmath258 , and the scheerbaum - like factor , @xmath259 , in .
the unit is in mev for @xmath257 and @xmath258 , and in mev@xmath260 for @xmath259 .
the models are fss2 and fss , and @xmath261 and cont .
imply the @xmath262 and continuous choices for intermediate spectra , respectively , used in the @xmath3-matrix calculations .
[ cols="^,^ , > , > , > , > " , ] we list in table [ table5 ] the depths of the effective local potentials @xmath263 for positive energies , @xmath264 and 50 mev , and the @xmath265-wave phase shifts @xmath266 obtained by solving the schrdinger equation .
the exact phase shift @xmath267 ( exact ) is also shown , which is obtained from the born kernel . here
we again find the effective local potentials are too shallow , especially for the effective @xmath268 forces .
the phase shift difference is about 5 - @xmath269 for the effective @xmath268 forces , while 3 - @xmath270 for the quark - model @xmath3-matrix interactions .
this implies that we need a readjustment of the effective local potentials of the order of 10 mev , in order to reproduce the correct magnitude of the phase shifts .
finally , we will make a brief comment on the choice of the @xmath53 value in the present framework . if we choose a smaller @xmath53 , the @xmath8 s.p .
potential in the symmetric nuclear matter becomes shallower and the @xmath0 central interaction becomes more attractive .
the first feature reduces the magnitude of the starting energy in the @xmath268 @xmath3-matrix , resulting in more attractive @xmath268 @xmath3-matrix .
although this change is compensated by the smaller volume in the phase - space integral to calculate the @xmath8 s.p .
potential , such a mechanism does not work in the present calculation of the @xmath0 interaction .
thus the @xmath0 central interaction after the @xmath5-cluster folding becomes more attractive for a smaller @xmath53 .
the same situation is observed in fig.4 of ref.@xcite for their @xmath271 potential .
for example , if we change @xmath29 to @xmath272 ( which corresponds to the 70@xmath273 of the normal density ) , the depth of the @xmath8 s.p .
potential , @xmath274 , in table [ table2 ] is reduced from @xmath275 mev to @xmath276 mev for the model fss2 in the continuous choice . on the other hand ,
the depth of the effective local potential , @xmath263 , and @xmath277 in table [ table3 ] changes from @xmath278 mev to @xmath279 mev and from @xmath280 mev to @xmath281 mev , respectively .
component , the @xmath282 values in table [ table3 ] change from @xmath283 mev to @xmath284 mev for fss2 and from @xmath285 mev to @xmath286 mev for fss , when @xmath287 is used in the continuous choice .
] this implies that the self - consistent mechanism of the starting - energy dependence , which is not properly taken into account in this paper , is in fact very important .
this finding is in accord with the importance of the brueckner rearrangement effect discussed in ref.@xcite .
since the purpose of the present study is not to examine the change of the @xmath5-cluster , it would be safe to assume the standard value @xmath29 , in order to examine the qualitative features of the @xmath22 interaction . , but for fss2 . ] , but for fss2 . ] as seen from fig.[fig1a ] , the @xmath1 and @xmath2 interactions are repulsive .
this does not mean that the @xmath16 and @xmath17 potentials are always repulsive .
the structure of spin - isospin factors for the @xmath1 and @xmath2 systems is especially simple ( see ) , which is due to the spin - isospin saturated character of the @xmath5-particle .
it is , therefore , important to examine each isospin component separately , in order to gain some insight to other possibilities of unknown hypernuclei .
here we discuss some qualitative features of these interactions , based on the symmetry properties of the @xmath38 interactions predicted by the quark - model interactions , fss and fss2 .
when the interaction is repulsive , the transcendental equation sometimes does not have its solution , since the square of the local momentum , @xmath288 , becomes negative . since the extension of the wigner transform to the negative @xmath288
is not easy numerically , we only discuss the zero - momentum wigner transform in this subsection . , but for fss2 . ] , but for fss2 . ]
figures [ fig9a ] and [ fig9b ] show the isospin components with @xmath12 and @xmath14 for the @xmath1 central interaction , predicted by fss and fss2 , respectively .
we find that both interactions have some amount of attraction originating from the @xmath289 channel of the @xmath12 @xmath10 interaction .
this channel becomes attractive due to the very strong @xmath268@xmath10 coupling by the one - pion exchange tensor force . on the other hand ,
the @xmath289 state of the @xmath14 channel is strongly repulsive due to the pauli principle at the quark level .
we find from figs .
[ fig9a ] and [ fig9b ] that this repulsion is so strong that almost no attraction from the @xmath12 channel remains in the @xmath1 interaction . ,
but for fss2 .
] , but for fss2 . ] , but for fss2 . ] , but for fss2 . ] on the other hand , the two isospin channels , @xmath12 and 3/2 , yield fairly large @xmath30 force for the @xmath1 interaction .
this can be seen in figs.[fig11a ] and [ fig11b ] . in the isospin @xmath14 channel
, @xmath290 states are classified to the flavor symmetric channel with the @xmath291 label ( 22 ) .
it is , therefore , plausible that the same mechanism as the @xmath24 @xmath15 channel yields very strong @xmath30 force .
on the other hand , a part of the @xmath30 force from the isospin @xmath12 channel is due to the flavor - exchange process between @xmath292 and @xmath293 @xmath291 configurations .
this process is accompanied with the spin flip between @xmath294 and 1 , and yields very strong @xmath32 force .
the @xmath30 and @xmath32 forces reinforce each other in just opposite way to the @xmath268 interaction , and the resultant @xmath1 @xmath30 force becomes almost 3/5 of the @xmath31 @xmath30 force , as seen from @xmath259 in table [ table1 ] .
this ratio is almost the same as that of the scheerbaum factors @xmath295 in symmetric nuclear matter .
( see table [ table2 ] . )
figures [ fig13a ] and [ fig13b ] show the isospin components with @xmath13 and @xmath15 for the @xmath2 central interaction , predicted by fss and fss2 , respectively . here again we find the situation that the attractive nature of the @xmath13 component is largely canceled by the repulsion in the @xmath15 channel . however , this cancellation is not strong especially in the model fss , and we have almost 5 mev attraction around @xmath296 fm . in fss2 , the height of the central repulsion in the @xmath15 channel is almost 20 mev , and we can expect a few mev attraction in the surface region .
these long - range attractions may have some influence to the atomic orbit between @xmath297 and @xmath5 .
the @xmath2 potential rather similar ( though more attractive ) to that by fss is presented by myint and akaishi in ref.@xcite .
it should be noted that the origin of the repulsion in the @xmath15 channel is the pauli forbidden state @xmath293 in the @xmath298 state and the almost pauli forbidden state ( 30 ) in the @xmath289 state . however , the coupling with the @xmath299 channel is very important , which may cause the long - range attraction even in the @xmath15 channel .
the @xmath30 components of the @xmath2 interaction is repulsive , which is clearly seen in figs.[fig15a ] ans [ fig15b ] .
this @xmath30 force is fairly strong , especially for fss .
the magnitude is almost 1/3 of the @xmath31 @xmath30 force with the opposite sign .
these features are very similar to the @xmath30 force in symmetric nuclear matter , as seen in table [ table2 ] .
the @xmath4 quark - model baryon - baryon interaction ( fss2 , fss ) @xcite is a unified model which describes all the baryon - octet baryon - octet ( @xmath38 ) interactions in a full coupled - channel formalism . for the nucleon - nucleon ( @xmath24 ) and hyperon - nucleon ( @xmath19 ) interactions ,
all the available experimental data are reasonably reproduced .
it is , therefore , interesting to study @xmath22 interactions in a microscopic framework under a simple assumption of the @xmath249 harmonic - oscillator shell - model wave function for the @xmath5-cluster . in this study , we have used the result of @xmath3-matrix calculations for symmetric nuclear matter , as an input for the two - body interactions for the @xmath5-cluster folding . since the resultant @xmath22 interactions are rather insensitive to the fermi - momentum @xmath53 in the @xmath3-matrix calculations ( except for the @xmath268 @xmath30 force for fss ) , we have assumed @xmath29 .
the other @xmath3-matrix parameters , the center - of - mass ( c.m . ) momentum @xmath72 of two interacting particles and the starting energy @xmath51 , are treated unambiguously in the total c.m .
frame of the @xmath22 system .
this can be achieved by considering the transformation of the matrix elements in the momentum representation from the initial ( @xmath300 ) and final ( @xmath301 ) momenta to the momentum transfer ( @xmath302 ) and the local momentum @xmath98 .
if one uses this transformation at the level of @xmath3-matrix , the procedure of the @xmath5-cluster folding becomes extremely simple both for the central and @xmath30 components .
the @xmath22 interaction , @xmath303 with @xmath304 , represented by @xmath305 and @xmath306 is then transformed back to the @xmath22 born kernel , @xmath307 , through the inverse transformation .
this procedure is also convenient to calculate the wigner transform , @xmath308 , which is simply a fourier transform of @xmath303 . by solving the transcendental equation for @xmath308
, we can obtain an energy - dependent local potential of the @xmath22 system in the wkb - rgm approximation @xcite . in this paper
, we have applied the present formalism to the @xmath0 , @xmath1 and @xmath2 systems .
applications to the @xmath31 system will be discussed in a separate paper , since this system involves an extra nucleon - exchange term , in addition to the direct and knock - on terms . in the @xmath0 system ,
the wkb - rgm approximation is rather poor due to the very strong momentum dependence of the exchange knock - on term .
this term appears even for the effective @xmath268 force , which is traced back to the strangeness exchange processes .
we find that the @xmath0 central potentials predicted by various quark - model @xmath3-matrices are very similar to each other , irrespective of a specific model , fss2 or fss , and the @xmath262 or continuous choice for intermediate spectra . at the bound - state energy , @xmath309 mev , they are long - range local potentials with the wine - bottle shape , having the depth less than 30 mev .
they are very similar to the @xmath0 potential obtained from the effective @xmath268 potential nd @xcite , simulating the nijmegen hard - core model d. the @xmath0 bound - state energies are calculated by solving the lippmann - schwinger equation of the @xmath0 born kernel .
these are @xmath280 mev for fss2 ( cont ) and @xmath310 mev for fss ( cont ) , when @xmath29 is used .
it should be noted that the depth of the single - particle potential for @xmath8 in symmetric nuclear matter is 48 mev for fss2 ( cont ) and 46 mev for fss ( cont ) @xcite .
the fact that the predicted @xmath0 bound - state energies are by no means too large , in comparison with the experimental value @xmath311 , implies that the proper treatment of the c.m .
motion of the @xmath0 system is very important .
it is also important to note that , in the present @xmath3-matrix approach , the @xmath268@xmath10 coupling by the very strong one - pion exchange tensor force is explicitly treated , the lack of which is known to lead to the so - called overbinding problem of the @xmath0 bound state .
the energy loss predicted by the @xmath5-cluster rearrangement effect @xcite through the starting - energy dependence of the @xmath3-matrix needs further detailed analyses . on the other hand , the @xmath0 @xmath30 potentials are rather model dependent . in the model fss2 ,
the depth of the @xmath30 potential is @xmath312 mev , while in fss about @xmath313 mev .
the interaction range of the fss @xmath30 potential is also very short , which leads to a small scheerbaum - like factor @xmath314 for the @xmath0 @xmath30 force .
we will show in a separate paper , the strength of the @xmath30 force is further reduced for smaller values of @xmath53 , if fss is used .
the very small spin - orbit splitting of the @xmath315 @xcite can be reproduced in the @xmath316 faddeev calculation , using the @xmath317 rgm kernel and the @xmath0 @xmath30 born kernel predicted from the fss @xmath3-matrix with @xmath318 .
based on the reasonable reproduction of the @xmath0 interaction properties , we have examined the real parts of the @xmath1 and @xmath2 interactions in the wigner transform technique .
since these interaction are repulsive , we have examined only the zero - momentum wigner transform as the first step . in the @xmath1 interaction , the attractive effect from the isospin @xmath12 @xmath10 channel is completely cancelled out by the repulsion from the @xmath14 @xmath10 channel .
the origin of this strong repulsion is the @xmath319 @xmath289 channel , which contains the almost pauli - forbidden @xmath291 ( 30 ) component for the most compact @xmath320 configuration . on the other hand ,
the @xmath2 interaction is less repulsive because of the appreciable attraction originating from the @xmath13 @xmath11 channels .
the @xmath2 zero - momentum wigner transform predicted by fss yields about @xmath321 mev attraction around @xmath296 fm , while the attraction of fss2 is about @xmath322 mev .
these long - range attractions may have some relevance to the formation of atomic bound states for the @xmath323 system .
as to the spin - orbit interaction , the two isospin channels of the @xmath10 interaction give fairly strong attractive @xmath1 @xmath30 forces , yielding almost 3/5 of the @xmath31 @xmath30 force . on the other hand ,
@xmath2 @xmath30 force is repulsive and the magnitude is @xmath324 of the @xmath31 @xmath30 force .
we will show in the next paper , the present @xmath31 @xmath30 force is consistent with the observed @xmath325-wave splitting of the @xmath326 and @xmath327 excited states of @xmath328 .
it should be noted that the overall repulsive character of the @xmath1 and @xmath2 interactions is related to the spin - isospin saturated character of the @xmath5-cluster .
the strong isospin dependence of the @xmath10 and @xmath11 interactions , namely , repulsive for the @xmath319 and @xmath329 channels and attractive for the @xmath330 and @xmath331 channels , leads to a possibility of attractive features in some particular spin - isospin channels for systems of @xmath16 , @xmath17 and the @xmath332-shell clusters @xcite .
one of the examples of such systems is the isospin @xmath12 and spin @xmath294 state of @xmath18 , in which the strong repulsion of the @xmath319 @xmath289 channel does not contribute @xcite and a quasi - bound state is in fact observed experimentally @xcite .
applications of the present approach to such systems are under way .
* acknowledgments * this work was supported by grants - in - aid for scientific research ( c ) from the japan society for the promotion of science ( jsps ) ( grant nos .
18540261 and 17540263 ) .
in this appendix , we will define the invariant @xmath3-matrix for the most general @xmath38 interaction .
the @xmath38 channels in the bra side ( @xmath333 ) and ket side ( @xmath5 ) are specified by @xcite @xmath334 ss_zyii_z ; \cp\ \ , \nonumber \\ \alpha & = & \left[\frac{1}{2}(11)a_1\,\frac{1}{2}(11)a_2\right ] s^\prime { s_z}^\prime yii_z ; { \cp}^\prime\ \ , \label{a1}\end{aligned}\ ] ] in the isospin basis . for example , the spin - flavor functions are given by @xmath335^\cp_{ii_z}\ \
, \nonumber \\ & & [ b_1 b_2]^\cp_{ii_z}=\frac{1}{\sqrt{2(1+\delta_{c_1,c_2 } ) } } \left\{[b_1 b_2]_{ii_z}+\cp ( -1)^{i_1+i_2-i}[b_2 b_1]_{ii_z}\right\ } \ , \label{a2}\end{aligned}\ ] ] where the isospin - coupled flavor wave functions , @xmath336_{ii_z}$ ] , are lexicographically ordered and the first baryon is numbered always 1 and the second 2 ( i.e. , @xmath337_{ii_z}=[b_2(1 ) b_1(2)]_{ii_z}$ ] ) .
the flavor symmetry basis , @xmath336^\cp_{ii_z}$ ] with @xmath338 and @xmath339 , gives @xmath340_{ii_z } & = & \sqrt{\frac{1+\delta_{c_1 , c_2}}{2 } } \sum_\cp \left[b_1 b_2\right]^\cp_{ii_z}\ \
, \nonumber \\ \left[b_3 b_4\right]_{ii_z } & = & \sqrt{\frac{1+\delta_{a_1 , a_2}}{2 } } \sum_{\cp^\prime } \left[b_3 b_4\right]^{\cp^\prime}_{ii_z}\ \ . \label{a3}\end{aligned}\ ] ] the matrix element of the @xmath3-matrix in the isospin basis and its partial - wave decomposition are given by @xcite @xmath341^\cp_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) | \chi_{s^\prime{s_z}^\prime } [ b_3 b_4]^{\cp^\prime}_{ii_z}\rangle \nonumber \\ & = & \sum^\prime_{jm\ell \ell^\prime } 4\pi g^j_{\gamma s\ell,\alpha s^\prime{s_z}^\prime}(p , p^\prime ; k , \omega ) \sum_m \langle \ell m s s_z | jm \rangle \,y_{\ell m}(\widehat{\bp } ) \nonumber \\ & & \times \sum_{m^\prime } \langle \ell^\prime m^\prime s^\prime { s_z}^\prime \label{a4}\end{aligned}\ ] ] here the prime symbol on @xmath342 implies that the summation is only for such quantum numbers that satisfy the generalized pauli principle : @xmath343 we multiply with @xmath344 and @xmath345 from the left- and right - hand sides , respectively , and take a sum over @xmath346 and @xmath347
. then we find @xmath348^\cp_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) | [ b_3 b_4]^{\cp^\prime}_{ii_z}\rangle \nonumber \\ & & = \sum^\prime_{jm\ell \ell^\prime s s^\prime } 4\pi\,g^j_{\gamma s\ell,\alpha s^\prime{s_z}^\prime}(p , p^\prime ; k , \omega)\,\cy_{(\ell s)jm}(\widehat{\bp } ; { \rm spin } ) \,\cy^*_{(\ell^\prime s^\prime)jm}(\widehat{\bp}^\prime ; { \rm spin } ) \ \ , \nonumber \\ \label{a6}\end{aligned}\ ] ] where @xmath349_{jm}$ ] is the angular - spin function for the two - baryon system .
let us denote the two - baryon channels with @xmath350 , @xmath351 , and consider @xmath352_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) -g(\bp , -\bp^\prime ; k , \omega ) p_\sigma p_f | [ b_3 b_4]_{ii_z } \rangle \ \ . \label{a7}\end{aligned}\ ] ] we can rewrite this by using eqs.([a3 ] ) and ( [ a6 ] ) : @xmath353^\cp_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) | [ b_3 b_4]^{\cp^\prime}_{ii_z}\rangle \right .
\nonumber \\ & & \left . -\langle
[ b_1 b_2]^\cp_{ii_z } | g(\bp , -\bp^\prime ; k , \omega)p_\sigma
p_f | [ b_3 b_4]^{\cp^\prime}_{ii_z}\rangle \right\}\ \ . \label{a8}\end{aligned}\ ] ] here the second term in the brackets gives the same contribution as the first term , due to the condition .
thus we find that is actually invariant @xmath3-matrix , which allows the expressions @xmath354_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) -g(\bp , -\bp^\prime ; k , \omega ) p_\sigma p_f | [ b_3 b_4]_{ii_z } \rangle \nonumber \\ & & = \sqrt{(1+\delta_{c_1,c_2})(1+\delta_{a_1,a_2 } ) } \sum_{\cp , \cp^\prime } \langle [ b_1 b_2]^\cp_{ii_z } | g(\bp , \bp^\prime ; k , \omega ) | [ b_3 b_4]^{\cp^\prime}_{ii_z}\rangle \nonumber \\ & & = \sqrt{(1+\delta_{c_1,c_2})(1+\delta_{a_1,a_2 } ) } \sum^\prime_{jm\ell \ell^\prime s s^\prime \cp \cp^\prime } 4\pi\,g^j_{\gamma s\ell,\alpha s^\prime{s_z}^\prime}(p , p^\prime ; k , \omega ) \nonumber \\ & & \qquad \times \cy_{(\ell s)jm}(\widehat{\bp } ; { \rm spin } ) \,\cy^*_{(\ell^\prime s^\prime)jm}(\widehat{\bp}^\prime ; { \rm spin } ) \nonumber \\ & & = \sqrt{(1+\delta_{c_1,c_2})(1+\delta_{a_1,a_2 } ) } \left [ g_0 + g_{ss}\,(\bfsigma_1\cdot \bfsigma_2 ) + h_0\,i\widehat{\bn}\cdot ( \bfsigma_1+\bfsigma_2 ) \right .
\nonumber \\ & & \left .
\qquad + h_-\,i\widehat{\bn}\cdot ( \bfsigma_1-\bfsigma_2 ) + \cdots \right]\ \ . \label{a9}\end{aligned}\ ] ] the eight independent invariant functions , @xmath355 , @xmath356 , etc . , are expressed by some combinations of the partial - wave components of the @xmath3-matrix , @xmath357 , which are explicitly given in appendix d of ref.@xcite .
99 y. fujiwara , y. suzuki and c. nakamoto , prog .
( 2006 ) ( in press ) y. fujiwara , m. kohno , c. nakamoto and y. suzuki , y. fujiwara , t. fujita , m. kohno , c. nakamoto and y. suzuki , y. fujiwara , k. miyagawa , m. kohno , y. suzuki and h. nemura , y. fujiwara , k. miyagawa , y. suzuki , m. kohno and h. nemura , y. fujiwara , k. miyagawa , m. kohno and y. suzuki , y. fujiwara , k. miyagawa , m. kohno , y. suzuki , d. baye and j .- m .
sparenberg , y. fujiwara , m. kohno , k. miyagawa and y. suzuki , y. fujiwara , m. kohno , k. miyagawa , y. suzuki and j .- m .
sparenberg , y. fujiwara , h. nemura , y. suzuki , k. miyagawa and m. kohno , y. fujiwara , y. suzuki , k. miyagawa , m. kohno and h. nemura , m. kohno , y. fujiwara , t. fujita , c. nakamoto and y. suzuki , y. fujiwara , m. kohno , t. fujita , c. nakamoto and y. suzuki , m. kohno , y. fujiwara and y. akaishi , x. campi and d. w. sprung , m. kohno , s. nagata and n. yamaguchi , m. kohno and d. w. l. sprung , e. hiyama , m. kamimura , t. motoba , t. yamada and y. yamamoto , c. daskaloyannis , m. grypeos and h. nassena , s. nakaichi - maeda and e. w. schmid , r. guardiola and j. navarro , i. n. filikhin , a. gal and v. m. suslov , t. harada , s. shinmura , y. akaishi and h. tanaka , s. okabe , t. harada and y. akaishi , t. yamada and k. ikeda , k. s. myint and y. akaishi , c. b. dover and a. gal , th .
a. rijken and y. yamamoto , nucl - th/0608074 r. r. scheerbaum , c. nakamoto , y. suzuki and y. fujiwara , y. fujiwara , m. kohno , c. nakamoto and y. suzuki , h. horiuchi , ; k. aoki and h. horiuchi , y. suzuki and k. t. hecht , ; h. akikawa , h. tamura , t. yamada and k. ikeda , r. s. hayano , t. nagae , | we calculate @xmath0 , @xmath1 and @xmath2 potentials from the nuclear - matter @xmath3-matrices of the @xmath4 quark - model baryon - baryon interaction .
the @xmath5-cluster wave function is assumed to be a simple harmonic - oscillator shell - model wave function .
a new method is proposed to derive the direct and knock - on terms of the interaction born kernel from the hyperon - nucleon @xmath3-matrices , with explicit treatments of the nonlocality and the center - of - mass motion between the hyperon and @xmath5 .
we find that the @xmath4 quark - model baryon - baryon interactions , fss and fss2 , yield a reasonable bound - state energy for @xmath6 , @xmath7 mev , in spite of the fact that they give relatively large depths for the @xmath8 single - particle potentials , 46 @xmath9 48 mev , in symmetric nuclear matter .
an equivalent local potential derived from the wigner transform of the nonlocal @xmath0 kernel shows a strong energy dependence for the incident @xmath8-particle , indicating the importance of the strangeness - exchange process in the original hyperon - nucleon interaction . for the @xmath1 and @xmath2 potentials ,
we only discuss the zero - momentum wigner transform of the interaction kernels , since these interactions turn out to be repulsive when the two isospin contributions for the @xmath10 and @xmath11 interactions are added up .
these components show a strong isospin dependence : they are attractive in the isospin @xmath12 ( @xmath1 ) and @xmath13 ( @xmath2 ) components and repulsive in @xmath14 ( @xmath1 ) and @xmath15 ( @xmath2 ) components , which indicate that @xmath16 and @xmath17 potentials could be attractive in some particular systems such as the well - known @xmath18 system .
@xmath0 potential , @xmath19 interaction , quark model baryon - baryon interaction , @xmath3-matrix , single - particle spin - orbit potential of hyperons |
three dimensional low - thrust optimal orbit transfer has attracted much inquiry focused on trajectory optimization using orbital elements . due to strong nonlinearity of differential equations in gaussian form with orbital elements , it is often difficult to obtain the optimal solution numerically in this system .
some of the earliest works on the orbit transfer between neighboring elliptic orbits and on the transfer between coplanar and coaxial ellipses were presented by edelbaum @xcite .
however , his elements as well as the keplerian elements all contain singularity .
to avoid the singularity , arsenault @xcite firstly introduced the equinoctial elements .
broucke and cefola @xcite developed nonsingular equinoctial orbital elements using the lagrange and poisson brackets of keplerian elements .
kechichian @xcite presented an application of these nonsingular elements to solve the minimum time rendezvous problem with constant acceleration .
chobotov @xcite considered more cases in minimum time transfer , including the comparison between exact solutions and approximate solutions obtained by the averaging technique .
gerffroy and epenoy @xcite made further progress in both minimum time and fuel transfer problems using the averaging technique with the constraints of earth s oblateness and shadow effect taken into account .
more recent works using the numerical averaging technique were presented by tarzi and speyer et al.@xcite , who provided a quick and accurate numerical approach for a wide range of transfers , including orbital perturbations such as earth s oblateness and shadow effect .
besides the strong nonlinearity , an additional disadvantage in using the equinoctial elements is the complexity of the equations in this coordinate system . with this system ,
kepler s equation must be solved by iteration to get the eccentric longitude at each integration step .
hence , equinoctial elements present challenges for trajectory optimization .
walker @xcite put forth another important development in the study of equinoctial elements when he used the stroboscopic method to modify orbital elements .
he also altered differential equations into an approximative form , containing five dependent variables and one independent variable , as a means of achieving faster computation performance without solving kepler s equation .
roth @xcite introduced the stroboscopic method to obtain a higher order approximation for small perturbed dynamical systems , which depends on several slow variables and one fast variable .
haberkorn and gergaud @xcite investigated the application of the modified orbital elements by using the homotopy method .
cui et al .
@xcite used sequential quadratic programming under their lyapunov feedback control law , which was based on a function made up of modified elements , to obtain the optimal - lyapunov solution without optimal transfer .
gao and li @xcite made the similar work to optimize the lyapunov function but never reached the optimal solution based on their lyapunov control law .
an advanced technique using the lyapunov - based controller to solve the low - thrust orbit transfer problem in cartesian coordinates was introduced and rigorously proved by chang et al .
. this technique is based on the fact that a non - degenerate keplerian orbit is uniquely described by its specific angular momentum and laplace vectors .
the resulting lyapunov function provides an asymptotically stabilizing feedback controller , such that the target elliptic keplerian orbit becomes a locally asymptotically stable periodic orbit .
however , the lyapunov - based transfer trajectory is not optimal in every sense . in this paper , the lyapunov - based transfer presented in @xcite
shall be called chang - chichka - marsden ( ccm ) transfer to distinguish it from any other lyapunov - based transfers .
the motivation behind this paper is to use the ccm transfer trajectory as an initial guess for optimization in order to obtain the optimal transfer solution utilizing cartesian coordinates . using this method avoids the numerical disadvantages due to strong nonlinearity and complexity in the use of orbital elements .
this paper reviews the ccm transfer method @xcite in section 2 . in section 3 , a means to translate keplerian elements into specific angular momentum and laplace vectors
is presented .
section 4 presents the proposed approach and the optimality ( kkt ) conditions for the minimum fuel consumption problem in cartesian coordinates .
specifically , the chebyshev - gauss pseudospectral method is introduced to illustrate how the continuous optimal control problem can be reduced to a discretized nonlinear programming problem .
finally , in section 5 , numerical simulations are carried out to make detailed comparisons between the optimal results using cartesian coordinates and those using orbital elements with the same initial guess .
it shows that the use of cartesian coordinates makes it easier to obtain the correct optimal solution .
this section summarizes and reviews the chang - chichka - marsden ( ccm ) transfer in @xcite .
this transfer employs lyapunov - based controllers to achieve asymptotically stable transfers between elliptic orbits in a two - body problem .
this paper assumes that the configuration space is @xmath0 .
let @xmath1 be the tangent space of @xmath2 , and @xmath3 be the coordinates on @xmath4 .
the equations of motion are given by @xmath5 of which the solutions are regarded as the _ keplerian orbits _ , where @xmath6 is the gravitational constant .
the specific energy @xmath7 is defined by @xmath8 define @xmath9 by @xmath10 where @xmath11 is the specific angular momentum vector and @xmath12 is the laplace vector .
the laplace vector @xmath12 is related to the eccentricity vector @xmath13 as follows : @xmath14 the three quantities @xmath15 , @xmath11 and @xmath12 satisfy the following two identities : @xmath16 define the sets @xmath17 the following proposition is from @xcite . the following hold : + 1 .
@xmath18 is the union of all non - degenerate elliptic keplerian orbits .
@xmath19 and @xmath20 .
the fiber @xmath21 consists of a unique ( oriented ) non - degenerate elliptic keplerian orbit for each @xmath22 .
[ prop1 ] the equation of the motion with a specific control force is given by @xmath23 where @xmath24 is the control force .
define a metric @xmath25 on @xmath26 by @xmath27 with @xmath28 an arbitrary parameter , and @xmath29 .
let @xmath30 be the open ball of radius @xmath31 centered at @xmath32 in the @xmath25-metric and @xmath33 its closure .
let @xmath34 be the pair of the angular momentum and the eccentricity vector of a target elliptic orbit .
define a lyapunov function @xmath35 on @xmath36 by @xmath37 along the trajectory of ( [ eq : newton ] ) there is @xmath38 hence , @xmath39 where @xmath40 take a controller as follows : @xmath41 with an arbitrary function @xmath42 .
then , @xmath43 the following proposition is proven in @xcite using lasalle s invariance principle @xcite . [ prop:2 ]
let @xmath44 be the pair of the specific angular momentum and laplace vectors of the target elliptic orbit .
take any closed ball @xmath45 of radius @xmath46 centered at @xmath47 contained in the set @xmath48 defined in ( [ set : d ] ) .
then every trajectory starting in the subset @xmath49 of @xmath50 remains in that subset and asymptotically converges to the target elliptic orbit @xmath51 in the closed - loop dynamics ( [ eq : newton ] ) with the control law in ( [ eq : control ] ) the choice of the parameter @xmath52 in the lyapunov function @xmath35 plays a crucial role in determining the transfer trajectory @xcite .
it determines the relative weighting between the two quadratic terms in the function @xmath35 in ( [ eq : lyapunov ] ) . with a small @xmath52 the shape of the trajectory will adjust to that of the target orbit first because a more weight is on @xmath53 . on the other hand , with a large @xmath52 , the normal direction of the trajectory plane will adjust to that of the target orbit plane first because a more weight is on @xmath54 .
the parameter @xmath52 also determines the shape of the region of attraction since @xmath52 determines the shape of the ball @xmath55 in the metric @xmath25 .
additionally , the ccm transfer works well for parabolic transfer , although the success of the transfer is proven exclusively for elliptic orbits only in proposition [ prop:2 ] .
this section presents the transform of the six keplerian elements to specific angular momentum and laplace vectors for convenient reference .
the state vector at periapsis to derive the transform in cartesian coordinates with the earth at the origin .
let @xmath56 the periapsis of the orbit in the geocentric equatorial frame is determined by @xmath57 in the perifocal frame , @xmath58 the transformation matrix @xcite from the perifocal frame into the geocentric equatorial frame is given by @xmath59_{pe}=\left [ \begin{array}{ccc } \cos\omega \cos\omega - \sin\omega \sin\omega\cos i & -\cos\omega \sin\omega -\sin\omega \cos i \cos\omega & \sin\omega \sin i \\
\sin\omega \cos\omega + \cos\omega \cos i \sin\omega & -\sin\omega \sin\omega+\cos\omega \cos i \cos\omega & -\cos\omega \sin i \\ \sin i \sin\omega & \sin i \cos\omega & \cos i \end{array } \right].\ ] ] the state vector in the geocentric equatorial frame is found by carrying out the matrix multiplications @xmath60_{pe}\{\mathbf r\}_p , \{\dot{\mathbf r}\}_e=[\mathbf q]_{pe}\{\dot{\mathbf r}\}_p.\ ] ] thus , the components of @xmath61 and @xmath62 are derived . then using the identity @xmath63 , the specific angular momentum and laplace vectors in the geocentric equatorial frame
are computed as follows : @xmath64 @xmath65 on equatorial orbits @xmath66 , they simplify to @xmath67 on circular orbits @xmath68 , they simplify to @xmath69
the ccm transfer trajectory is used as an initial guess to support the trajectory optimization in the open - loop system using the direct chebyshev - gauss pseudospectral transcription method and a nonlinear programming solver .
the cartesian coordinates and the modified orbital elements in optimization are compared .
let @xmath70 denote the cartesian coordinates in the geocentric equatorial frame .
the minimum fuel consumption problem is given as follows : @xmath71 \\ & \| \mathbf u(t ) \|
\leq f_{\rm max } \ , , \ , \forall t \in [ t_0,t_f ] \\ & \mathbf x(t_0 ) \ , \text { fixed}\\ & \mathbf l_t , \mathbf a_t \ , \text { fixed}\\ \end{array } \right.\ ] ] with @xmath72_{6
\times 6 } , \quad \quad d=-\frac{\mu}{\| \mathbf r \|^3 } , \quad \quad \mathbf u=[0,0,0,f_1,f_2,f_3]^{t},\ ] ] where @xmath73 denotes the state vector , @xmath74 is the @xmath75 identity matrix and @xmath76 the control vector .
the boundary conditions are given by @xmath77 for the minimum time problem , the cost function @xmath78 shall be used . to reduce the continuous optimal control problem ( ocp ) into a discretized non - linear programming ( nlp ) problem ,
the pseudospectral method is used with second - kind chebyshev points . the transformation to express the ocps in the time interval @xmath79 $ ] is given by @xmath80 use lagrange interpolation polynomials with @xmath81 points as follows : @xmath82 where @xmath83 is the initial boundary point , and @xmath84 , @xmath85 , are the @xmath86 collocation points , which are the zeros of the second - kind chebyshev polynomial @xmath87 as expressed below : @xmath88 the weights of the chebyshev - gauss quadrature in this case are given by @xmath89 then the nlp problem can be obtained as ( see @xcite , @xcite ) @xmath90=\mathbf 0 , \\ & \| \mathbf u(\tau_k)\|-f_{\rm max } \leq 0 , \\ & \mathbf s_0(\mathbf x(\tau_0))= \mathbf 0 , \\ & \mathbf s_{l}(\mathbf x(\tau_f))= \mathbf 0 , \\ & \mathbf s_{a}(\mathbf x(\tau_f))=\mathbf 0.\end{aligned}\ ] ] where @xmath91 indicates the @xmath91th collocating point . the augmented cost function with the constraints combined via lagrange multipliers
is given by @xmath92\right ) .
\end{split}\end{aligned}\ ] ] the remaining karush - kuhn - tucker ( kkt ) conditions at the collocating points are given by @xmath93 now , the optimal control problem can be solved by using well - established nlp algorithms .
the same notations as @xcite are utilized here to describe the modified equinoctial orbit elements .
the state variables @xmath94 are defined as @xmath95 where the true longitude @xmath96 is the fast independent variable and the other five are slow dependent variables .
the control variables @xmath97 in rtn ( radial - tangential - normal ) coordinates are defined with @xmath98 then @xmath99 with @xmath100 the system equation is given by @xmath101 the minimum fuel consumption orbit transfer problem can be written as @xmath102 \\ & \| \mathbf f(t ) \| \leq f_{\rm max } \ , , \ , \forall t \in [ t_0,t_f ] \\ & \mathbf x_m(t_0 ) \ , \text { fixed}\\ & \mathbf x_m(t_f ) \ , \text { fixed}\\ \end{array } \right.\ ] ] the optimality conditions expressed in terms of modified elements are not provided here since those presented above in cartesian coordinates also apply to modified elements .
using an example , this section illustrates transfers from a low - earth orbit ( leo ) to a geosynchronous orbit ( geo ) in terms of minimum transfer time and minimum fuel consumption .
the numerical data used are from @xcite , with the exception of the final time in the minimum fuel case .
the initial point is given by @xmath103 the target orbit is given by @xmath104 with the control constraint @xmath105 km / sec@xmath106 and the initial time @xmath107 . in the minimum fuel case , the fixed final time @xmath108 is @xmath109 hr . canonical units are used in simulations , where 806.812 sec = 1 canonical time unit ; 6378.140 km = 1 canonical distance unit ; @xmath110 km / sec@xmath111 = 1 canonical acceleration unit ; and the gravitational parameter @xmath112 . in the following ,
all units are canonical unless otherwise indicated .
the initial and final conditions in cartesian coordinates are given by the initial and final conditions in modified elements are given by @xmath115 @xmath116 with @xmath117 . in the minimum fuel case ,
the fixed final time is @xmath118 , but in the minimum time case the final time is free , so it must first be adjusted according to the final time guess of @xmath119 .
the ccm controller given in ( [ eq : control ] ) is applied with the weighting @xmath120 and the function @xmath121 to obtain a transfer trajectory that provides an initial guess for optimization .
then tomlab / propt is utilized together with the pseudospectral method and snopt solver to optimize the trajectory on the matlab platform . to compare the different coordinate systems ,
all the collocating points are located in one phase .
the optimal results are listed in table [ table : c ] , and the minimum fuel and minimum time transfer trajectories are shown in fig .
[ fig : transfer_c ] and fig .
[ fig : transfer_time_c ] , respectively . for the minimum fuel case , fig .
[ fig : force comparison c ] compares the force magnitude between the two coordinate systems presented above .
[ fig : la c ] and fig .
[ fig : r_dr_c ] show the time history of @xmath122 @xmath123 , and @xmath124 , and fig .
[ fig : force_c ] details the time history of the control force in the cartesian coordinates system . in the time
history plots , the dotted lines represent the initial guesses and the solid lines represent final trajectories . | this paper presents a simple approach to low - thrust optimal - fuel and optimal - time transfer problems between two elliptic orbits using the cartesian coordinates system . in this case ,
an orbit is described by its specific angular momentum and laplace vectors with a free injection point .
trajectory optimization with the pseudospectral method and nonlinear programming are supported by the initial guess generated from the chang - chichka - marsden lyapunov - based transfer controller .
this approach successfully solves several low - thrust optimal problems .
numerical results show that the lyapunov - based initial guess overcomes the difficulty in optimization caused by the strong oscillation of variables in the cartesian coordinates system .
furthermore , a comparison of the results shows that obtaining the optimal transfer solution through the polynomial approximation by utilizing cartesian coordinates is easier than using orbital elements , which normally produce strongly nonlinear equations of motion . in this paper , the earth s oblateness and shadow effect are not taken into account .
chang - chichka - marsden lyapunov - based transfer , trajectory optimization , cartesian coordinates |
the mechanism of core - collapse supernovae has been an open question for decades . in the past 15 years , a lot of detailed numerical simulations have been done by many authors , paying attention to various aspects of the dynamics ( see , for example , @xcite for the latest review of the subject ) .
spherically symmetric models have incorporated full general relativity and solved the radiation - hydrodynamical equations with realistic equations of state and neutrino physics @xcite .
these computations demonstrated consistently that the stalled shock wave is highly unlikely to be revived by neutrino heatings in the spherically symmetric dynamics .
multi - dimensional simulations have been also making a remarkable progress over the times ( see @xcite and references therein ) .
local non - spherical motions of convection - type are common in the envelope as inferred from the spectra and light curves @xcite , and are believed to occur also in the core .
the global asymmetry as observed in the ejecta of sn1987a is also supposed to be common to the core - collapse supernovae @xcite . motivated by these observations , the implications of the asymmetry for the explosion mechanism have been investigated intensively and extensively ( see again @xcite and references therein ) .
recently , the instability of the standing accretion shock wave was discovered by @xcite in the context of core - collapse supernova and its nature has been studied both analytically @xcite and numerically @xcite .
the ramifications to the global asymmetry of the ejecta as mentioned above and the high kick velocities of young pulsars @xcite were discussed by @xcite in more realistic simulations .
since it has been demonstrated @xcite over the years that these instabilities tend to facilitate the shock revival by neutrino heatings , attempts have been done with little success @xcite so far to show by realistic multi - dimensional simulations that an explosion is indeed induced .
this situation may change .
very recently , @xcite have proposed a new possibility to induce explosions .
they have done long - term simulations of axisymmetric 2d radiation - hydrodynamics .
although some microphysics are incomplete and the formulation is essentially non - relativistic , their models with no equatorial symmetry and artificial inner boundary condition being imposed have obtained a shock revival after a rather long delay ( @xmath6ms after bounce ) not by the neutrino heatings but by what they call the `` acoustic '' mechanism .
they have claimed that the turbulence of accreted matter induces mainly @xmath0-mode oscillations in the proto neutron star , which in turn emit acoustic waves that are dissipated as they propagate outward to become shock waves .
this is a quite new idea that has eluded the discovery by multi - dimensional numerical studies for many years , the reasons for which are , the authors think , that previous computations were not long enough and/or put too restrictive conditions such as the artificial inner boundary that has been frequently employed to reduce the cfl constraint on numerical time steps . @xcite
have insisted that the merit of this mechanism is the self - regulation in such a way that the @xmath0-mode excitations and subsequent emissions of acoustic waves are maintained until the shock revival occurs .
this is an interesting possibility and certainly worth further exploration .
there are indeed a lot of things remaining to be clarified .
first of all , the mechanism seems to be rather inefficient in transferring the energy .
inferred from the published results , the explosion energy appears to be substantially smaller than the canonical value .
since the mechanism constitutes of multiple steps , the efficiency of the couplings between different steps should be understood . for that purpose
, we think that the linear analysis utilizing an idealized model that mimics the realistic situation is more suitable than large scale simulations . in this paper
, we calculate the eigen frequencies and eigen functions of the @xmath0-modes in a non - rotating proto neutron star . assuming that the standing accretion shock instability ( sasi ) is responsible for the excitation of the @xmath0-modes in the acoustic mechanism
, we estimate the efficiency , formulating and solving the equations for forced oscillations of the @xmath0-modes . in so doing , we use the pressure fluctuations in sasi that we obtained in the previous numerical studies @xcite .
the emissions of acoustic waves by these excited @xmath0-modes are currently being investigated numerically and will be published elsewhere @xcite . the organization of the paper is as follows . in section [ form ] , we describe the models of proto neutron star and formulate the basic equations for the normal modes and their forced oscillations .
the sasi that we use as an external force to drive the oscillations of @xmath0-modes is also presented there .
we give the numerical results in section [ result ] .
we conclude the paper in section [ conclusion ] with some discussions .
we will investigate the @xmath0-mode excitations in the proto neutron star , one of the key steps in the acoustic mechanism , under the following assumptions : ( i ) a non - rotating proto neutron star is approximately in hydrostatic equilibrium .
( ii ) sasi in the accreting matter is responsible for the excitations of @xmath0-modes .
( iii ) the oscillation amplitudes in the proto neutron star are small compared with the radius . as shown in the numerical simulations done by @xcite , these assumptions are reasonable after the stagnation of a prompt shock wave until the excited @xmath0-modes grow sufficiently and their non - linear effects on the dynamics become significant .
then the dynamics of the proto neutron star can be treated by the linear perturbation theory formulated below .
the linear perturbation of stars can be described in terms of the lagrangian displacement @xmath7 , which obeys the equation of motion given as @xmath8 where @xmath9 and @xmath10 are , respectively , the self - adjoint operator and the external force driving the perturbations ( see , e.g. , friedman & schutz 1978 ; unno et al . 1989 ) .
here , we have assumed the unperturbed star to be static and spherically symmetric . assuming the lagrangian displacement to be @xmath11 where @xmath12 and @xmath13 are the eigen function and eigen frequency , respectively
, we can obtain the normal mode solutions , for which the eigen function and eigen frequency satisfy the following equation , @xmath14 here , @xmath15 denotes the mode index , which includes for spherical stars the angular eigen values @xmath1 and @xmath16 ( see below ) and the radial mode number @xmath17 .
we impose for the normal mode solutions the regularity condition at the stellar center and @xmath18 at the stellar surface , where @xmath19 stands for the lagrangian change of the pressure @xmath20 .
when the unperturbed star is static and spherically symmetric , the orthogonality of the eigen functions allows us to adopt the normalization @xmath21 where @xmath22 , @xmath23 and @xmath24 denote the density , mass and radius of the unperturbed star , respectively ( see , e.g. , unno et al .
any lagrangian displacement @xmath25 that satisfies the same boundary conditions as those for the normal modes is then expressed as the linear superposition of the normal modes : @xmath26 substituting eq .
( [ expanded_efun ] ) into eq .
( [ basic_eq1 ] ) and using eqs .
( [ basic_eq2 ] ) and ( [ norm_eq ] ) , we obtain the dynamical equation for the expansion coefficient @xmath27 as @xmath28 in this paper , we study adiabatic oscillations of a non - rotating spherical proto neutron star and employ the cowling approximation , neglecting perturbations of the gravitational potential ( see , e.g. , unno et al .
1989 ) . for spherical stars ,
the separation of the angular variables in the perturbation equations can be achieved by the use of the spherical harmonics in the spherical polar coordinates ( @xmath29 , @xmath30 , @xmath31 ) .
the lagrangian displacement vector @xmath25 is expressed in terms of the vector spherical harmonics as @xmath32 and the eulerian perturbation of pressure @xmath33 is given by @xmath34 where @xmath0 denotes the local gravitational acceleration .
( [ basic_eq2 ] ) is then reduced to @xmath35 @xmath36 where @xmath37 here , @xmath38 and @xmath39 stand for the adiabatic exponent and the stellar mass contained inside the radius @xmath29 , respectively .
the inner and outer boundary conditions are explicitly given by @xmath40 the total mode energy @xmath41 is defined as @xmath42 where @xmath43 is the eulerian perturbation of density , and @xmath44 and @xmath45 mean the sound velocity and the brunt - visl frequency defined , respectively , as @xmath46 in order to obtain eigen frequencies and eigen functions , eqs .
( [ mseq1 ] ) and ( [ mseq2 ] ) are solved numerically as an eigen value problem for @xmath47 , together with the boundary conditions ( [ bc1 ] ) and ( [ bc2 ] ) . a henyey - type relaxation method is employed ( see , e.g. , unno et al .
1989 ) . introducing new variables
@xmath48 and @xmath49 defined as @xmath50 @xmath51 we can rewrite the set of the pulsation equations ( [ mseq1 ] ) and ( [ mseq2 ] ) in a canonical form given by @xmath52 @xmath53 where @xmath54 is the lamb frequency defined as @xmath55 here , the function @xmath56 is positive definite and is defined by @xmath57\,.\ ] ] the canonical pulsation equations ( [ caneq1 ] ) and ( [ caneq2 ] ) are more appropriate to gain the insight into the qualitative properties of nonradial oscillations . ignoring the spatial variations of all the coefficients in eqs .
( [ caneq1 ] ) and ( [ caneq2 ] ) and employing a wkb - type local approximation , we obtain the eigen functions in the form , @xmath58 where @xmath59 means a radial wave number of the perturbations .
then the following dispersion relation is obtained : @xmath60 which implies that if @xmath61 , the wave number @xmath59 is real and the corresponding wave can propagate in the radial direction , whereas the wave number @xmath59 becomes purely imaginary and the amplitude of the perturbation grows or damps exponentially with @xmath29 when @xmath62 . in this study , we are concerned with the excitation of oscillations in the proto neutron star by the turbulence in accreting matter . in our formulation , we take into account the latter effect as pressure perturbations at the surface of proto neutron star .
hence we set @xmath63 in eq .
( [ basic_eq1 ] ) and adopt the time - dependent non - vanishing pressure perturbation given by @xmath64 instead of the ordinary condition @xmath18 at the proto neutron star surface . here
the function of time , @xmath65 , describes how the stellar surface is disturbed by the accreted matter and is taken from the numerical results obtained in our previous exploration of sasi ( see section [ sasi ] ) .
note that the boundary condition ( [ def_f ] ) is no longer the same as that for the normal modes , @xmath18 at @xmath66 .
however , we can still use an expansion by the normal modes , @xmath12 , for a part of the lagrangian displacement by introducing a new dependent variable @xmath67 defined as @xmath68 where @xmath69 substituting @xmath70 into eq .
( [ basic_eq1 ] ) , we obtain the master equation for @xmath67 given by @xmath71 in eq .
( [ def_zeta ] ) , @xmath72 is chosen so that @xmath67 could satisfy the same boundary conditions as those for the normal modes : @xmath73 and @xmath67 is regular at @xmath74 . as a result , we can express @xmath67 as a linear superposition of the eigen functions for the normal modes , @xmath75 the expansion coefficient @xmath27 is again determined by the equation similar to eq .
( [ dynamic_eq ] ) , @xmath76 where @xmath77 . in eq .
( [ dynamic_eq2 ] ) , the driving - force term @xmath78 is explicitly written as @xmath79 where @xmath80 and @xmath81 are constants determined by the background structure of the proto neutron star and the eigen functions for the normal modes and are given as @xmath82 \label{def_a } \
\\ b&=&\left({1\over\gamma}\right)_{r = r}\int_0 ^ 1\ !
dx \ , { \rho r^3\over m}x^{\ell+4 } { y^*_1\over \ell+3 } \ , .
\label{def_b}\end{aligned}\ ] ] here the normalized radius @xmath83 is introduced .
once we obtain @xmath27 , the lagrangian displacement vector @xmath25 is given by @xmath84 since @xmath72 does not satisfy , in general , the orthogonality , @xmath85 @xmath72 includes the components proportional to @xmath12 , and the expansion coefficient @xmath27 is only a part of the excited oscillation mode .
the true amplitude @xmath86 of the excited mode @xmath12 in @xmath25 can be obtained by the inner product between @xmath25 and @xmath12 as @xmath87 substituting eqs .
( [ def_efun ] ) and ( [ def_zeta ] ) into eq .
( [ def_bamp ] ) , we obtain @xmath88 it is reasonable to assume that no modes are excited in the proto neutron star initially , i.e. , @xmath89 and @xmath90 at @xmath91 .
accordingly , we start the integration of eq .
( [ dynamic_eq2 ] ) with the following initial conditions , @xmath92 as already mentioned above , we assume in this paper that oscillations in the proto neutron star are excited by the turbulence of accreting matter , which is supposed to be induced by sasi .
sasi is a non - local hydrodynamical instability that occurs in the accretion flow through the standing shock wave onto the proto neutron star @xcite .
although the mechanism is still controversial @xcite , sasi is probably driven by the cycle of inward advections of entropy- and vortex - perturbations and outward propagations of pressure fluctuations @xcite .
the prominent feature of sasi is the dominance of low @xmath1 modes both in the linear and non - linear regimes . in particular ,
the @xmath93 mode is thought to be a promising cause for the proper motions of young pulsars . in this paper , we regard sasi as a source of pressure perturbations exerted on the surface of the proto neutron star , that is , @xmath65 in eq .
( [ def_f ] ) .
we employ the numerical results obtained by @xcite , in which we have done 2d axisymmetric simulations of the accretion flows through the standing shock wave onto the proto neutron star .
the realistic equation of state by @xcite and the standard reaction rates @xcite for the neutrino absorptions and emissions on nucleons were employed .
the mass accretion rate and the neutrino luminosity are the parameters that specify a particular steady accretion flow and were held constant during the computations . the self - gravity of the accreting matter was ignored and only the gravitational attraction by the proto neutron star was taken into account .
the mass of the proto neutron star was also set to be constant ( 1.4m@xmath94 ) so that we could obtain completely steady background flows , to which velocity perturbations were added initially . in the following ,
we use the results for the model with the mass accretion rate of @xmath950.19m@xmath94/s and neutrino luminosity of @xmath96erg / s , which are typical for the post bounce core around the shock stagnation .
the neutrino spectra were assumed to take a fermi - dirac distribution with a vanishing chemical potential and a temperature of @xmath97mev for the electron - type neutrino and @xmath98mev for the electron - type anti - neutrino .
the random velocity perturbation of less than @xmath99 was added to the radial velocity initially .
we refer the readers to @xcite for more detailed formulations .
we observed sasi growing exponentially in the linear phase that lasts for @xmath100100ms and then saturated to enter the non - linear phase .
the @xmath93 feature was clearly seen in the non - linear stage for this model although other modes were also generated by non - linear mode couplings .
we expanded the eulerian pressure perturbation at the inner boundary by the spherical harmonic functions .
we used only the @xmath93 component in this study , since this was claimed to be the most important mode @xcite .
the power spectrum of this component is shown in fig .
[ fig4 ] . as can be seen
, the pressure perturbation is actually a series of impulsive forces that last for quite short times .
it should be noted that the linear analysis of sasi for this model gives the eigen frequency of @xmath101hz for the linear @xmath93 mode .
the non - linear couplings broaden the spectrum both to the lower and higher frequencies .
as seen later in section [ result ] , the spectrum has most of the power in much lower frequencies than the typical @xmath0-mode frequencies of several hundreds hz , which leads to a severe impedance mismatch in exciting @xmath0-modes . it should be mentioned that the model studied here is not fully self - consistent . for example , what we need in eq .
( [ def_f ] ) is the lagrangian perturbation of pressure while what we obtained above is the eulerian perturbation .
this is simply because the former is difficult to obtain from the numerical data .
although they are different , rigorously speaking , we think that the difference will not change the outcome qualitatively , since it is the typical frequency and spectral feature of sasi that matters in this study .
we should also bear in mind that the sasi spectrum employed in this study is just a single realization of essentially chaotic flows although we expect it is representative .
finally , the spherically symmetric , stationary proto neutron star that is adopted as the unperturbed state in this study is briefly described .
the model is based essentially on the results of detailed radiation - hydrodynamical simulation of core - collapse by @xcite , but we have modified it in the following way . we have developed a numerical code to compute the long - term post - bounce evolution in the quasi - steady approximation @xcite , the details of which will be published elsewhere . in this formalism , we solve separately but consistently three different regions in the collapsing star simultaneously : ( 1 ) the proto neutron star that cools in a quasi - static way .
we calculate both the equilibrium configuration and the neutrino transport in it with matter accretion taken into account .
( 2 ) the steady accretion flow from the standing shock wave down to the proto neutron star for the given neutrino luminosity obtained in the calculation ( 1 ) and matter accretion rate given by the next computation . (
3 ) the dynamical implosion of the envelope toward the standing shock wave .
the second and third regions are joined by the rankine - hugoniot condition at the shock wave consistently .
the code can follow the spherically symmetric quasi - static evolutions for more than 10 seconds and we have confirmed that the evolution is well reproduced for the first second , for which the result of the detailed dynamical simulations is available . in this study , we utilize the proto neutron star , which is obtained in this approximation and is in hydro - static equilibrium at @xmath100360ms after bounce .
physical quantities characterizing the proto neutron star model , that is , mass @xmath23 , radius @xmath24 , and kepler frequency @xmath102 are given in table [ table1 ] .
the profiles of density , pressure , specific entropy , and electron fraction are shown in fig .
[ fig0 ] as a function of radius .
it is noted that this spherically symmetric model contains a convectively unstable region around @xmath103 km . here
again , the proto neutron star model is not very consistent with the model for sasi .
as mentioned in the previous section , it is not our aim to construct a fully self - consistent model in this paper , where the elementary feature of the @xmath0-mode excitations is explored qualitatively from an experimental point of view . for that purpose ,
the slight inconsistency in the current model does not matter .
we first consider the oscillation properties of @xmath0-modes for the proto neutron star model described in the previous section . before giving the eigen functions , we look into the so - called propagation diagram ( see , e.g. , unno et al .
1989 ) , which is useful to see where in the star the wave with a particular frequency @xmath104 can propagate .
the diagram shows the square of the lamb frequency @xmath105 and the brunt - visl frequency @xmath106 as a function of the normalized radius @xmath107 . as explained in section [ form ] , since the waves can propagate when @xmath61 is satisfied , the region where this condition holds true is called the propagation zone , and the other regions in the diagram are referred to as an evanescent zone , in which the perturbations do not oscillate but grow or damp exponentially .
[ fig1 ] is the propagation diagram for the present proto neutron star model , in which we take @xmath93 to calculate the lamb frequency and the frequency is normalized according to eq .
( [ omgbar ] ) .
the propagation regions can be classified into two categories : one is the @xmath0-mode propagation zone indicated as hatched regions and the other is the @xmath20-mode propagation zone shown as a cross - hatched region in fig .
[ fig1 ] . as mentioned already ,
since the proto neutron star has a convective region around @xmath108 , which is an evanescent zone for @xmath0-modes , there are two disconnected @xmath0-mode propagation zones , one is near the center of the star and the other near the surface .
this implies that a particular @xmath0-mode is almost trapped in one of these two disconnected @xmath0-mode propagation zones .
in other words , the @xmath0-modes for the current proto neutron star model are essentially divided into two classes , i.e. , what we refer to as the core @xmath0-modes ( @xmath109-modes ) and the surface @xmath0-modes ( @xmath110-modes ) , whose amplitudes of the eigen functions have a peak near the center and the surface of the proto neutron star , respectively .
we now discuss the properties of the normal mode oscillations obtained numerically for the current proto neutron star model . in table
[ table2 ] , we summarize the key quantities for several low - order @xmath0-modes : the normalized angular eigen frequencies , @xmath111 , the corresponding frequencies in hz , @xmath112 , the mode energies , @xmath41 , and the integral quantities , @xmath80 and @xmath81 , defined by eqs .
( [ def_a ] ) and ( [ def_b ] ) .
the last two are important in evaluating the excitations of @xmath0-modes by external perturbations in the next section .
the mode energy gives us a rough measure of non - linear oscillations in the proto neutron star .
note that we employ the normalization condition for the eigen functions , eq .
( [ norm_eq ] ) , in evaluating the mode energy @xmath41 .
as expected , the mode energy is of the order of @xmath113ergs .
the typical frequency of low - order @xmath0-modes is several hundreds hz , and we can confirm that the numbers given in table [ table2 ] are comparable with those found by @xcite .
it should be also emphasized that these frequencies are much larger than the typical frequency , @xmath114100 hz , of sasi in the non - linear stage .
it is also recognized in table [ table2 ] that for @xmath115 the lowest mode is a surface mode and core modes and surface modes appear alternately as one goes to higher overtones . in fig . [ fig2 ] , the eigen functions , @xmath116 and @xmath117 in eqs .
( [ def_efun ] ) and ( [ def_ep ] ) , are shown for the @xmath0-modes with @xmath93 as a function of the normalized radius of the proto neutron star @xmath107 . in the figure
, the @xmath118-mode means the @xmath17-th radial overtone @xmath0-mode , which is characterized by @xmath17 radial nodes . as mentioned just above
, the @xmath119- and @xmath4-modes are a typical surface and core @xmath0-mode , respectively .
in fact , the eigen functions have a non - negligible amplitude either outside or inside the convection region , which is an evanescent zone for @xmath0-modes and located at @xmath120 km .
this is also true for other @xmath0-modes with different @xmath1 or @xmath17 . upon using these eigen functions , the quantities , @xmath80 and @xmath81 , which appear in eq .
( [ def_src ] ) and give the driving force for mode excitations , are obtained by integrating eqs .
( [ def_a ] ) and ( [ def_b ] ) .
although the general trend is that their absolute values decrease as the radial node number increases , there are exceptions that are critically important in determining which mode is most efficiently excited by external perturbations as explained below .
now we move on to the excitation by accreting matter of the @xmath0-modes considered just above . as mentioned before , we assume in this study that the @xmath0-modes are excited by the pressure fluctuations around the surface of the proto neutron star , which are induced by sasi in the accretion flow . the pressure fluctuation @xmath65 in eq .
( [ def_f ] ) is obtained from the numerical results on sasi and is displayed in fig .
[ fig4 ] as a power spectrum .
it should be repeated that the major peaks of the spectrum are observed in the frequencies much lower than those of the low - order @xmath0-modes given in table [ table2 ] .
in other words , the typical frequency of sasi , the driving force , is much lower than that of the low - order @xmath0-modes that should be excited in the proto neutron star .
it is true that these numbers are dependent on the models of proto neutron star and sasi as well as eos employed , but we expect that the mismatch of the frequencies is a rather generic issue . by making use of the pressure fluctuations shown in fig .
[ fig4 ] and solving eq .
( [ dynamic_eq2 ] ) together with the initial condition given in eq .
( [ initial_data ] ) , we calculate the time evolution of the mode amplitude , @xmath121 , for several @xmath0-modes . among them ,
the most efficiently excited @xmath0-mode is the @xmath122-mode with @xmath93 .
this is mainly because it has the largest @xmath80 as given in table [ table2 ] . in fig .
[ fig3 ] , we show the temporal evolution of the mode energy for the @xmath122-mode with @xmath93 . since the driving forces , that is , the pressure fluctuations induced by sasi are a collection of impulsive forces , the mode energy is fluctuating on a short time scale .
it is clear , however , that the mode is growing for @xmath100100ms initially and then is saturated .
the maximum value of the excited mode energy is @xmath123 for this mode , which is considerably smaller than the typical energy of non - linear oscillations , @xmath124erg , given in table [ table2 ] . in this sense ,
the excitation of the @xmath0-modes is inefficient , at least for the present model .
this is a direct consequence of the severe mismatch in the frequencies of the driving force , sasi , and the excited @xmath0-modes .
hence it is more appropriate to refer to what we see here as the forced oscillation rather than as the excitation .
it is noted , however , that the eulerian pressure perturbations at @xmath125 km are found to be of the same order as those found in @xcite .
the maximum values of the mode energy for other @xmath0-modes , @xmath126erg , are even smaller than that of the @xmath122-mode with @xmath93 for the proto neutron star model considered here .
we show in fig .
[ fig6 ] the temporal evolutions of the mode energy for the @xmath119- and @xmath4-modes with @xmath93 .
the essential feature is the same as for the @xmath4-mode shown in fig .
these modes also grow for about 100ms initially and then get saturated with large fluctuations on a short time scale .
it is emphasized again that all these excitations are non - resonant . in that case , the maximum amplitude of @xmath127 can be approximated by the maximum value of @xmath128 , which , for example , becomes @xmath129 for the @xmath4-mode and an - order - of - magnitude smaller for other modes with @xmath93 . from eq .
( [ emode ] ) , we then have approximate values of the mode energies , which are consistent with the numerical results given in figs . [ fig3 ] and [ fig6 ] .
inspired by a series of papers on the acoustic mechanism of core - collapse supernova by @xcite , we have calculated in this paper the eigen frequencies and eigen functions for several low - order @xmath0-modes in the proto neutron star , which were claimed to play an important role in the acoustic mechanism , and have studied also their excitations by sasi in the accreting matter .
we have constructed a proto neutron star model , which is supposed to represent a quasi - equilibrium configuration around a few hundred milliseconds after bounce , based on the detailed radiation - hydrodynamical simulations by @xcite .
we have employed the temporal variation of pressure on surface of the proto neutron star , which were obtained in the study on sasi by @xcite .
we have found that the @xmath0-modes are classified either as the surface @xmath0-mode or as the core @xmath0-mode according to the profile of the eigen function , since there is a convective zone in the proto neutron star , which is an evanescent zone for the @xmath0-modes .
the eigen frequencies of the low - order @xmath0-modes are several hundred hz , which are larger in general than the typical frequency of sasi , @xmath114100hz , in the non - linear regime . assuming that sasi is the driving force for the @xmath0-mode excitations , we have solved numerically the equations for the time evolution of the @xmath0-mode amplitude with the pressure perturbations at the surface of the proto neutron star taken into account .
we have observed that the @xmath0-mode amplitudes grow for @xmath100100ms initially and then get saturated . for the proto neutron star model considered in this paper , the @xmath122-mode , that is , the second overtone core @xmath0-mode with @xmath93 is found to have the largest mode energy at saturation . even for this mode ,
the mode energy , @xmath130erg , is much smaller than the typical energy , @xmath124erg , for non - linear oscillations , which can be estimated by imposing the natural normalization condition to the eigen functions .
this apparent inefficiency of the @xmath0-mode excitations is a direct consequence of the mismatch of the frequencies of the @xmath0-modes and sasi .
it is noted , however , that what is important for the acoustic mechanism may not be the mode energy .
in fact , the @xcite mentioned that the @xmath0-mode oscillations acted like a transducer that transformed the gravitational energy of accreted matter to the energy of out - going sound waves . in the present formulation , unfortunately , it is in principle impossible to obtain the energy that will be transferred to the out - going acoustic waves .
however , we may be able to infer it as follows . in the growth phase that lasts for @xmath100100ms
, the energy of accreting matter , or the work done by the external pressure perturbation in the present model , is transferred to the mode energy , as is apparent in fig .
[ fig3 ] . after saturation , on the other hand , the mode energy is almost unchanged . in this phase , then , it may be that the energy of the accreting matter is emitted just at the same rate as the accumulation rate .
if this is really the case , we can estimate that the emitted energy , possibly as out - going acoustic waves , will be @xmath131erg / s , which may not be a bad estimate , provided , in general , the explosion is considerably delayed ( @xmath132500ms ) in the acoustic mechanism .
it is obvious , however , that more elaborate formulation of the problem , which would include the out - going waves , is required .
numerical approaches from an experimental view point will be also useful @xcite .
there are many other improvements remaining to be done .
as repeated in the text , we have not attempted to make the model fully self - consistent in this paper , since the qualitative understanding of the @xmath0-mode excitations by sasi is the focus of the paper .
however , the dependence of the results on the proto neutron star model , eos , the progenitor model that determines the accretion rate , and so forth should be investigated in addition to making the model more self - consistent , all of which are currently under way .
this work is partially supported by the grant - in - aid for the 21st century coe program `` holistic research and education center for physics of self - organizing systems '' of waseda university and for scientific research ( 14740166 , 14079202 , young scientist ( b ) 17740155 ) of the ministry of education , science , sports and culture of japan .
blondin , j. m. , mezzacappa , a. , & demarino , c. 2003 , , 584 , 971 blondin , j. m. , & mezzacappa , a. 2006a , , 642 , 401 blondin , j. m. , & mezzacappa , a. 2006b , to be published in nature , astro - ph/0611680 buras , r. , janka , h. -th . ,
rampp , m. , & kifonidis , k. 2006 , a&a , 457 , 281 burrows , a. , livne , e. , dessart , l. , ott , c. d. , & murphy , j. 2006a , , 640 , 878 burrows , a. , livne , e. , dessart , l. , ott , c. d. , & murphy , j. 2006b , to be published in , astro - ph/0610175 burrows , a. , dessart , l. , ott , c. d. , & livne , e. , 2006c , to be published in the `` centennial festschrift for hans bethe , '' physics reports ( elsevier : holland ) , ed .
brown , e. van den heuvel , and v. kalogera , 2006 , astro - ph/0612460 bruenn , s. w. 1985 , , 58 , 771 foglizzo , t. , scheck , l. , & janka , h. -th .
2006 , , 652 , 1436 friedman , j. l. , & schutz , b. f. 1978 , , 222 , 281 herant , m. , benz , w. , hix , w. r .. , fryer , c. l. , & colgate , s. a. 1994 , , 435 , 339 janka , h. -th . , & mller , e. 1996 , a&a , 306 , 167 kotake , k. , sato , k. , & takahashi , k. 2006 , rept.prog.phys .
69 , 971 - 1144 liebendrfer , m. , rampp , m. , janka , h. t. , & mezzacappa , a. , 2005 , , 620 , 840 lyne , a. g. , & lorimer , d. r. , 1994 , nature , 369 , 127 nomoto , k. , tominaga , n. , umeda , h. , kobayashi , c. , & maeda , k. 2006 , nucl .
phys . , a 777 , 424 ohnishi , n. , kotake k. , & yamada , s. 2006a , , 641 , 1018 ohnishi , n. , kotake k. , & yamada , s. 2006b , in preparation rampp , m. , & janka , h. -th . 2000 , l , 539 , l33 scheck , l. , plewa , t. , janka , h .- th . , kifonides , k. , & mller , e. 2004 , , 92 , 011103 shen , h. , toki , h. , oyamatsu , k. , & sumiyoshi , k. 1998 , nucl .
a , 637 , 48 sumiyoshi , k. , yamada , s. , suzuki , s. , shen , h. , chiba , s. , & toki , h. 2005 , , 629 , 922 thompson t a , burrows a and pinto p a 2003 , 592 , 434 unno , w. , osaki , y. , ando , h. , saio , h. , & shibahashi , h. 1989 , nonradial oscillations of stars ( tokyo : univ .
tokyo press ) wang , l. , wheeler , j. c. , hflich , p. , khokhlov , a. , baade , d. , branch , d. , challis , p. , filippenko , a. v. , fransson , c. , garnavich , p. , kirshner , r. p. , lundqvist , p. , mcray , r. , panagia , n. , pun , c. s. j. , phillips , m. m. , sonneborn , g. , & suntzeff , n. b. 2002 , , 579 , 671 watanabe , m. 2006 , master thesis ( waseda univ . ) yamasaki , t. , & yamada , s. 2006 , to be published in , astro - ph/0606581 ccccccrr & 1 & @xmath133 & 2.92 & 499 & 5.70 & 3.50e-03 & 1.72e-03 + & & @xmath134 & 1.43 & 245 & 1.30 & 2.13e-02 & @xmath1352.99e-04 + & & @xmath136 & 1.03 & 176 & 1.97 & 1.67e-03 & @xmath1351.54e-04 + & 2 & @xmath133 & 3.15 & 538 & 7.51 & 5.91e-03 & 9.97e-04 + & & @xmath134 & 2.24 & 383 & 3.15 & 2.02e-03 & @xmath1351.01e-04 + & & @xmath136 & 1.72 & 294 & 7.11 & 1.02e-03 & 2.38e-04 + | the so - called `` acoustic revival mechanism '' of core - collapse supernova proposed recently by the arizona group is an interesting new possibility . aiming to understand the elementary processes involved in the mechanism , we have calculated the eigen frequencies and eigen functions for the @xmath0-mode oscillations of a non - rotating proto neutron star .
the @xmath0-modes have relatively small eigen frequencies .
since there is a convective region inside the proto neutron star , the eigen functions are confined either to the exterior or interior of this evanescent region although some fraction of eigen functions penetrates it because of their long wave lengths .
the possible excitation of these modes by the standing accretion shock instability , or sasi , is discussed based on these eigen functions .
we have formulated the forced oscillations of @xmath0-modes by the external pressure perturbations exerted on the proto neutron star surface .
the driving pressure fluctuations have been adopted from our previous computations of the axisymmetric sasi in the non - linear regime .
we have paid particular attention to low @xmath1 modes , since these are the modes that are dominant in sasi and that the arizona group claimed played an important role in their acoustic revival scenario . here
@xmath1 is the index of the spherical harmonic functions , @xmath2 . although the frequency spectrum of the non - linear sasi is broadened substantially by non - linear couplings , the typical frequency is still much smaller than those of @xmath0-modes , the fact leading to a severe impedance mismatch . as a result , the excitations of various @xmath0-modes are rather inefficient and the energy of the saturated @xmath0-modes is @xmath3erg or smaller , with the @xmath4 mode being the largest in our model . here
the @xmath4 mode has two radial nodes and is confined to the interior of the convection region .
the energy transfer rate from the @xmath0-modes to out - going sound waves is estimated from the growth of the @xmath0-modes and found to be @xmath5erg / s in the model studied in this paper . |
the formation of mesoscopic ( nano / micro ) scale patterns ( interpreted broadly ) in equilibrium systems is often due to a competition between interactions favoring different microscopic structures . as
an example suppose we begin with a short ranged potential favoring local `` alignment '' of the microscopic constituents of the system , e.g. nearest neighbor ferromagnetic interactions of an ising spin system . at low temperature
this would lead to essentially all spins pointing in the same direction .
if we now add a long range interaction which does not like this ordering then the system will do the best it can by forming mesoscopic domains with different aligments .
such a competition occurs in many systems .
it is well illustrated in patterns observed in the low temperature structure of thin films .
these quasi-2d systems show a wide range of ordering effects including formation of striped states , reorientation transitions , etc .
@xcite , see figure 1 .
@xmath0 , with @xmath1 . from m. seul and r. wolfe , pra ( 1992 ) ] the origins of these phenomena can , in many cases , be traced to the competition between short ranged exchange ( ferromagnetic ) interactions , favoring a homogeneous ordered state , and long ranged dipole - dipole type interactions , which oppose such ordering on the scale of the whole sample @xcite .
this type of competitive interaction is believed responsible for many of the observed patterns in a great variety of systems , including thin magnetic films @xcite , micromagnets @xcite , diblock copolymers @xcite , anisotropic electron gases @xcite , langmuir monolayers @xcite , lipid monolayers @xcite , liquid crystals @xcite , polymer films @xcite , polyelectrolytes @xcite , charge - density waves in layered transition metals @xcite , superconducting films @xcite , colloidal suspensions and cell membranes @xcite .
many of these systems are characterized by low temperature phases displaying periodic mesoscopic patterns , such as stripes or bubbles .
the simplest models to describe such systems are ising or `` soft spin '' models with a short range ferromagnetic interaction and a power law long range antiferromagnetic pair potential @xcite .
the zero temperature phase diagram of these models has been thoroughly investigated over the last decade and a sequence of transitions from an antiferromagnetic homogeneous state to periodic striped or lamellar phases with domains of increasing sizes has been predicted , as the strength of the ferromagnetic coupling is increased from zero to large positive values .
these theoretical predictions are mostly based on a combination of variational techniques and stability analysis : they start by _ assuming _ a periodic structure , proceed by computing the corresponding energy and finally by comparing that energy to the energy of other candidate structures , usually by a combination of analytical and numerical tools .
these calculations give an excellent account of some of the observed `` universal '' patterns displayed by the aforementioned systems .
however they run the risk of overlooking complex microphases that have not been previously identified @xcite .
this risk is particularly significant in cases , as those under analysis , where dynamically ( e.g. in monte carlo simulations ) the domain walls separating different microphases are very long lived as the temperature is lowered @xcite . to develop a complete ab initio theory of pattern formations
it is necessary to be able to first _ prove _ periodicity of the ground state and then proceed with a variational computation within the given ansatz .
the problem is not simple .
most of the mathematically rigorous techniques developed for obtaining the low temperature phase diagram of spin systems , e.g. , the pirogov
sinai theory @xcite , depend on the interaction being short range . only methods based on reflection positivity @xcite or on convexity @xcite
seem applicable to the kind of potentials considered here .
in this paper we review recent progress on the construction of periodic ground states in one and two dimensional spin systems , both in the case of discrete and continuum systems @xcite .
the analysis is based on reflection positivity methods @xcite . in the cases where it applies
, it gives a full justification of the variational calculations based on the periodicity assumption .
the paper is organized as follows : in the next section we describe the class of discrete and continuum spin models we will be concerned with ; next , we present our main results about the zero temperature phase diagram of these models ; then we give a sketch of the proof and , finally , we draw our conclusions .
the simplest discrete spin model describing the class of systems we are interested in is an ising model with the following hamiltonian : h^(1)_=-j _ < , > _ _
+ k_,[2.1]where : @xmath2 is a @xmath3-dimensional cubic box with ( say ) open boundary conditions ; the first summation is over all pairs of nearest neighbor sites in @xmath4 , while the second summation is over pairs of distinct sites of @xmath4 ; if @xmath5 , @xmath6 is the ising spin variable ; @xmath7 , in order to guarantee the summability of the long range potential .
a physically interesting case is the one corresponding to @xmath8 and @xmath9 , in which case the long range term mimics the effect of the 3d dipolar interactions among out - of - plane magnetic moments constrained to a two dimensional surface as in a thin magnetic film .
the ground states of ( [ 2.1 ] ) are well understood in two limiting cases : if @xmath10 , the ground state is ferromagnetic , consisting of spins all aligned either up or down ; if @xmath11 , then the ground state is antiferromagnetic , displaying nel order of period 2 . for general @xmath12 and @xmath13
the ground state is not known .
a variational calculation supports the conjecture that in @xmath8 , @xmath9 , the ground state is periodic and striped for all @xmath14 with @xmath15 larger than a critical value @xmath16 .
in particular , it is remarkable that the checkerboard state has higher energy than the striped one .
another natural class of discrete spin models is described by the following hamiltonian : h^(2)_= - j _ , _ _ + k _ _ ^dip ( - ) _ , [ 2.1a]where @xmath17 are classical heisenberg spins and w^dip_ij ( ) = -_i _ j = ( _ ij - 3 ) .[2.1b]also in this case , in the limiting cases @xmath11 or @xmath10 the ground state can be determined exactly : it is ferromagnetic if @xmath10 ; it displays an in - plane uniaxial af order and is continuously degenerate if @xmath11 . in the general case
the ground state is unknown . on the mathematical side ,
the problem of _ proving _ periodicity is very difficult @xcite .
there are very few models where one can demonstrate rigorously that energy minimizers are periodic .
note that the problem is not trivial already in 1d .
it is sometimes convenient to consider effective continuum descriptions of the same spin system .
if @xmath18 @xmath19 and @xmath20 is the region where @xmath21 , one can consider the following model : h^(3)_= 2j | x_+|+k _ d*x*d*y * , p > d.[2.2]the case of `` soft '' local magnetization @xmath22 is often considered , too : h^(4)_= _ dx + k _ d * x*d * y * .[2.3]in the two equations above @xmath23 is an ultraviolet regulator , playing the role of the lattice spacing in the previous discrete model . in the case of the soft spin model
, @xmath24 is a symmetric double well potential , with two minima located at @xmath25 : typical examples to keep in mind are @xmath26 or @xmath27 or f()=\{. [ 2.4]where @xmath28 .
the gradient term in ( [ 2.3 ] ) represents the cost of a transition between two phases ( the homogeneous magnetized phases induced by the short range exchange interaction ) , while the term @xmath24 represents the local free energy density of a homogeneous system in a mean field approximation .
similarly to the discrete case , the minimizers of ( [ 2.2 ] ) and ( [ 2.3 ] ) can be easily determined in the limiting cases where only the attractive or only the repulsive interactions are present . in the presence of a competition , it is again conjectured , on the basis of variational computations , that the minimizers are periodic and that they display periodic striped ( or lamellar ) order @xcite .
both the models in this and in the previous section can be extended to the case of non zero magnetic field , in which case the expected phase diagram is even more complex , e.g. , periodic bubbled states are expected at high enough magnetic field @xcite .
the rigorous results concerning these more complicated cases are still very partial , and we will not consider them explictly in this paper .
we have proven a number of results in the 1d case , both for the discrete and the continuum models .
the higher dimensional case is in many respects still open .
partial results include an example of a 2d dipole system with in - plane dipoles , reminiscent of model ( [ 2.1a ] ) , that , in the presence of a special nearest neighbor ferromagentic interaction , has periodic ground states displaying striped periodic order . in the following we want to describe these results , starting from the case of one dimension . in one spatial dimension , we have a quite complete picture of the ground states of models @xmath29 , @xmath30 and @xmath31 . regarding the models with hamiltonian @xmath29 and @xmath30 , our main result can be summarized as follows @xcite .
+ + * theorem 1 . *
_ let @xmath32 .
for any @xmath33 and any @xmath34 , the specific ground state energies @xmath35 and @xmath36 of , respectively , @xmath37 and @xmath30 in the thermodynamic limit are given by : & & _
|| e_0^(1 ) ( ) = _ he_1(h ) , + & & _
|| e_0^(3 ) ( ) = _ h_+e_3(h ) , where @xmath38 and @xmath39 are the specific energies of a striped periodic configuration of period @xmath40 , obtained by extending periodically over the whole line the functions @xmath41 , @xmath42 , and @xmath43 , @xmath44 $ ] , respectively . in the presence of periodic boundary conditions ,
the only ground states of @xmath30 are the optimal periodic striped configurations ; in the case of the discrete ising model , the same conclusion is valid , if the ring has a length divisible by the optimal period . _ + * remarks . *
+ 1 ) the optimal period @xmath40 increases with @xmath12 ; when @xmath11 , @xmath45 in the discrete and @xmath46 in the continuum , i.e. , as @xmath47 the oscillations become wild and the period shrinks to zero . depending on the values of @xmath48 , @xmath12 and @xmath13 , the ground state has finite periodicity or is ferromagnetic : if @xmath49 and @xmath15 smaller than a suitable @xmath50 the period is finite and diverges at @xmath51 ; if @xmath52 the ground state is fm .
if @xmath53 , then @xmath54 and the period will increase as @xmath15 increases , becoming _ mesoscopic / macroscopic_. + 2 ) the proofs are based on a generalized notion of reflection positivity ( rp ) , see @xcite and below for the description of reflection positivity . in the present case one needs first to describe the states as collections of blocks , and then apply rp to the effective model of interacting blocks .
+ 3 ) the proof works equally well for a larger class of long range potentials @xmath55 : it is enough that @xmath55 is the laplace transform of a positive measure , i.e. , @xmath56 .
this class of interactions include , besides the power laws , the exponentials and arbitrary positive linear combinations of exponentials .
+ 4 ) inclusion of temperature is not trivial : block rp is lost for @xmath57 .
one expects for @xmath34 a gibbs measure , whose typical configurations are close to the periodic ground states determined above .
this is `` easy '' to prove for @xmath11 and @xmath58 and is probably not true for @xmath59 when the potential is not absolutely summable . (: for purely _
ferromagnetic _ power law potentials @xmath60
, there is a finite temperature spontaneous magnetization for @xmath61 ) .
+ 5 ) as mentioned above , we are unable at present to extend our results to the case where there is a magnetic field acting on the system or there is a non zero specified magnetization .
this is true both for the lattice , where very complex structures are expected ( e.g. , for @xmath11 the existence of a `` devil s staircase '' has been proved @xcite ) , and for the continuum where one expects simple periodic structures with blocks of alternate lengths @xcite .
+ the proof of theorem 1 works exactly in the same way both for @xmath62 and for @xmath63 .
the extension to @xmath64 requires a refinement of the block rp ideas , see @xcite .
the final result , stated in an informal way , is the following ( see @xcite for a mathematically rigorous statement , making precise the choice of boundary conditions and the notion of infinite volume minimizers ) .
+ + * theorem 2 . *
_ if @xmath32 , all the minimizers of @xmath64 are either _ simply periodic _ , of finite period @xmath65 , with zero average , or of constant sign
( and are constant if @xmath24 is convex on @xmath66 ) . by `` simply periodic '' we mean that within a period the minimizer has only one positive and one negative region , with the negative part obtained by a reflection from the positive part
_ + * remarks . *
+ 1 ) similarly to the `` hard '' spin case , the proof works even if the power law potential is replaced by a @xmath55 that is the laplace transform of a positive measure , in particular in the case of an exponential interaction . of course , as in theorem 1 , the value of the period and , in this case , the shape of the minimizer within one period , is obtained by a variational computation , whose result depends on @xmath55 , @xmath12 and @xmath13 .
+ 2 ) one can provide explicit examples of cases where the system displays a transition from a homogeneous to a periodic non homogeneous state .
for instance , if @xmath67 and @xmath68 , the previous result combined with an explicit computation imply that the minimizer is constant for @xmath69 and periodic with zero average for @xmath70 .
+ 3 ) the result of theorem 2 can be extended to a larger class of free energy functionals .
in particular , it can be applied to the study of effective 1d models for martensitic phase transitions , see @xcite . as mentioned above ,
the case of two or more dimensions is in many respects still open .
partial results include : + 1 ) lower bounds on the specific ground state energy of ( [ 2.1 ] ) and
( [ 2.2 ] ) , agreeing at lowest order with the energy of the striped case ; + 2 ) an example of a 2d dipole system with in - plane dipoles and a special exchange interaction , whose ground states can be proved to be striped and periodic .
+ + for the precise statement and the proof of claim ( 1 ) , we refer to @xcite ( let us just mention that the proof is based on apriori bounds on the energy of peierls contours ) . here
we want to describe in more detail the result mentioned in item ( 2 ) , which is , as far as we know , the only example of a spin model in two or more dimensions with real dipole interactions , for which existence of periodic striped order has been proved . the two dimensional spin model that we consider is a modification of model @xmath71 , defined by the following hamiltonian : h^(2 ) _ = _ _ ^dip ( - ) _ - _ , ( j _ x _ y + ( _ x _ y)^2 ) , [ 2d]where @xmath4 is a 2d square lattice , @xmath72 and @xmath73 are in - plane spins , whose allowed directions are only @xmath74 , @xmath75 , @xmath76 and @xmath77 . note that the @xmath78 term has the effect of encouraging alignment or antialignment but this term alone can not create periodic order .
our main result can be summarized as follows . + + * theorem 3 .
* _ let @xmath8 . for @xmath79 and @xmath78 large enough , the specific ground state energy of @xmath80 in the thermodynamic limit is given by : _ || e^(2)_0 ( ) = _ h + e_2(h ) , where @xmath81 is the specific energy of a striped configuration of period @xmath40 , consisting of stripes of uniformly polarized columns , all of size @xmath82 , and with alternate up / down polarization .
on a torus of side divisible by the optimal period , the only ground states are the optimal periodic striped configurations , either displaced vertically or horizontally . _ + + * remarks .
* + 1 ) the condition on @xmath78 is not uniform in @xmath12 .
it is unclear whether the same result should be valid for large @xmath78 , uniformly in @xmath12 , or even up to @xmath83 .
+ 2 ) the proof of theorem 3 is based on an extension of the ideas of the proof of theorem 1 .
we first show that , for @xmath84 , the system preferes to have the columns all completely polarized ( because of the anisotropy of the dipolar potential ) ; this means that , for the purpose of computing the ground state energy , we can restrict to 1d confirgurations of up or down columns , and at this point we can apply theorem 1 .
next we show that if @xmath85 is sufficiently large , then in the ground state there are no perpendicular neighboring spins , and this concludes the proof .
see @xcite for details .
as mentioned in previous sections , the proofs of theorem 2 and 3 extend the ideas of the proof of theorem 1 , which is based on a generalized notion of reflection positivity .
let us clarify here what we mean by reflection positivity , and how can one apply it to the problem of determining the ground state of @xmath29 , at least in the simple case @xmath11 .
let us consider the hamiltonian h_=_-n < x < yn_x_y v(y - x),in the presence of periodic boundary conditions and with @xmath86 , we can rewrite : v(x)=_0^ ( e^-x + e^-(2n - x)).therefore , h_()=h_l()+h_r ( ) + _ x =- n+1 ^ 0_y=1^n _ 0^d ( ) _ x_y ( e^-(y - x)+e^-(2n - y+x ) ) , [ 3.1]where : @xmath87 , @xmath88 and & & h_l()=_-n < x < x<0 _ x_x
v(x-x ) , + & & h_r()=_0y < yn _ y_y v(y-y).if @xmath89 is the configuration with @xmath90 and @xmath91 is defined as a_()=_y=0^n_y e^-y , we can rewrite ( [ 3.1 ] ) in the form : h_()=h_l()+h_r ( ) -_0^d ( ) e^,[3.2]now
the crucial remark is that the integral in the right hand side of ( [ 3.2 ] ) defines a scalar product between the configurations @xmath92 and @xmath89 .
therefore , defining = _ 0^d ( ) e^and using that , for any scalar product , 12 ( + ) , we get h_()12(h_l()-+h_r ( ) ) + 12(h_l()- + h_r()),which means h_()12 ( h_(_l)+h_(_r ) ) , with @xmath93 and @xmath94 . in other words , given any configuration @xmath92 , at least one of the two configurations obtained from @xmath92 by reflection around @xmath95 has better ( or equal ) energy than the original configuration .
so , if we want to reduce the energy , we can keep reflecting about all possible bonds ; proceeding like this we end up with the configuration @xmath96 in the presence of a nearest neighbor ferromagnetic interaction , we proceed in a similar fation , but we only reflect about the bonds separating a plus from a minus spin . after repeated refelctions we are left with a configuration consisting of a sequence of blocks of polarized spins , all of the same size and with alternate polarization . for more details , see @xcite ( see also @xcite and @xcite for a corrected discussion about the checkerboard estimate ) .
in this note we reviewed some recent rigorous results about existence of periodic striped states for a number of 1d and 2d spin systems , described by discrete or continuum models .
the proofs are based on a generalized notion of reflection positivity , and require the long range interaction to be _ reflection positive _ , i.e. , the laplace transform of a positive measure .
the short range interaction needs to be among nearest neighbor sites .
the proof applies to 1d systems or higher dimensional systems which can be proven to display 1d ground states by apriori energy estimates .
open problems include the inclusion of magnetic fields , of a non zero temperature and , most importantly , the proof that the ground states of the @xmath3dimensional spin systems described by eqs .
( [ 2.1 ] ) , ( [ 2.1a ] ) , ( [ 2.2 ] ) and ( [ 2.3 ] ) are translational invariant in @xmath97 coordinate directions .
so far , this claim has only been proven for @xmath98 in ( [ 2d ] ) .
there are good hopes to extend the proof to a new class of 2d models , relevant for the description of martensitic phase transitions @xcite .
let us also mention that these models with competing interactions are closely related to a class of systems considered by lebowitz and penrose in @xcite .
they considered the case when the pair potential is the sum of a short range interaction , @xmath99 , favoring phase segregation on the macroscopic scale and a long range ( kac type ) interaction favoring a uniform density , e.g. @xmath100 , @xmath3 the space dimension .
lebowitz and penrose proved that , in the limit @xmath101 , this competition will result in the system breaking up into a `` foam '' consisting of mesoscopic regions of the different phases .
these will have a characteristic length large compared to that of the short range potential and small compared to that of the long range potential @xmath102 .
this may give rise to the different kinds of patterns observed experimentally in many systems .
we have shown in @xcite that in one dimension when the long range kac potential is of the exponential type , or more generally is reflection - positive , then these droplets will form a periodic ground states with period of order @xmath103 . a heuristic argument suggests that the scaling of the patterns in three dimensions will be like @xmath104 @xcite .
as pointed out in @xcite the effect of the long range repulsive interaction is similar to that of surfactants which lower the surface tension between the two phases .
we plan to explore further the general phenomena of pattern formation due to competing interactions which occurs in many systems beyond those described earlier , e.g. langmuir monolayers , lipid monolayers , liquid crystals , two dimensional electron gases , diblock copolymers , etc .
this type of competition may also be responsible for some of the phenomona observed in confined water @xcite and in aqueous surfactant solutions @xcite .
a related kind of competition , due to geometric frustration , can induce the formation of striped periodic patterns in buckled collodial monolayers @xcite .
as pointed out in @xcite , this is related to stripe formation in compressible antiferromagnents on a triangular lattice @xcite .
the work of jll was supported by nsf grant dmr-044 - 2066 and by afosr grant af - fa 9550 - 04 - 4 - 22910 .
the work of ag and ehl was partially supported by u.s
. national science foundation grant phy-0652854 .
jll would like to thank pedro garrido and joaquin marro for their hospitality at their conference in granada .
a. desimone , r. v. kohn , f. otto and s. mller , in _ the science of hysteresis ii : physical modeling , micromagnetics , and magnetization dynamics _ ,
g. bertotti and i. mayergoyz eds . , pp . 269381 , elsevier ( 2001 ) . | there is a growing interest , inspired by advances in technology , in the low temperature physics of thin films .
these quasi-2d systems show a wide range of ordering effects including formation of striped states , reorientation transitions , bubble formation in strong magnetic fields , etc .
the origins of these phenomena are , in many cases , traced to competition between short ranged exchange ferromagnetic interactions , favoring a homogeneous ordered state , and the long ranged dipole - dipole interaction , which opposes such ordering on the scale of the whole sample .
the present theoretical understanding of these phenomena is based on a combination of variational methods and a variety of approximations , e.g. , mean - field and spin - wave theory .
the comparison between the predictions of these approximate methods and the results of montecarlo simulations are often difficult because of the slow relaxation dynamics associated with the long - range nature of the dipole - dipole interactions . in this note
we will review recent work where we prove existence of periodic structures in some lattice and continuum model systems with competing interactions .
the continuum models have also been used to describe micromagnets , diblock polymers , etc .
address = dip.to di matematica , universit di roma tre , l.go s. l. murialdo 1 , 00146 roma , italy address = department of mathematics and physics , rutgers university , piscataway , nj 08854 usa address = departments of mathematics and physics , princeton university , princeton , nj 08544 usa |
molecular oxygen , , is a dominant component of earth s atmosphere ( 21% by volume ) .
because it is a byproduct of photosynthesis ( and also a reactant in cellular respiration ) , it is considered as a potential marker for biological activity on terrestrial - like exoplanets ( e.g. , * ? ? ?
atomic oxygen is the third most abundant element in the universe ( following h and he ) ; however , it is still unknown what fraction of oxygen is contained within the deceptively simple in interstellar and circumstellar material .
gas - phase has recently been observed in - situ in the coma of comet 67p / churyumov - gerasimenko ( * ? ? ?
* hereinafter comet 67p / c - g ) by the rosina instrument on board the _ rosetta _ spacecraft ( rosetta orbiter spectrometer for ion and neutral analysis , * ? ? ?
is strongly correlated with and is present at an average level of @xmath5% relative to , making it the fourth most abundant molecule in the comet , following , , and .
the authors argue that does not originate from gas - phase chemistry in the coma but from direct sublimation from or within the comet surface .
the strong correlation with suggests that the is trapped within the bulk ice matrix of the comet , which provides constraints concerning the chemical origin of the ice .
processing of the cometary surface by solar wind particles and uv radiation has been ruled out by the authors , because the penetration depth ( a few @xmath6 m to m ) is not sufficient to process material throughout the bulk .
this process has been postulated to be responsible for the -rich , yet tenuous , atmospheres of several of the icy moons of saturn and jupiter ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
upon each pass into the inner solar system , comet 67p / c - g loses several meters of surface ice ; hence , the surface revealed today is likely pristine .
a reanalysis of data from the neutral mass spectrometer on board the _ giotto _ probe which did a fly - by of comet 1p / halley in 1986 , confirmed the presence of at a level similar to that seen in 67p / c - g @xcite .
this suggests that is not only an abundant molecule in comets , but is also common to both jupiter - family comets , such as 67p / c - g , and oort cloud comets , such as 1p / halley , which have different dynamical behaviours and histories .
the 67p / c - g observations strongly suggest that was present within the ice mantle on dust grains in the presolar nebula prior to comet formation .
this then raises the question whether was abundant in icy dust mantles entering the protoplanetary disk of the young sun , or whether the conditions in the comet - forming zone of the early solar system were favourable for formation and survival .
upper limits on the abundance of ice in molecular clouds obtained with the _ infrared space observatory _ ( iso ) and ground - based instruments are rather conservative ( / @xmath7 , * ? ? ?
* ; * ? ? ?
* ) . is a diatomic homonuclear molecule with zero electric dipole moment ; hence it does not possess electric dipole - allowed rotational transitions which makes it difficult to detect in cold environments via remote sensing .
therefore , gas - phase has been particularly elusive in interstellar clouds , early attempts to detect gas - phase in molecular clouds with the _ submillimeter wave astronomy satellite _ ( swas ) and _ odin _ resulted in upper limits only , @xmath8 relative to @xcite .
more recent and higher sensitivity observations with _ herschel _ allowed a deep search for towards sources considered true solar system progenitors : low - mass protostars .
a deep upper limit was determined towards the well - studied protostar , ngc 1333-iras 4a , ( / @xmath9 , * ? ? ?
detailed modelling of the chemistry throughout the well - characterised envelope of iras 4a demonstrates that the material entering the protoplanetary disk , both gas and ice , is likely poor in molecular oxygen . for a / abundance of @xmath10 , the inferred limit would correspond to a / abundance ratio of @xmath11 % .
this picture is consistent with laboratory experiments that have shown that ice is efficiently hydrogenated at low temperatures and converted into and ices ( @xmath12 k , * ? ? ?
* ; * ? ? ?
this makes the close association of with in 67p / c - g an even stronger enigma .
however , _
herschel _ did reveal the presence of gas - phase in two sources : orion ( @xmath13 , * ? ? ?
* ; * ? ? ?
* ) and @xmath14 oph a ( @xmath15 , * ? ? ?
* ; * ? ? ?
orion is a region of active star formation and the location of the gas - phase emission coincides with a clump of very warm ( @xmath16
k ) and dense gas , a so - called ` hot spot ' , which may have recently been subjected to shocks ( e.g. , * ? ? ?
these conditions are not representative of those expected in the molecular cloud from which the sun formed . on the other hand , @xmath14 oph a is a dense core in the more quiescent @xmath14 oph molecular cloud complex , which stands out from other low - mass star - forming regions by exhibiting emission from relatively warm molecular gas ( @xmath17 k , * ? ? ?
* ; * ? ? ?
subsequent observations of @xmath14 oph a have also determined the presence of related gas - phase species , and , at an abundance level on the order of @xmath18 that of @xcite .
these molecular ratios show reasonable agreement with those seen in 67p / c - g with rosina ( @xmath19 , @xmath20 , * ? ? ?
the chemically related species , ( ozone ) , was not detected in the comet coma with a very low upper limit , @xmath21 with respect to . in summary , despite being a particularly elusive molecule in interstellar and circumstellar environments , there apparently do exist conditions which are favourable for the formation of and related species at abundance ratios similar to that observed in ices in comet 67p / c - g . by assuming that all the energy deposited into water ice by high energy particles is used to convert into , @xcite claimed that radiolysis of water - containing interstellar ices in molecular clouds is the only mechanism that produces in high abundances .
however , laboratory experiments of cold interstellar ice analogs show that can also be efficiently formed through non - energetic surface chemistry before being converted to water ( see * ? ? ?
* ) while the production of through water radiolysis should be accompanied by a more efficient production of , in contradiction with the low abundance of observed in 67p / c - g . here
we investigate the formation and survival of ice using a variety of sophisticated astrochemical models , taking an extended chemical network including the formation and destruction pathways of into account , in order to elucidate the origin of cometary , and help explain its strong correlation with water ice and the low abundances of its chemically related species .
we explore and discuss several different origins : i ) synthesis in ice mantles in dark clouds ( `` primordial '' origin ) , ii ) formation and survival _ en route _ from the protostellar envelope into the disk and subsequent delivery into the comet - forming zone , and ( iii ) _ in - situ _ formation of within the protoplanetary disk prior to comet formation .
this work differs from that presented in @xcite because we consider all possible chemical pathways between and other o - bearing species , including , , , and . in section
[ chemistry ] we describe the interstellar chemistry of molecular oxygen , in sections 3 to 5 we systematically discuss each scenario , presenting the necessary evidence for or against each hypothesis , and in section [ summary ] we summarise our main findings .
two main processes have been invoked for the formation of molecular oxygen in the interstellar medium : i ) gas - phase formation via neutral - neutral chemistry , and ii ) formation via association reactions on / within icy mantles of dust grains .
the observations towards both @xmath14 oph a and 67p / c - g , in conjunction with known chemical pathways studied in the laboratory , present several challenges for astrochemical models .
first , the reproduction of the relatively high / ice ratio simultaneously with the very low / ice ratio , and second , the ratios of / and / produced in the gas phase , assuming that chemistry on or within the ice mantle is responsible for the observed gas - phase ratios .
figure [ scheme_o2 ] summarises the chemical reactions involved in the formation and destruction of molecular oxygen which are discussed here .
gaseous is thought to form primarily via the barrierless neutral - neutral reaction between o and oh in cold and warm gas . due to its importance , this reaction has been well studied both experimentally and theoretically .
the rate coefficient has a negligible temperature dependence , with a recommended value ( based on theoretical calculations and experiments ) between @xmath23 and @xmath24 @xmath25 s@xmath4 at 10 k , and an experimentally - constrained value of @xmath26 @xmath25 s@xmath4 at 140 k decreasing to @xmath27 @xmath25 s@xmath4 at 300 k ( see * ? ? ?
* for a discussion on the rate coefficient ) .
the formation of in cold dark clouds is initiated by the high initial abundance assumed for atomic oxygen , inducing an efficient ion - neutral chemistry that also forms oh . in warm environments ( @xmath28 k ) , e.g. , the inner regions of protostellar envelopes or the inner , warm layers of protoplanetary disks , oh and o are mostly produced through warm neutral - neutral chemistry driven by the photodissociation of water sublimated from interstellar ices .
the gas - phase formation of the chemically - related species , , is inefficient under interstellar conditions , as it requires three - body association of + @xcite ; thus , despite this reaction possessing a negligible reaction barrier , it only proceeds under the high - density conditions found in planetary atmospheres and in the inner midplanes of protoplanetary disks .
solid in dark clouds is involved in the surface chemistry reaction network leading to the formation of water ice @xcite .
is formed through atomic o recombination on ices and efficiently reacts with either atomic o or atomic h to form or , respectively , eventually leading to the formation of water .
the hydrogenation of also leads to the formation of , in addition to dominating the destruction of ice .
laboratory experiments of interstellar ice analogues studying water formation suggest that the o + o , o + , and h + reactions , involved in the formation and destruction of , all have small or negligible reaction barriers . @xcite and @xcite independently studied the efficiency of the + h reaction , with both studies concluding that this reaction is effectively barrierless , contradicting the earlier quantum calculations by @xcite for the gas - phase reaction which predicted an activation barrier of 1200 k. the reactivity of the o + o and o + reactions is still a matter of debate , and
is discussed in section [ imp_params ] .
the reaction , @xmath29 , is considered unlikely to occur on grain surfaces under dark cloud conditions because of its relatively high activation energy barrier , 2000 k , as experimentally determined for the gas - phase reaction @xcite .
dark clouds , the inner regions of protostellar envelopes , and the comet - forming regions of protoplanetary disks , are all well - shielded from external sources of uv radiation ( @xmath30 mag ) ; however , water ice can be photodissociated by cosmic - ray - induced uv photons produced by the excitation of molecular hydrogen by electrons generated by cosmic - ray ionisation of @xcite . water ice photodissociation has been extensively studied in the laboratory @xcite and in molecular dynamics ( md ) simulations @xcite .
the md simulations show that water ice which is photodissociated generates oh and h photoproducts that move through the ice due to their excess energy . each photodissociation event can lead to various chemical outcomes ( e.g. , direct desorption into the gas phase or recombination followed by desorption or trapping ) , the probabilities for which are dependent upon the depth into the ice mantle ( and fully tabulated in * ? ? ?
the detection of following the uv irradiation of cold water ice also supports water ice photodissociation into o + or o + h + h photoproducts @xcite .
laboratory experiments show that the bombardment of cold water ices with ionizing energetic particles can result in the formation of and other chemically related species from the destruction of water ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the production of and through irradiation of water ice by energetic particles depends on the projectile penetration depth .
low - energy ions , for example , only penetrate the few dozen outermost ice layers , where h and can easily escape , favouring an efficient production of relative to .
the yield of o@xmath22 production therefore tends to decrease with the energy of the irradiating particles from a few @xmath31 molecule ev@xmath4 for kev protons to @xmath32 for mev ions ( see * ? ? ?
irradiation of energetic ions during the condensation of water molecules can dramatically enhance the production of up to / abundances ratios of @xmath33 % @xcite .
o@xmath22 formed in the ice mantle under dark cloud conditions ( @xmath34 k ) can be returned to the gas - phase via a multitude of non - thermal desorption processes ( e.g. , * ? ? ?
those mechanisms which have been quantified in the laboratory for include photodesorption by cosmic - ray - induced uv photons @xcite , and desorption induced by exothermic chemical reactions ( i.e. , chemical desorption , * ? ? ?
* ; * ? ? ?
photodesorption of was found to be triggered by photodissociation , with returned to the gas - phase with yields of @xmath35 molecules per incident photon , for a radiation spectrum appropriate for the cosmic - ray - induced uv field and pure ice @xcite .
is also detected in the experiments by @xcite with yields a factor of a few lower than those for .
is not seen in the experiments by @xcite due to the lower fuv fluences in the synchrotron experiments .
the probability of chemical desorption depends strongly on the type of reaction and on the substrate and can vary between 0 and 80% . the chemical desorption efficiency of the o + o reaction was found to be @xmath36% in experiments of formation via oxygen recombination on bare olivine - type surfaces @xcite .
however , in experiments with higher oxygen coverage , the efficiency was reduced to an estimated upper limit of @xmath37% probably due to an efficient dissipation of the energy released by the exothermic reaction into the water ice @xcite .
the standard chemical desorption efficiencies assumed in this work are the theoretical values computed by @xcite for the submonolayer regime on bare grains .
however , they should be regarded as upper limits .
the o + o reaction has a high theoretical probability of 68% while reactions + h , + h , and + h show much lower theoretical chemical desorption probabilities of 0.5 - 2 % , in agreement with the experimental upper limits .
we explore in section [ rhoopha ] the impact of the chemical desorption efficiencies on the gas phase abundances of and its chemically related species .
when data are not available , the chemical desorption probability is fixed to 1.2% @xcite .
the binding energies of to a variety of surfaces , including dust - grain analogues and water ice , have been measured in the laboratory ( @xmath38 k , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this low binding energy makes a particularly volatile species , expected to desorb at temperatures similar to co. in temperature - programmed desorption ( tpd ) experiments with layered on top of , and fully mixed with , water ice , a fraction of is found to remain trapped within the ice matrix and released at higher temperatures @xcite .
the trapped fraction depends upon the deposition temperature with a greater fraction of volatiles trapped within the water ice when deposited at lower temperatures @xcite .
the formation and survival in dark clouds and protoplanetary disks depends on a number of parameters , which are linked in turn to various physical and chemical conditions : + 1 ) the gas phase abundance ratio between h and o atoms that accrete onto grains governs the competition between hydrogenation reactions leading to and and association reactions between o atoms , forming or @xcite . for dark cloud conditions ,
the atomic h abundance in the gas phase is a balance between its formation , which occurs via ionisation followed by dissociative electron recombination , and its conversion back into via recombination reactions on grain surfaces . at steady state and assuming a sticking probability of 1 ,
the density of h is therefore given by the ratio between these two processes @xcite @xmath39 where @xmath40 is the thermal velocity of atomic hydrogen , @xmath41 the cosmic ray ionisation rate , and @xmath42 and @xmath43 the abundance and the cross section of interstellar grains .
the absolute number density of atomic h is therefore independent of the total density and increases linearly with the cosmic ray ionisation rate . since the initial number density of atomic o increases linearly with the total number density for a fixed oxygen abundance , the atomic h / o abundance ratio increases ( decreases ) linearly with the total density ( cosmic - ray ionisation rate ) .
\2 ) the surface mobility of o atoms governs the reactivity of the o + o and o + reactions .
the surface mobility of o atoms occurs mostly through thermal hopping and depends exponentially on the dust temperature @xmath44 , and their diffusion energy @xmath45 .
astrochemical models which treat grain - surface chemistry usually scale the diffusion energy to the binding energy of the considered species @xmath46 , using a fixed value for the diffusion - to - binding energy ratio @xmath47 ( e.g. , * ? ? ?
as discussed by several authors @xcite , @xmath48 and @xmath47 strongly depend upon the ice morphology and composition .
the mobility of atomic oxygen on interstellar ice analogues has recently been investigated by several experimental groups @xcite who conclude that atomic o has a higher binding energy than the value of 800 k estimated by @xcite .
theoretical calculations and experiments studying the diffusion of molecules ( co or ) or heavy atoms ( o ) on several types of substrates suggest that species diffuse with low diffusion - to - binding energy ratios of the order of @xmath49% @xcite .
however , experiments focusing on formation via h recombination on surfaces suggest a higher diffusion - to - binding energy ratio between 50 and 80% @xcite . the diffusion - to - binding energy ratio likely has a distribution of values that depend upon the substrate ( bare or ice - coated ) , the species under consideration ( light atom , heavy atom , molecule ) , the ice morphology ( porous , compact , crystalline , or amorphous ice ) , and the dominant composition of the chemically - active surface layer ( , , or ) .
\3 ) the activation barriers of the o + o and o + reactions directly govern the reactivity of the two reactions . @xcite
derive an upper limit of 150 k for the reaction barrier for o + o and o + in an experimental study on an amorphous silicate surface .
however , the presence of an activation barrier for the latter reaction has been invoked by several authors ( see * ? ? ?
for example , @xcite require an activation barrier of 500 k for the o + reaction in order to reproduce the results of laboratory experiments in thick ices with their microscopic monte - carlo model .
here we explore the effects of the parameter choices for these three key aspects of the chemistry .
three state - of - the - art gas - grain astrochemical models have been used in this work to study the formation and survival of molecular oxygen from dark clouds to the solar system : 1 ) the multi - phase model by @xcite to study the formation of in dark clouds ; 2 ) the multi - phase model by @xcite to study the formation of during the formation of protoplanetary disks ; 3 ) the two - phase model by @xcite to study the formation of in protoplanetary disks . the multi - phase gas - grain taquet and furuya models couple the gas phase and ice chemistries with the approach developed by @xcite to follow the multi - layer formation of interstellar ices and to determine the gas - ice balance .
several sets of differential equations governing the time - evolution of abundances are considered : one for gas - phase species , one for surface ice - mantle species , and one ( or several ) for bulk ice - mantle species .
the equations governing chemical abundances on the ice surface and in the bulk ice are linked by an additional term that is proportional to the rate of growth or loss of the grain mantle . as a consequence ,
surface species are continuously trapped in the bulk because of the accretion of new species in dark clouds . following @xcite , the chemically - active surface
is limited to the top four monolayers .
the bulk ice mantle is considered to be chemically inert .
the original three - phase model considered in the taquet model assumes that the inert bulk ice mantle has a uniform molecular composition . in order to accurately follow the ice evolution in warm conditions ,
the furuya model considers a depth - dependent molecular composition , through the division of the inert bulk ice mantle into five distinct phases ( for details , see * ? ? ?
* and references therein ) .
radiolysis , i.e. the bombardment of ( ionizing ) energetic particles depositing energy into the ice , and/or photolysis , i.e. the irradiation of ultraviolet photons breaking bonds , can trigger chemistry within the bulk mantle of cold interstellar ices .
we have investigated the impact of the uv photolysis induced by secondary uv - photons on the bulk ice chemistry and the formation and survival of by activating the bulk chemistry and assuming the same ice parameters as for the surface chemistry ( same diffusion and binding energies , same chemical reactions ) . in our model , the formation of from photodissociation is a multi - step process , starting from the production of oxygen atoms from water or oh photodissociation followed by their recombination .
we find that can not be efficiently produced in the bulk through ice photolysis as the photodissocation of the main ice components not only produces oxygen atoms , that recombine together to form , but also hydrogen atoms that react with to reform water even if ice photodissociation would go directly to rather than h since there are other molecules like , or that produce hydrogen atoms that are very mobile .
overall , activating the bulk chemistry decreases the abundance of highly reactive species like o atoms or radicals but does not affect the main ice species .
laboratory experiments show that can be efficiently formed through radiolysis of ices without overproducing only if the radiolysis occurs as water is condensing onto a surface ( * ? ?
* see section [ icechem ] ) .
however , in molecular clouds water ice is mostly formed _ in - situ _ at the surface of interstellar grains through surface reactions involving hydrogen and oxygen atoms .
this happens prior to the formation of the presolar nebula , i.e. the cloud out of which our solar system was formed , and it is possible that the comet - forming zone of the sun s protoplanetary disk inherited much of its water ice from the interstellar phase @xcite . for the radiolysis mechanism to occur in the presolar nebula , water ice would first need to be completely sublimated and then recondensed prior to comet formation .
luminosity outbursts induced by instabilities in the disk of the solar nebula can potentially provide a scenario for efficient production in the ice matrix through sudden evaporation of water ice followed by fast recondensation .
we consider this scenario less likely because the cosmic - ray ionisation rate is thought to be impeded near the disk midplane with respect to interstellar values ( e.g. , * ? ? ?
energetic ionizing particles from the ( pre)solar wind are also expected to be significantly attenuated close to the disk midplane by the intervening large column of material ( @xmath50 g @xmath51 ) between the central star and the comet - forming zone beyond @xmath52 au .
@xcite explored the formation through radiolysis of water within interstellar ices in the solar nebula to explain the high abundance of observed in comet 67p / c - g .
however , they concluded that the galactic cosmic - ray flux is not sufficient to produce the observed ratio of / over the lifetime of the presolar nebula .
the gas - phase chemical network used by the taquet model is the non - deuterated version of that from @xcite , the basis for which is the 2013 version of the kida chemical database @xcite .
it has been further updated to include warm gas - phase chemistry involving water and and ion - neutral reactions involving ozone .
the network also includes the surface chemistry of all dominant ice components ( , co , , , , , ) , as well as those important for water ( e.g. , , , and ) .
several new surface reactions were added involving and reactive species such as n , o , oh , , and , following the nist gas - phase chemical database .
the gas - ice chemical network of @xcite , based on the osu 2006 network , is used with the furuya model .
the gas phase and surface networks are more suited to the high density and warm temperatures conditions found in protostellar envelopes .
it has therefore been supplemented with high - temperature gas - phase reactions from @xcite and includes the formation of many complex organic molecules .
it is consequently more expansive than the network used in the taquet model .
the gas - phase chemical used in the walsh model is based on the 2012 release of the umist database for astrochemistry ( udfa ; * ? ? ?
* ) , supplemented by direct x - ray ionisation reactions , x - ray - induced ionisation and dissociation processes , and three - body reactions . the grain surface chemical network of @xcite is used .
input parameters assumed for the three types of astrochemical models are listed in table [ input_table ] .
unless otherwise stated , this table gives the standard values for the physical parameters : the cosmic ray ionisation rate @xmath41 , the flux of secondary uv photons ; the grain surface parameters : the dust - to - gass mass ratio @xmath53 , the grain diameter @xmath54 , the volumic mass of grains @xmath55 , the surface density @xmath56 , the diffusion - to - binding energy ratio @xmath57 , the number of chemically active monolayers @xmath58 , and the sticking coefficient of species heavier than h and .
the elemental abundances of species correspond to the set ea1 from @xcite . l c input parameters & values + + @xmath41 ( s@xmath4 ) & @xmath59 + @xmath60(sec .
uv ) ( @xmath51 s@xmath4 ) & @xmath61 + + @xmath53 & 0.01 + @xmath54 ( @xmath6 m ) & 0.2 + @xmath55 ( g @xmath2 ) & 3 + @xmath56 ( @xmath51 ) & @xmath62 + @xmath57 & 0.5 + @xmath58 ( mls ) & 4 + @xmath63 ( heavy species ) & 1 + + @xmath64 ( ) & 0.5 + @xmath64 ( ) & 0.09 + @xmath64 ( ) & @xmath65 + @xmath64 ( ) & @xmath66 + @xmath64 ( ) & @xmath67 + @xmath64 ( ) & @xmath68 + @xmath64 ( ) & @xmath69 + @xmath64 ( ) & @xmath70 + @xmath64 ( ) & @xmath71 + @xmath64 ( ) & @xmath72 + @xmath64 ( ) & @xmath73 + [ input_table ]
here we investigate whether the observed in 67p / c - g has a dark cloud origin , using the chemistry of ice and gas described in the previous section . for this purpose
, we use the taquet astrochemical model presented in section 2.4 .
the appendix presents a first parameter study , in which several surface and chemical parameters are varied , in order to reproduce the low abundances of the chemically related species , , and with respect to seen in comet 67p / c - g .
the low abundance of and relative to can be explained when a small activation barrier of @xmath74 300 k is introduced for the reactions o + and h + , in agreement with the monte - carlo modelling of @xcite .
however , the abundance of is still overproduced by one order of magnitude , suggesting that other chemical processes might be at work .
a second parameter - space study is then conducted to determine the range of physical conditions ( e.g. , dust temperature , number density , and cosmic - ray ionisation rate ) over which ice and gas ( and those for chemically - related species , , , and ) reach abundances ( relative to water ice ) similar to that seen in 67p / c - g .
finally , the case of @xmath14 oph a , where gas - phase has been detected in the gas phase , is revisited with the same chemical model .
the low temperature , in conjunction with the low flux of uv photons found in interstellar dark clouds , promotes the formation of interstellar ices .
the ice chemical composition depends on various physical and chemical parameters as discussed in section [ imp_params ] . to investigate the formation and survival of under dark cloud conditions
, a model grid is run in which the total density of h nuclei , @xmath75 , the gas and dust temperature , @xmath44 ( assumed to be equal ) , the cosmic ray ionisation rate , @xmath41 , and the visual extinction , @xmath76 are varied following the methodology described in @xcite .
values explored in the model grid are listed in table [ grid_table ] , resulting in 500 models in total . in these models ,
the `` standard '' set of chemical parameters derived in the appendix are assumed ( see table [ grid_table ] ) .
( see section [ modelgrid ] ) .
the thick dashed lines or the solid boxes refer to the abundances observed in the comet 67p / c - g.,title="fig : " ] ( see section [ modelgrid ] ) .
the thick dashed lines or the solid boxes refer to the abundances observed in the comet 67p / c - g.,title="fig : " ] the abundances of all species in the reaction network are evolved from their assumed initial abundances ( see section [ astrochem ] ) as a function of time only , i.e. , assuming constant physical conditions .
figure [ modelgrids1 ] shows the distribution of abundances of solid , and the chemically related species , , , and , relative to water ice , at the free - fall time , @xmath77 , defined as @xmath78 where @xmath79 is the gravitational constant and @xmath80 is the proton mass .
@xmath77 varies across the grid from @xmath81 to @xmath82 yr .
cores can have longer lifetimes , e. g. due to magnetic support , up to 10 @xmath77 .
however , assuming a longer timescale does not change our conclusions because interstellar ices form in a timescale similar to @xmath77 .
the results show that the formation and survival of solid , and other reactive species , in interstellar ices , is strongly dependent upon the assumed physical conditions .
the model grid shows a large dispersion of final abundances of solid from @xmath83 to @xmath84 relative to water ice ( top panel of figure [ modelgrids1 ] ) . due to its lower reactivity , hydrogen peroxide , ,
shows a slightly more narrow final abundance dispersion , with most of the models predicting values between @xmath32 and @xmath85 ( 1 % ) with respect to water ice ( see bottom panel of fig .
[ modelgrids1 ] ) .
is mostly formed in the ice mantle via the hydrogenation of , and is converted into via a subsequent barrierless hydrogenation reaction , being a precursor of ; hence , its final abundance is governed by that of ice , and therefore follows a similar trend but lower by four orders of magnitude due its high reactivity .
ozone , formed from molecular oxygen via the + o reaction also displays a broad distribution of abundances but most of the models predict abundances lower than @xmath32 relative to water , due to the small o + barrier .
figure [ modelgrids2 ] shows the distribution of the final abundance of solid relative to water ice , for the range of assumed values for each physical parameter varied in the model grid .
high abundances ( @xmath86% relative to water ice ) are obtained only for those models with high densities ( @xmath87 @xmath2 ) .
as discussed in section [ imp_params ] , higher gas densities result in a lower gas - phase h / o ratio , thereby increasing the rate of the association reaction between o atoms to form ice , and correspondingly decreasing the rate of the hydrogenation reactions , o + h and + h , which compete with ice formation , and destroy ice once formed , respectively .
an intermediate temperature of 20 k is also favoured because it enhances the mobility of oxygen atoms on the grain surfaces whilst at the same time allowing efficient sublimation of atomic h. this additionally enhances the rate of oxygen recombination forming , with respect to the competing hydrogenation reactions .
models with lower temperatures of 10 or 15 k can also reproduce the / of 4% if a high density of @xmath88 @xmath2 is considered . moreover , because the density of gas - phase h atoms increases linearly with the cosmic - ray ionisation rate , @xmath41 , a low value of @xmath41 also tends to favour the survival of ice . on the other hand
, the visual extinction does not have a strong impact on the abundance of solid as the distributions of abundances obtained for the five visual extinction values are very similar .
thus , the final ice abundances depend more strongly upon the assumed gas density , temperature , and cosmic - ray ionisation rate , and high ice abundances occur when the initial atomic h / o ratio is low ( @xmath89 ) . to illustrate further the crucial impact of the density and the cosmic ray ionisation rate on the chemical composition of ices , figure [ horatio ] shows the evolution of the abundances of and its chemically related species with respect to water ice as a function of the initial atomic h / o abundance ratio induced by a variation of the total density ( assuming a constant @xmath41 of @xmath59 s@xmath4 ) or a variation of the cosmic ray ionisation rate ( assuming @xmath90 @xmath2 ) at @xmath91 and @xmath92 k. according to equation ( [ eq_nh ] ) , the initial atomic h / o abundance ratio follows the expression @xmath93 assuming the grain parameter values listed in table [ input_table ] .
for each temperature case , the evolution of the abundance ratios with the initial atomic h / o abundance ratio follows similar trends , suggesting that the initial atomic h / o abundance ratio , and consequently the @xmath94/@xmath41 ratio , is the dominant parameter for the formation and survival of and its chemically related species in dark clouds .
the formation of ice is strongly inhibited ( @xmath95% ) for high initial h abundances ( [ h]/[o]@xmath96 ) induced by high cosmic - ray ionisation rates and/or low densities , as it increases the rate of conversion of ice to ice . for low cosmic - ray ionisation rates or high densities inducing initial h / o ratios lower than @xmath85 ,
the formation of h atoms in the gas phase is no longer dominated by ionisation followed by dissociative recombination but by neutral - neutral reactions involving o atoms .
the abundances of and other chemically - related species are consequently no longer influenced by @xmath41 nor @xmath97 and remain constant .
the results here demonstrate that a high abundance of , at a level similar to that measured in 67p / c - g , seems to require an initial h / o abundance ratio lower than @xmath98 ( depending on the temperature ) or , according to equation ( [ horatio_eq ] ) @xmath99 assuming the initial abundances listed in table [ input_table ] .
@xmath2 , dashed lines ) and different densities ( and assuming @xmath100 s@xmath4 , solid lines ) at @xmath91 k ( top ) and @xmath101 k ( bottom ) .
the `` standard '' values of other parameters , listed in table [ grid_table ] , are assumed .
the solid boxes refer to the abundances observed in comet 67p / c - g . ]
figure [ poplay ] shows the chemical composition of the ice obtained for the model using the physical conditions that best reproduce the observations in comet 67p / c - g ( @xmath102 @xmath2 , @xmath103 k , @xmath104 s@xmath4 ) , and the chemical parameters derived in the appendix .
the fractional composition in each ice monolayer is plotted as function of monolayer number , i.e. the ice thickness that grows with time . at such a high density
( @xmath105 @xmath2 ) , hydrogenation reactions are less efficient due to the lower relative abundance of atomic h , and the freezeout timescales are sufficiently fast that reactive species can be trapped in the ice mantle before conversion into more stable molecules , like .
the higher temperature ( 21 k ) also enhances the mobility of heavier species , such , to increase the relative abundance of ice species such as and . as a consequence ,
the most abundant species are water and carbon dioxide .
ice is mostly present in the innermost layers of the ice mantle and decreases in relative abundance towards the ice surface , reflecting the initial low ratio of h / o in the gas phase , but tends to be well mixed with ice .
in contrast , is mostly formed in the outer part of the ices , allowing an efficient sublimation , explaining its weak correlation with water in 67p / c - g .
the @xmath14 oph a core , located at a distance of 120 pc , constitutes the best test case for the water surface network and the production of in dark clouds because it is the only interstellar source so far where gas - phase , , and have been detected @xcite .
the parameter study presented in the previous section suggests that the physical conditions of @xmath14 oph a , a high density ( @xmath106 @xmath2 ) , and a relatively warm gas temperature ( @xmath107 k ) and dust temperature ( @xmath108 k ) , derived by @xcite are consistent with those which facilitate the formation and survival of ice .
, , , and are mostly , and potentially only , produced via surface chemistry ; hence their gas - phase abundances depend on their formation efficiency in interstellar ices and on the probability of desorption upon formation through chemical desorption ( which is the dominant non - thermal desorption mechanism for these species in dark cloud conditions ) . as explained in section
[ gasice ] , the chemical desorption probabilities assumed in this work are the theoretical values computed by @xcite and @xcite for more than 20 reactions involved in the water and methanol chemical networks and vary between 0 and 70% .
when data are not available , the chemical desorption probability is fixed to 1.2% @xcite .
figure [ rhoopha_gas ] shows the temporal evolution of the gas phase abundances of , , , and when the theoretical chemical desorption probabilities from @xcite , considered as our standard values , are assumed .
the high chemical desorption probability of the reaction o + o ( 68 % ) allows for an efficient evaporation of in the gas phase upon formation on ices , inducing maximal abundances of a few @xmath32 obtained at 8000 years . at longer timescales ,
the production on ices is limited and the gas phase abundance of decreases sharply in a few @xmath61 yr due to its efficient freeze - out induced at the high density @xmath109 @xmath2 .
the surface reactions + h and + h forming and have a lower chemical desorption probability of 1.4 and 0.5 % respectively . these values are nevertheless high enough to produce gaseous abundances of and larger than @xmath110 . as a consequence
, the model fails to simultaneously reproduce the gaseous abundances of , , and derived in @xmath14 oph a since the predicted and abundances are higher than the observations by one order of magnitude when the predicted abundance reaches the observed value of @xmath111 at a time of @xmath112 years . instead , their abundances are fit at a slightly longer time of @xmath113 years .
oph a physical conditions and the chemical parameters derived in the appendix . ]
@xcite also performed a comprehensive modelling of the gas - ice chemistry occurring for the physical conditions found in @xmath14 oph a by focusing on and .
their chemical network is similar to that used in this work but they used a two - phase model where the entire bulk ice is assumed to be chemically reactive , and adopted a high chemical desorption probability of 10 % for all surface reactions .
their model therefore predicts a high abundance of gaseous and , typically higher than @xmath110 for the first @xmath114 yr of their simulation , and finds good agreement with the observations , with abundances of @xmath115 , at @xmath116 , which is 10 times longer than the free - fall timescale expected at this density . at this timescale ,
the predicted abundance is one order of magnitude lower than the observed value of @xmath117 , a similar result as our standard model .
figure [ rhoopha_gas2 ] shows the gas phase abundances of , , and obtained when the predicted abundance of reaches the abundance observed toward @xmath14 oph a by decreasing the chemical desorption probability of all reactions with respect to their standard theoretical value .
the model using a normalized chemical desorption probability of 1 is the standard model .
it can be seen that the , , and abundances can be simultaneously reproduced when chemical desorption probabilities lower than the standard values by a factor of 500 are used , giving absolute values of @xmath74 0.001 % for the reactions h + and h + .
yr toward @xmath14 oph a when decreasing the chemical desorption probability of all reactions with respect to their standard theoretical value . ]
the models presented and discussed in the preceding section show that ( and chemically related species ) can be efficiently formed under dark cloud conditions , reaching abundance levels ( relative to water ice ) similar to that observed in comet 67p / c - g as long as the density is high , the ionisation rate is low , and the temperature is warm . here
, we discuss the role of chemistry during protostellar collapse and protoplanetary disk formation on the observed abundance of in 67p / c - g .
material en route from the protostellar envelope into the disk is subject to increasing temperatures and uv radiation generated by the central ( proto)star . we address the following two questions : ( i ) can gas and/or ice efficiently form during the formation of protostars and disks if the material composition is initially poor in molecular oxygen ? and ( ii ) can ice co - formed with ice in the prestellar stage be delivered to the comet - forming zone in young protoplanetary disks without significant chemical processing ? to address those questions , the chemical evolution from prestellar cores to forming disks is calculated . for the protostellar disk formation model , the axisymmetric semi - analytical two - dimensional model developed by @xcite and adjusted by @xcite is adopted .
briefly , the model describes the temporal evolution of the density and velocity fields following inside - out collapse and the formation of an accretion disk described by the @xmath118-viscosity prescription @xcite .
additional details can be found in the original papers .
the vertical structure of the disk is calculated assuming hydrostatic equilibrium .
the dust temperature and uv radiation field , which are critical for the chemistry , are calculated at each time step by solving the radiative transfer with radmc-3d .
outflow cavities are included by hand in a time - dependent manner ( see * ? ? ?
* for details ) .
initially , the core has a power - law density distribution @xmath119 , where @xmath120 is the distance from the center of the core , with an outer boundary of @xmath121 au and a total mass of 1 @xmath122 .
two values for the initial core rotation rate are investigated : @xmath123 s@xmath4 and @xmath124 s@xmath4 , corresponding to cases 3 and 7 in @xcite , respectively .
the model follows the physical evolution until the end of the main accretion phase when the gas accretion from the envelope onto the star - disk system is almost complete .
fluid parcels from the envelope to the disk are traced in the physical model , and the furuya astrochemical model is used to follow the gas - ice chemical evolution calculated along each individual trajectory with the parameters described in section [ astrochem ] .
a molecular cloud formation model is run to determine the composition of the gas and ice in the parent molecular cloud @xcite .
the chemistry is then evolved for an additional @xmath125 yr under prestellar core conditions to compute the abundances at the onset of collapse . the prestellar core density , temperature , and visual extinction are set to @xmath126 @xmath2 , 10 k , and 10 mag , respectively . at the onset of collapse ,
most oxygen ( @xmath127% ) is contained in icy molecules , e.g. , and co ice .
the gas and ice abundances with respect to hydrogen nuclei are only @xmath128 and @xmath129 , respectively , while the gas and ice abundances are @xmath130 and 10@xmath131 , respectively .
hence , the models using this set of initial abundances have a negligible ice abundance .
note that the gas abundance in both the molecular cloud formation stage and the prestellar core stage is lower than a few @xmath132 , which is consistent with the upper limits of the observationally derived gas abundance toward nearby cold ( @xmath133 k ) clouds @xcite .
following on from the preceding section , we also explore whether ice co - formed with ice in the prestellar stage can be delivered to the comet - forming midplanes of protoplanetary disks without significant alteration . to do this
, we also run models with an artificially increased initial ice abundance , set to be 5% of that for ice .
figure [ fig : disk_formation1 ] shows the spatial distributions of fluid parcels at the final time of the simulation in models with @xmath123 s@xmath4 ( infall dominated , top panels ) and @xmath124 s@xmath4 ( spread dominated , lower panels ) . for the case in which the ice mantle is poor in ice at the onset of collapse , it is found that ( i ) some gaseous can form ( up to @xmath7410@xmath134 ) depending on the trajectory paths ( left panels ) , and ( ii ) ice trapped within ice does not efficiently form en route into the disk ( middle panels ) .
given that most elemental oxygen is in ices ( and ) at the onset of collapse , gaseous forms through photodissociation / desorption of ice by stellar uv photons in the warm ( @xmath13520 k ) protostellar envelope , followed by subsequent gas - phase reactions ( e.g. , o + oh ) .
the middle panels of figure [ fig : disk_formation1 ] show that the majority of parcels in each disk have a low final / ice ratio , @xmath136 .
however , the upper layers of the larger ( i.e , higher @xmath137 case ) disk do have several parcels with a / ice ratio higher than @xmath85 ( see panel ( e ) in figure [ fig : disk_formation1 ] ) .
analysis of the ice composition shows that the ice is associated with ice rather than with . upon water ice photodissociation ,
the warm temperatures encountered through the protostellar envelope mean that ice ( re)formation is more favorable than that for ice .
this is due to the weak binding energy of atomic hydrogen : the reaction to form ice ( via , e.g. , co + oh ) proceeds faster that that for reformation ( e.g. , h + oh ) as atomic hydrogen escapes back into the gas phase before it can diffuse and react with oh . figure [ fig : disk_formation2 ] shows the correlation among the abundances of ice , ice , and ice in the model with @xmath138 s@xmath4 . in regions where ice is relatively abundant ( @xmath1351 % of ice ) , the ice abundance is higher than or comparable to the ice abundance .
hence , these results show that it is difficult to form ice which is closely associated with ice during the process of core collapse and disk formation . for the case that the simulations begin with an appreciable fraction of ice ( 5% relative to water ice ) , the / ratio throughout both disks is largely preserved .
this is indicated by the relatively homogenous distribution of orange points in panels ( c ) and ( f ) in figure [ fig : disk_formation1 ] .
hence , which has a prestellar or molecular cloud origin , is able to survive the chemical processing en route into the comet - forming regions of protoplanetary disks .
trajectories which are an exception to this rule , are those which have been most exposed to stellar radiation ; however , these trajectories are predominantly in the upper and closer - in layers of each protoplanetary disk and likely do not contribute to the composition of the comet - building material .
this is consistent with the earlier finding by @xcite that most water ice is delivered to protoplanetary disks without alteration or sublimation .
s@xmath4 and the very low initial ice abundance .
, width=302 ]
the simulations in the previous section show that can be produced in the gas phase in the intermediate layers of relatively warm forming disks ( see panel ( a ) in figure [ fig : disk_formation1 ] ) , with an abundance a few percent that of water ice ( i.e. , a fractional abundance of @xmath139 with respect to @xmath140 ) . the origin of the gas - phase is driven by photoprocessing of water ice by stellar uv photons en route into the disk , which releases photofragments required for forming ( o and oh ) into the gas - phase .
relatively high abundances of gas - phase are also predicted in the inner regions of protoplanetary disks around already formed stars ( e.g. , * ? ? ?
* ; * ? ? ?
the origin of gas - phase in these models is similar to that in forming disks , except that the release of photofragments of water ice photodissociation occurs over the lifetime of the disk ( @xmath141 yr ) and is driven by the uv photons generated near the disk midplane by the interaction of cosmic rays with .
persists in the gas - phase near the disk midplane because its volatility is such that it can not freezeout at the midplane temperatures within a few 10 s of au ( typically @xmath142 k ) .
figure [ fig : disko2-h2o ] shows the fractional abundance of gas ( left ) and ice ( right ) as a function of disk radius and height for a protoplanetary disk around a t tauri star ( data from * ? ? ?
similar abundances are seen for disks around both cooler ( i.e. , m dwarf ) and hotter stars ( i.e. , herbig ae ) stars , except that the water snowline is shifted to smaller and larger radii , respectively ( see * ? ? ?
the results show that gas can reach an abundance a few percent of that of water ice in the comet - formation zone ( @xmath143 au ) .
the main issue with this scenario is whether a mechanism exists whereby gas - phase formed near the disk midplane , in either forming disks or more evolved disks , can become entrapped within , and thus associated with , the water - rich ice mantle , as seen in comet 67p / c - g .
observational and theoretical studies suggest that the luminosity evolution of low - mass stars is highly variable , with frequent and strong eruptive bursts , followed by long periods of relative quiescence ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
such luminosity outbursts could have a strong impact on the morphology and the chemical composition of ices near the protoplanetary disk midplane .
the sudden temperature variations induced by short luminosity outbursts could gradually recycle the content of ices into the gas phase and modify their chemical structure via rapid and efficient freeze - out .
if the luminosity outburst is sufficiently strong , warm gas - phase formation of molecular oxygen could be triggered by the evaporation of water ice , if the peak temperature during the outburst is higher than @xmath144 k. might then be recondensed together with water post outburst , if the cooling timescale is shorter than the freeze - out timescale i.e. , @xmath145 , and also if the temperature reached after post outburst is lower than the condensation temperature of ( @xmath146 k ) .
an increase in temperature from @xmath147 to @xmath148 k during an outburst , corresponds roughly to an increase in luminosity by a factor of @xmath149 assuming that the temperature in the disk and the central luminosity are linked through stefan - boltzmann s law . the recent hydrodynamical model by @xcite shows that a dozen of such strong luminosity outbursts , with typical durations of @xmath150 yr , may occur during the disk lifetime .
the exact number depends on the physical properties of the collapsing core and the disk .
the scenario of formation and recondensation of induced by a series of outburst events in disks is investigated by a series of outbursts occurring every @xmath61 yr for a total timescale of @xmath114 yr .
the astrochemical model and chemical network used are described in section [ astrochem ] , while the assumed physical conditions are for a single point , motivated by protoplanetary disk models .
initial ice abundances are the median values derived by @xcite from interstellar ice observations towards low - mass protostars . thus it is assumed that the ice mantles are initially poor in .
the pre - outburst and post - outburst temperature is set to 20 k , corresponding approximately to the freeze - out temperature of .
protoplanetary disk models suggest that the corresponding midplane density at this point is @xmath151 @xmath2 ( e.g. , * ? ? ?
* ; * ? ? ?
* ) ; however , the exact relation between the dust temperature and gas density near protoplanetary disk midplanes depends on numerous factors including disk surface density ( or mass ) , stellar spectral type , and the dust properties .
gas phase formation of is triggered by the photodissociation of water into h and oh and consequently , is highly dependent on the assumed cosmic - ray ionisation rate , @xmath41 , which is thought to be impeded near the disk midplane with respect to interstellar values ( e.g. , * ? ? ?
the impact of @xmath41 on the formation of is investigated by considering two values which cover the possible range , @xmath152 and @xmath153 s@xmath4 .
the freeze - out timescale of a neutral species @xmath154 onto grains is given by @xmath155 where @xmath53 is the dust - to - gass mass ratio , @xmath55 the volumic mass of grains , @xmath54 the mean grain diameter , and @xmath156 the weight of species @xmath154 .
grain growth is expected to occur near protoplanetary disk midplanes . @xcite
predict an average size of 1 @xmath6 m with a dust - to - gass mass ratio of 0.01 in the midplane but the average size sharply decreases with altitude .
therefore , two grain sizes are considered @xmath157 and 1 @xmath6 m . for a fixed dust - to - gas mass ratio , @xmath158 , a larger grain size will increase the freezeout timescale , @xmath159 , relative to the cooling timescale , @xmath160 , due to the reduction in total available dust - grain surface area .
on the other hand , an increase in @xmath158 , perhaps due to settling and/or radial drift , will increase the total available grain surface area and will reduce the freezeout timescale .
six models are run to investigate the impact of various parameters on the formation and the recondensation of during outbursts which last a timescale , @xmath161 . 1 . standard model with @xmath162 @xmath6 m , @xmath163 s@xmath4 , @xmath164 yr , @xmath165 k. 2 .
same as 1 but with @xmath157 @xmath6 m .
3 . same as 1 but with @xmath166 s@xmath4 .
4 . same as 1 but with @xmath167 yr . 5 . same as 1 but with @xmath168 k. 6 .
same as 1 but with an initial abundance of 5 % relative to water .
model 6 is included to test the hypothesis that primordial , formed during the molecular cloud stage , survives both transport into the the forming protoplanetary disk and luminosity outbursts in the disk midplane .
for the standard set of parameters ( i.e. , model 1 ) , the freezeout timescale following a burst is @xmath169 yr ( see equation [ freezeout ] ) .
this is likely longer than the cooling timescale ( from @xmath170 100 to 20 k ) , expected to be shorter than the duration of the outburst ( @xmath171 yr ; * ? ? ?
figure [ outburst_fig ] shows the fractional composition of ices in each monolayer as a function of monolayer , for the six models described above .
each line represents the abundance of each species prior to the next outburst ( i.e. , following each period of cooling and quiescence ) .
in general , regardless of the assumed physical parameters , the volatile component of the ice mantle increases with time .
the ice profile is composed of two main parts : i ) a volatile - free deeper ice mantle mostly composed of water ice and other non - volatile species , such as , that primarily remain on grains during the outbursts because the temperature reached during the outburst is only slightly higher than their evaporation temperature , and ii ) an upper ice mantle composed of water ice , but also of volatile species , such as co , , and , that freezeout during the post - outburst cooling .
the deeper and volatile - free ice mantle increases in mass / depth with the duration of the outburst ( compare panels a ) and d ) in figure [ outburst_fig ] ) and with decreasing grain size that increases the surface area of dust ( compare panels a ) and b ) in figure [ outburst_fig ] ) .
the fraction of trapped in the ice mantle increases with the outburst duration , the grain size , and the cosmic - ray ionisation rate , @xmath41 , all parameters which favour the formation of gaseous from water during the outburst . increasing the grain size decreases the total grain cross - sectional area and therefore the accretion rate of gas - phase species onto the ice mantle , allowing water and other species to spend more time in the gas phase for reaction
. a higher @xmath41 increases the production of oh from the photodissociation of water vapour , necessary to form . allowing the peak temperature during outburst to reach values higher than the evaporation temperature of water ice ( @xmath172 k , panel e ) in figure [ outburst_fig ] ) , results in full sublimation of the ice mantle prior to recondensation .
however , the enhanced abundance of water released into the gas phase does not significantly enhance the abundance of formed , and subsequently trapped . in the ice mantle .
the luminosity outburst period and duration considered here are potentially too short and also too infrequent to reproduce the high amount of observed in 67p / c - g .
maximum abundances of a few @xmath173% only are predicted .
results for the calculation with 5% of ( relative to water ice ) already present in the ice , show that can survive and be efficiently trapped within the water - rich ice mantle following a series of luminosity outbursts .
hence , may become associated with water ice in the disk midplane via release and recondensation driven by outbursts . however , other volatile species such as co and are also trapped within the water ice , which is in contradiction with the observations towards 67p / c - g .
co and are shown to be depleted in 67p / c - g relative to interstellar values , and the molecules are not strongly correlated with water in the comet coma , converse to the case for @xcite .
note also that interstellar ices produced after luminosity outbursts are likely amorphous in structure @xcite .
in this work , sophisticated astrochemical models are used to investigate the chemical and physical origin of molecular oxygen in comet 67p / c - g as observed with _
the observations show that molecular oxygen is not only strongly associated with water , but is also the fourth most abundant molecule in the coma , at @xmath744% of the water abundance @xcite .
we tested the formation and survival of ( and related species ) in models covering a range of dark cloud physical conditions ( temperature , density , and cosmic - ray ionisation rate ) .
we found that the efficiency of the formation of molecular oxygen increases for higher densities ( @xmath174 @xmath2 ) , moderate temperatures ( @xmath147 k ) , and moderate ionisation rates ( @xmath175 s@xmath4 ) .
these conditions lower the ratio of h / o in the gas phase , thereby impeding the conversion of ice into ice .
these parameters are found to be in good agreement with the physical conditions for the dark cloud @xmath14 oph a , one of the two interstellar regions where has been detected .
the high abundances do not require photolysis of bulk ice as the main process but are the result of surface reactions building up the ice layers .
we next tested whether molecular oxygen can be efficiently formed in the ice mantle during protostellar disk formation . for models in which the initial ice composition is assumed to be poor in ,
can produced only through gas phase chemistry induced by processing of the water - rich ice mantle by stellar uv radiation in the intermediate - layers of the protoplanetary disk , with abundance levels similar to that in 67p / c - g but ices in the disk midplane remain poor in . for models in which the ice mantles were originally abundant in molecular oxygen ( @xmath37% relative to water ) , for both disk models , we find that the oxygen is delivered to the comet - forming zone without sublimation nor alteration . hence , if molecular oxygen has a primordial origin as suggested by the dark cloud models , then it can survive transport into the protoplanetary disk .
given that gas - phase can form near protoplanetary disk midplanes , and reach abundances relative to water ice similar to that in 67p / c - g , we finally tested whether luminosity outbursts which increase the local temperature to @xmath176 k , aid the formation and entrapment of gas - phase into the water - rich ice mantle .
although laboratory experiments show that can be efficiently formed within water ices during the ice recondensation through radiolysis @xcite , we consider this less likely because the cosmic - ray ionisation rate and energetic particles from the ( pre)solar wind are expected to be significantly attenuated near the disk midplane .
it is found that the maximum amount of formed during luminosity outbursts and then trapped within the ice mantle during the cooling depends on several parameters , such as grain size , ionisation rate , or the outburst duration , but never exceeds @xmath177% .
assuming an initial abundance of @xmath178% relative to water ice results in an efficient trapping of within the water - ice mantle due to the fast cooling after the outburst .
however , in that case also other volatile species , such as co and , become trapped , which is in contradiction with observations towards 67p / c - g . in summary ,
the models presented here favour the scenario that molecular oxygen in 67p / c - g has a primordial origin ( i.e. , formed in the molecular cloud ) and has survived transport through the protostellar envelope and into the comet - forming regions of protoplanetary disks .
the `` primordial '' origin of is in good agreement with the conclusions of @xcite .
however , while @xcite invoked radiolysis to efficiently convert water ice to , we find that the entrapment and strong association with water ice combined with low abundance of species like , , or can alternatively be explained by an efficient formation at the surface of interstellar ices through oxygen atom recombination in relatively warmer ( @xmath179 k ) and denser ( @xmath87 @xmath2 ) conditions than usually expected in dark clouds . the weak correlation of co and with water seen in 67p / c - g is explained by a later formation of these species in dark clouds with respect to and water .
this picture would therefore be consistent with the physical and chemical properties of our solar system , such as the presence of short - lived radio isotopes in meteorites or the orbits of solar system planets , suggesting that our solar system was born in a dense cluster of stars ( see * ? ? ?
the authors thank t. lamberts , e. bergin , and the rosina team , especially k. altwegg , m. rubin , and a. bieler , for fruitful discussions and comments on the manuscript and m. persson for making figure 1 .
astrochemistry in leiden is supported by the european union a - erc grant 291141 chemplan , by the netherlands research school for astronomy ( nova ) , by a royal netherlands academy of arts and sciences ( knaw ) professor prize .
is supported by the research fellowship from the japan society for the promotion of science ( jsps ) . c. w. acknowledges support from the netherlands organization for scientific research ( nwo , program 639.041.335 ) . 99 acharyya , k. , fuchs , g. w. , fraser , h. j. , van dishoeck , e. f. , & linnartz , h. 2007 , , 466 , 1005 adams , f. c. 2010 , , 48 , 47 , k. , balsiger , h. , bar - nun , a. et al .
2015 , science , 347 , 27 atkinson , r. , baulch , d. l. , cox , r. a. et al .
2004 , atmospheric chemistry and physics , 4 , 1461 andersson , s. , al - halabi , a. , kroes , g.j . , &
van dishoeck , e. f. 2006 , , 124 , 4715 andersson , s. & van dishoeck , e. f. 2008 , , 491 , 907 arasa , c. , koning , j. , kroes , g .- j . ,
walsh , c. , & van dishoeck , e. f. 2015 , , 575 , a121 balsiger , h. , altwegg , k. , boschler , p. , et al .
2007 , , 128 , 745 baragiola , r. a. , atteberry , c. l. , dukes , c. a. et al .
2002 , _ nuclear instruments and methods in physics research b _ , b , 720 bergeron , h. , rougeau , n. , sidis , v. , et al .
2008 , j. phys .
a , 112 , 11921 bergman , p. , parise , b. , liseau , r. , et al . 2011a , , 527 , a39 bergman , p. , parise , b. , liseau , r. , et al .
2011b , , 531 , l8 bieler , a. , altwegg , k. , balsinger , h. , et al .
2015 , nature , 526 , 678 cassen , p. , & moosman , a. 1981 , icarus , 48 , 353 cazaux , s. , minissale , m. , dulieu , f. and hocuk , s. 2016 , , 585 , a55 chen , j - h . ,
goldsmith , p. f. , viti , s. et al .
2014 , , 793 , 111 cleeves , l. i. , adams , f. c. , & bergin , e. a. 2013 , , 772 , 5 , l. i. , bergin , e. a. , alexander , c. m. o. et al .
2014 , science , 345 , 1590 collings , m. p. , dever , j. w. , fraser , h. j. , mccoustra , m. r. s. , & williams , d. a. 2003 , , 583 , 1058 collings , m. p. , anderson , m. a. , chen , r. , et al .
2004 , , 354 , 1133 collings , m. p. , frankland , v. l. , lasne , j. , et al .
2015 , , 449 , 1826 congiu , e. , minissale , m. , baouche , s. , et al .
2014 , in faraday discussions , 168 , 151 cuppen , h. m. and herbst , e. 2007 , , 668 , 294 cuppen , h. m. , ioppolo , s. , romanzin , c. , & linnartz , h. 2010a , phys .
( incorporating faraday transactions ) , 12 , 12077 drozdovskaya , m. n. , walsh , c. , visser , r. , harsono , d. , & van dishoeck , e. f. 2014 , , 445 , 913 du , f. and parise , b. 2012 , , 538 , a91 dulieu , f. 2011 , iau symposium , 280 , 405 eistrup , c. , walsh , c , & van dishoeck , e. f. 2016 , , in preparation fayolle , e. c. , bertin , m. , romanzin , c. , et al .
2013 , , 556 , a122 fuchs , g. , acharyya , k. , bisschop , s. e. , et al .
2006 , faraday disc .
, 133 , 331 furuya , k. , aikawa , y. , nomura , h. , hersant , f. , & wakelam , v. 2013 , , 779 , 11 furuya , k. , aikawa , y , hincelin , u. , hassel , g. , bergin , e. a. , vasyunin , a. i. , & herbst , e. , 2015 , 584 , a124 furuya , k. , drozdovskaya , m. , visser , r. et al .
2016 , submitted to garrod , r. t. , & herbst , e. 2006 , , 457 , 927 garrod , r. t. , wakelam , v. , & herbst , e. 2007 , 467 , 1103 , r. t. , weaver , s. l. w. and herbst , e. 2008 , , 682 , 283 - 302 goldsmith , p. f. , melnick , g. j. , bergin , e. a. , et al .
2000 , , 539 , l123 goldsmith , p. f. , liseau , r. , bell , t. a. , et al .
2011 , , 737 , 96 hall , d. t. , strobel .
d. f. , feldman , p. d. , mcgrath , m. a. , & weaver , h. a. 1995 , nature , 373 , 677 hand , k. p. and carlson , r. w. 2011 , _ icarus _ , 215 , 226 harada , n. , herbst , r. , & wakelam , v. 2010 , , 721 , 1570 harsono d. , visser r. , bruderer s. , van dishoeck e. f. , & kristensen l. e. , 2013 , , 555 , a45 , t. i. and herbst , e. 1993 , , 263 , 589 - 606 he , j. , shi , j. , hopkins , t. , et al .
2015 , , 801 , 120 heays , a. n. , bosman , a. d. , & van dishoeck , e. f. 2016 , , submitted herbig , g. h. 1977 , , 217 , 693 , u. , wakelam , v. , hersant , f. et al .
2011 , , 530 , a61 ioppolo , s. , cuppen , h. m. , romanzin , c. , van dishoeck , e. f. , & linnartz , h. 2008 , , 686 , 1474 ioppolo , s. , cuppen , h. m. , romanzin , c. , van dishoeck , e. f. and linnartz , h. , 2010 , physical chemistry chemical physics ( incorporating faraday transactions ) , 12 , 12065 jaycock , m. j. & parfitt , g. d. 1986 , chemistry of interfaces ( wiley and sons , new york ) , l. j. and cuppen , h. m. 2014 , , 569 , a107 katz , n. , furman , i. , biham , o. , pirronello , v. , & vidali , g. 1999 , , 522 , 305 hartmann , l. & kenyon , s. j. 1985 , , 299 , 462 kouchi , a. , yamamoto , t. , kozasa , t. , kuroda , t. , & greenberg , j. m. 1994 , , 290 , 1009 lamberts , t. , cuppen , h. m. , ioppolo , s. and linnartz , h. 2013 , phys .
, 15 , 8287 larsson , b. , liseau , l. , pagani , l. , et al .
2007 , , 466 , 999 liseau , r. , larsson , b. , bergman , p. , et al .
2010 , , 510 , a98 liseau , r. , goldsmith , p. f. , larsson , b. , et al .
2012 , , 541 , a73 loeffler , m. j. , raut , u. , vidal , r. a. , baragiola , r. a. , and carlson , r. w. 2006 , _ icarus _ , 180 , 265 lynden - bell , d. & pringle , j. e. 1974 , , 168 , 603 matar , e. , congiu , e. , dulieu , f. , momeni , a. , & lemaire , j. l. 2008 , , 492 , l17 matich , a. j. , bakker , m. g. , lennon , d. et al .
1993 , the journal of physical chemistry , 97 , 10539 - 10553 , d. , walsh , c. , markwick , a. j. , cordiner , m. a. , smith , k. and millar , t. j. 2013 , , 550 , a36 melius , c. f. , and blint , r. j. 1979 , chem .
, 64 , 183 melnick , g. j. & kaufman , m. j. 2015 , , 806 , 227 minissale , m. , congiu , e. and dulieu , f. 2014 , , 140 , 074705 minissale , m. & dulieu , f. 2014 , , 141 , 014304 minissale , m. , dulieu , f. , cazaux , s. , & hocuk , s. 2016 , , 585 , a24 mousis , o. , ronnet , t. , brugger , b. et al .
2016 , accepted in apjl miyauchi , n. , hidaka , h. , chigai , t. , nagaoka , a. , watanabe , n. , & kouchi , a. 2008 , chem .
, 456 , 27 noble , j. , congui , e. , dulieu , f. , & fraser , h. j. 2012 , , 421 , 768 berg , k. i. , linnartz , h. , visser , r. , & van dishoeck , e. f. 2009 , , 693 , 1209 berg , k. i. , boogert , a. c. a. , pontopiddan , k. m. , et al .
2011 , , 740 , 109 pagani , l. , olofsson , a. o. h. , bergmann , p. , et al .
2003 , , 402 , l77 parise , b. , bergman , p. , & du , f. 2012 , , 541 , l11 perets , h. b. , biham , o. , manic , g. , et al .
2005 , , 627 , 850 , k. m. , fraser , h. j. , dartois , e. et al .
2003 , , 408 , 981 - 1007 prasad , s. s. & tarafdar , s. p. 1983
, , 267 , 603 rimola , a. , taquet , v. , ugliengo , p. 2014
, , 572 , a70 rubin , m. , altwegg , k. , balsiger , h. , et al .
2015a , science , 348 , 232 rubin , m. , altwegg , k. , van dishoeck , e. f. & schwehm , g. 2015b , , 815 , l11 shakura , n. i. , & sunyaev , r. a. 1973 , , 24 , 337 shu f. h. , 1977 , , 214 , 488 , m. t. , simpson , w. c. and orlando , t. m. 1998 , , 394 , 554 - 556 slanger , t. g. & black , g. 1982 , , 77 , 2432 snellen , i. a. g. , de kok , r. j. , le poole , r. , brogi , m. , & birkby , j. 2013 , , 764 , 182 spencer , j. r. , calvin , w. m. , & person , m. j. 1995 , journal for geophysical research , 100 , 19049 taquet , v. , ceccarelli , c. , & kahane , c. 2012 , , 538 , a42 taquet , v. , peters , p. , kahane , c. et al .
2013 , , 550 , a127 taquet , v. , charnley , s. b. , sipil , o. 2014 , , 791 , 1 , b. d. , loeffler , m. j. , raut , u. , fam , m. and baragiola , r. a. 2005 , , 644 , l141 teolis , b. d. , jones , g. h. , miles , p. f. , et al .
2010 , _ science _ , 330 , 1813 terebey , s. , shu , f. h. , & cassen , p. 1984
, , 286 , 529 tielens , a. g. g. m. , & hagen , w. , , 114 , 245 tielens , a. g. g. m. , & allamandola , l. j. 1987 , interstellar processes , eds .
d. j. hollenbach , & h. a. thronson , 397 , a. g. g. m. 2005 , the physics and chemistry of the interstellar medium ( cambridge university press ) tielens , a. g. g. m. 2013 , rev .
mod . phys . ,
85 , 1021 vandenbussche , b. , ehrenfreund , p. , boogert , a. c. a. , et al .
1999 , , 346 , l57 van harrevelt , r & van hemert , m. c. 2008 , j. phys . chem .
a , 112 , 3002 van dishoeck , e. f. , herbst , e. , & neufeld , d. a. 2013 , chem .
, 113 , 9043 vasyunin , a. i. , wiebe , d. s. , birnstiel , t. et al .
2011 , , 727 , 76 vasyunin , a. i. and herbst , e. 2013 , , 762 , 86 visser , r. , van dishoeck , e. f. , doty , s. d. , & dullemond , c. p. 2009 , , 495 , 881 visser , r. , doty , s. d. , & van dishoeck , e. f. 2011 , , 534 , a132 vorobyov , e. i. & basu , s. 2005 , 633 , l137 vorobyov , e. i. & basu , s. 2015 , , 805 , 115 , v. and herbst , e. 2008 , 680 , 371 - 383 wakelam , v. , herbst , e. , loison , j-.c .
, et al . 2012 , , 199 , 21 walch , s. and duchovic , r. 1991 , j. chem .
phys . , 94 , 70687075 walsh , c. , millar , t. j. , nomura , h. , et al .
2014 , , 563 , 33 walsh , c. , nomura , h. , & van dishoeck , e. f. 2015 , , 582 , a88 westley , m. s. , baragiola , r. a. , johnson , r. e. , & baratta , g. a. 1995 , nature , 373 , 405 yildiz , u. a. , acharyya , k. , goldsmith , p. f. , et al .
2013 , , 558 , a58 zhen , j. & linnartz , h. 2014 , , 437 , 3190 , w. , jewitt , d. and kaiser , r. i. 2006 , , 639 , 534 - 548
a set of models is run , in order to investigate the impact of various surface and chemical parameters on the chemical composition of interstellar ices and to assess whether the abundances of , , , and observed in comet 67p / c - g can be reproduced simultaneously . in each model ,
the `` standard '' values of the input parameters , listed in table [ grid_table ] , are assumed and only one of the parameters is varied in turn .
in particular , the physical conditions assumed here are conditions that favour a high production of , i.e. a high density @xmath180 @xmath2 and a warm temperature @xmath101 k , according to the discussion in section [ imp_params ] .
the impact of two surface parameters , the diffusion - to - binding energy ratio , @xmath47 , and the binding energy for on the formation and survival of ice and the chemically related species is investigated first . following the discussion in section [ imp_params ] , models in which the diffusion - to - binding energy ratio ranges between 0.3 and 0.8 and for which the binding energy of atomic oxygen ranges between 800 and 1700 k have been run .
figure [ appendix_fig1 ] shows the abundances of , , , and in the solid phase ( relative to water ice ) , at a time of @xmath181 yr for the different values of the input chemical parameters .
the abundance of ( and that for chemically related species ) tends to decrease as @xmath47 is increased , both in the gas phase and in the ice mantle .
the formation rate of is governed by the mobility of o atoms . due to their relatively high binding energy ( @xmath182 k for the standard model ) , o atoms
can diffuse efficiently only if @xmath183 .
higher values strongly inhibit the mobility , leading to a low abundance of , , , and ( see left panel in figure [ appendix_fig1 ] ) .
decreasing the binding energy of o atoms from 1700 to 800 k increases their mobility but also increases the rate of evaporation at 20 k. this then limits the conversion from o to and , leading to a negligible dependence of the final abundances upon the assumed binding energy for atomic oxygen ( see right panel in fig .
[ appendix_fig1 ] ) . the high abundance of seen in 67p / c - g can be reproduced for dense and warm conditions and assuming the standard values for the diffusion - to - binding energy ratio and the binding energy of atomic o. however , the abundances of , , are overproduced by more than two orders of magnitude when standard values for the activation barriers of surface reactions are assumed .
yr , when the gas phase abundance of the standard model reaches the abundance observed in @xmath14 oph a , for different values of the surface parameters @xmath47 ( left ) and @xmath48(o ) ( right ) .
the thick dashed lines refer to the abundances observed in the comet 67p / c - g.,title="fig:",scaledwidth=33.0% ] yr , when the gas phase abundance of the standard model reaches the abundance observed in @xmath14 oph a , for different values of the surface parameters @xmath47 ( left ) and @xmath48(o ) ( right ) .
the thick dashed lines refer to the abundances observed in the comet 67p / c - g.,title="fig:",scaledwidth=33.0% ] the impact of the activation barriers of the reactions o + , h + , and h + on the abundances of , , and in ices is explored next .
figure [ appendix_fig2 ] shows how the abundances of , , and in the gas phase ( top row ) and in the solid phase ( bottom row ) relative to vary with the activation barriers of these three reactions . as described in section [ imp_params ] , surface reactions involving have small or negligible reaction barriers but the activation barrier of the reaction o + remains uncertain .
due to the relatively high masses of o and , the activation barrier can be only overcome thermally . at a dust temperature of 21 k
, the reaction probability exponentially decreases from 1 ( @xmath184 k ) to @xmath185 ( @xmath186 k ) . as a consequence ,
a small activation barrier of 250 - 300 k favours the formation and survival of , , and with respect to and allows us to reproduce the low / abundance ratio seen in comet 67p / c - g .
this result is in good agreement with the results of @xcite who needed to introduce an activation barrier of 500 k in their microscopic monte - carlo model to explain the slow formation of ozone observed in laboratory experiments of @xcite .
the o + reaction still takes place and induces the evaporation of into the gas phase via chemical desorption , explaining the high abundance of in the gas phase .
however , this occurs at a slower rate than the reactions destroying ice through barrierless surface reactions . in the model results , a high abundance of ice is consistently accompanied by similar abundances of and ice , because the h + reaction is assumed to be barrierless , following the laboratory experiments of @xcite and @xcite .
however , quantum chemistry calculations by @xcite show that the reaction in the gas phase has an activation barrier whose exact value depends on the incoming angle of the molecule .
@xcite introduced a small activation barrier of 200 - 400 k for this reaction in their monte - carlo model to reproduce the chemical composition observed in cold ices produced in the laboratory experiments by @xcite .
the activation barrier of the reaction h + was therefore varied between 0 k ( our standard value ) and 1200 k , the energy computed for gas phase conditions by @xcite . the transmission probability through quantum tunelling
was computed assuming a rectangular barrier with a standard width of 1 @xmath187 . as shown in the middle panel of fig .
[ appendix_fig2 ] , the abundance of ice decreases sharply with the activation barrier of the reaction h + , even for moderate values because its rate of formation becomes much lower than its rate of destruction while the decrease of the abundance of solid is more limited .
a small activation barrier of 300 k , similar to the values found by @xcite , is therefore sufficient to reproduce the low abundance of relative to observed in comet 67p / c - g .
the standard activation barrier of the h + reaction of 2500 k and the associated transmission probability computed with the eckart model , follow the quantum chemical calculations for gas phase conditions presented in @xcite . as for other reactions showing high activation barriers ,
the exact value of their barrier is highly uncertain and could be lowered in interstellar ices , due to the van der waals interactions between the neighbouring water molecules and the reactants , as shown for instance by @xcite for the case of the co + h and + h reactions .
@xcite decreased the activation barrier for the h + to 800 - 1200 k in their monte - carlo model to reproduce the abundance in laboratory cold ices .
the right panel of fig .
[ appendix_fig2 ] present the evolution of the , , and abundances relative to for different values of the activation barrier of the reaction h + , and assuming an activation barrier of 300 k for the reactions o + and h + .
the model tends to overpredict the abundance of relative to the abundance observed in 67p , unless the reaction h + is assumed to be effectively barrierless . | molecular oxygen has been confirmed as the fourth most abundant molecule in cometary material ( / @xmath0 % ) and is thought to have a primordial nature , i.e. , coming from the interstellar cloud from which our solar system was formed .
however , interstellar gas is notoriously difficult to detect and has only been observed in one potential precursor of a solar - like system . here , the chemical and physical origin of in comets is investigated using sophisticated astrochemical models .
three origins are considered : i ) in dark clouds , ii ) during forming protostellar disks , and iii ) during luminosity outbursts in disks .
the dark cloud models show that reproduction of the observed abundance of and related species in comet 67p / c - g requires a low h / o ratio facilitated by a high total density ( @xmath1 @xmath2 ) , and a moderate cosmic ray ionisation rate ( @xmath3 s@xmath4 ) while a temperature of 20 k , slightly higher than the typical temperatures found in dark clouds , also enhances the production of .
disk models show that can only be formed in the gas phase in intermediate disk layers , and can not explain the strong correlation between and in comet 67p / c - g together with the weak correlation between other volatiles and .
however , primordial ice can survive transport into the comet - forming regions of disks . taken together , these models favour a dark cloud ( or " primordial ) origin for in comets , albeit for dark clouds which are warmer and denser than those usually considered as solar system progenitors .
[ firstpage ] ism : abundances ism : molecules astrochemistry protoplanetary discs stars : formation comets : individual : 67p / c - g |
w. veech s celebrated dichotomy @xcite applies in the setting of his initial examples : gluing together two copies of a regular @xmath1-gon by identifying opposite sides by translation results in a translation surface @xmath3 having exactly two types of directions for linear flow completely periodic or uniquely ergodic .
indeed , a completely periodic direction @xmath4 enjoys the stronger property that the surface decomposes into a collection of cylinders in the direction , and the moduli of these cylinders are rationally related .
directly related to this , there is a parabolic @xmath5 real matrix fixing the inverse of the slope , thus @xmath6 ; veech shows that for these examples , the obvious rotational symmetry and a parabolic element generate the full ( fuchsian ) group of derivatives of the ( orientation preserving ) affine diffeomorphisms of the surface .
after calta @xcite , and calta
smillie @xcite , one is strongly motivated to normalize by taking a surface in the @xmath7-orbit of @xmath3 whose planar presentation has its vertices in the trace field of @xmath8 for then also the set of slopes in which the @xmath9-invariant vanishes coincides with this field .
here , we apply work of a. leutbecher and others on cusps of related fuchsian groups to easily reach the sobering implication that even surfaces of optimal dynamics have non - parabolic directions with vanishing @xmath9-invariant . in a certain sense ,
the differing planar presentations of the surfaces emphasize different number fields , with the original veech construction being directly related to @xmath10 .
thus the choice of normalized surface requires understanding the relationships of various trigonometric fields .
the literature reveals some minor confusion about these relationships and hence , in an appendix , we address the various possibilities for containments of number fields generated over @xmath11 by trigonometric values . for each of veech s @xmath8 we give a @xmath7-equivalent surface normalized so that its periodic direction field in the sense of calta and smillie
equals the trace field . with this in hand , we use number theoretic results to find explicit non - parabolic directions with vanishing saf - invariant . in particular , we show the following .
[ main ] all directions of vanishing saf - invariant are parabolic on @xmath0 if and only if the trace field of @xmath0 is at most a quadratic extension of the field of rational numbers .
thus , exactly for @xmath12 . [ thmegs ] the set of indices for which the surface @xmath0 has an affine pseudo - anosov diffeomorphism whose contracting direction has vanishing saf - invariant includes @xmath13 . [ remone ] our proof of theorem [ main ] proceeds by first replacing @xmath0 by a natural degree two quotient , clearly identifying normalized surfaces equivalent to the these surfaces so as to apply arithmetic results of the school of a. leutbecher on parabolic fixed points of the hecke groups .
these groups are commensurable , that is conjugate up to finite index , with the veech groups of the @xmath8 .
this reduces us to a finite number of unresolved cases ; for these , we construct explicit `` special '' hyperbolic matrices and thus affine pseudo - anosov diffeomorphisms whose contracting direction has vanishing saf - invariant .
these examples also complete leutbecher s project .
in fact , the positive statement of theorem [ main ] follows from results now already known : mcmullen @xcite has shown ( using equivalent terminology ) that any veech surface of quadratic trace field has the property that all its directions of vanishing saf - invariant are parabolic . that the @xmath0 of higher degree trace field no longer enjoy this property leads us to presume that no veech surface of cubic or higher trace field has only parabolic directions have vanishing saf - invariant .
this is another indication that , in genus larger than 2 , the situation is much more complicated than the rigid structure in genus 2 discovered by calta and mcmullen . in the following section
, we briefly recall background .
further background material appears in the initial piece of section [ sectrisurf ] .
that section ends with a presentation of the normalized form of the veech examples . in section [ secapplyingleut ] , we apply work of leutbecher and of later authors to reduce to a finite number of cases .
section [ secnonparab ] rules out the remaining cases by providing explicit examples of special affine pseudo - anosov maps on veech surfaces .
finally , in the appendix , based largely on the paper @xcite of d. h. lehmer , we clarify the relationship between various trigonometric fields .
the second - named author will keep available on his home page a mathematica notebook containing the basic computations for the examples for @xmath14 of subsection [ subeven ] . at the time of the submission of this manuscript ,
the following is a url for this file http://oregonstate.edu/ schmidtt / ourpapers / arnoux / ascode.nb the present note remained in nascent form for several years .
indeed , the results of section [ secapplyingleut ] are referred to in @xcite .
we have benefited from the prompting of p. hubert and from discussions with him , e. lanneau and j. schmurr .
we briefly recall terminology and give more detailed motivation for this work .
a _ translation surface _ is a real surface with an atlas such that , off of a finite number of points , transition functions are translations . from the euclidean plane
, this punctured surface inherits a flat metric , and this metric extends to the complete translation surface , with ( possibly removable ) conical singularities at the punctures . due to the transition function
being translations , directions of linear flow on a translation surface are well - defined .
post - composing the coordinate function of a chart from the atlas of a translation surface with any element of @xmath15 results in a new translation surface .
this action preserves both the underlying topology and the types of the conical singularities .
related to this , an _ affine diffeomorphism _ of the translation surface is a homomorphism that restricts to be a diffeomorphism on the punctured flat surface whose derivative is a _ constant _ @xmath16 real matrix .
w. veech @xcite showed that for any compact translation surface , the matrices that arise as such derivatives of ( orientation- and area - preserving ) affine diffeomorphisms form a discrete subgroup of @xmath17 , thus form a fuchsian group , now referred to as the _ veech group _ of the surface .
furthermore , veech showed that if this group is a lattice ( that is , if it is of finite co - area ) , then the directions for flow on the translation surface enjoys the aforementioned dichotomy .
such surfaces are now referred to as _
veech surfaces_. perhaps the main reason that w. thurston introduced the notion of affine diffeomorphism is that it allows visualization of the action of a pseudo - anosov diffeomorphism , see @xcite an affine diffeomorphism is a pseudo - anosov diffeomorphism exactly when its derivative is a hyperbolic matrix ( thus of trace greater than 2 in absolute value ) .
the two real fixed points of this matrix give the expanding and contracting directions of the diffeomorphism .
we refer to these diffeomorphisms as _ affine pseudo - anosov _ diffeomorphisms .
an interval exchange is an interval map that acts by translating subintervals .
the sah - arnoux - fathi invariant , or simply _ saf - invariant _ , of an interval exchange map is @xmath18 if the interval exchange takes intervals of lengths @xmath19 and translates them by the various @xmath20 .
this invariant defines a homomorphism from the group ( under composition ) of interval exchange maps to the additive group @xmath21 , hence it vanishes whenever the interval exchange map is completely periodic .
it can be proved that a minimal exchange map whose translation lengths generate a rank 2 @xmath22-module has non - vanishing invariant .
this is in particular the case if the translation lengths belong to a fixed quadratic field ; hence any interval exchange map that is quadratic in this sense and has a null invariant is either periodic or reducible , see @xcite .
note that this is no longer the case in higher degree : for any @xmath23 , there are minimal exchange maps with lengths in an algebraic field of degree @xmath24 and a null invariant , @xcite .
when we consider a flow preserving a transverse measure on a surface , to any tranverse interval we can associate a first return map which is an interval exchange map .
it is not difficult to prove that , as long as the interval cuts every orbit of the flow , the invariant does not depend on the choice of the interval , but only on the flow , see @xcite .
if the flow is minimal , the invariant is the same for all choices of the interval ; otherwise , it is the sum of the invariants of the minimal components .
kenyon and smillie @xcite introduced an invariant for translation surfaces , the @xmath25-invariant , that takes values in @xmath26 .
they showed that appropriate projections of the @xmath25-invariant of the surface give the saf - invariants associated to the directions @xmath4 and the surface .
if the direction is _ completely periodic _ , in the sense that every orbit of the flow is either periodic , or a saddle connection , then the saf - invariant of this direction vanishes . in particular , the aforementioned parabolic directions have vanishing saf - invariant .
calta and smillie in @xcite show that if a translation surface has at least three directions with vanishing saf - invariant , then upon appropriately normalizing the surface , the set of slopes of vanishing saf - invariant forms the union of infinity with a field .
they call this the _ periodic direction field _ , and say that the surface is in _
standard form_. each translation surface has a very natural set of directions a _ holonomy _ direction for the surface is a direction for which some closed cycle exists .
every completely periodic direction is certainly a holonomy direction , but in general the opposite inclusion does not hold : for example , there could be a closed cycle separating two minimal components . slightly more subtly ,
a _ holonomy vector _ is defined by using the so - called developing map , and gives a flat representative of a cycle .
kenyon and smillie @xcite defined the _ holonomy field _ to be the smallest field over which the set of holonomy vectors is contained in a two - dimensional vector space .
gutkin and judge @xcite defined the _ trace field _ of a translation surface to be the field extension of the rationals generated by the traces of derivatives of the affine diffeomorphisms of the surface ; this is clearly a conjugacy invariant , and does not depend on the choice of a particular translation surface in the @xmath27-orbit .
calta and smillie @xcite show that a veech surface has equality of its trace , holonomy and periodic direction fields .
results of calta @xcite and of mcmullen @xcite show that on any genus two veech surface every direction of vanishing saf - invariant is a parabolic direction .
this result uses the fact that the degree of the periodic direction field is bounded by the genus .
indeed , as already alluded to in remark [ remone ] , mcmullen @xcite shows that if the trace field of a veech surface is quadratic , then upon appropriately normalizing , the parabolic directions form the full trace field . on the other hand , even in genus two , there are many translation surfaces ( with two singularities ) with directions that are completely periodic but not parabolic @xcite .
also , there are examples of genus two surfaces with a direction of vanishing saf - invariant but for which the surface decomposes into a connected sum of tori , each of which is `` irrationally foliated '' in the given direction @xcite . in this case
there are two minimal components with nonzero , but opposite , invariant , so that the global invariant vanishes .
whereas in genus two this is the only manner to realize a non - completely periodic direction with vanishing saf - invariant ( see the corollary to theorem 3.7 of @xcite ) , in genus three and higher arnoux and yoccoz @xcite already gave examples of affine pseudo - anosov diffeomorphisms of vanishing saf - invariant ; see @xcite for recent work related to these examples . for none of these earlier examples of non - parabolic directions with vanishing saf - invariant
is the underlying surface a veech surface . given any fuchsian group , the _ trace field _
is the field extension of @xmath11 generated by the traces of the elements of the group . in general , the eigenvalues belong to a quadratic extension of this field , but it can happen that an element of this trace field is the fixed point of a hyperbolic matrix @xmath28 in the group .
the standard fixed point formula shows that this is the case if and only if @xmath29 is a square in this trace field .
this curious phenomenon has been noted in other contexts , see in particular @xcite , where they call such @xmath28 _ special hyperbolic_. we call an affine pseudo - anosov diffeomorphism _ special _ if its derivative is a special hyperbolic matrix . recall that the larger eigenvalue of this matrix is called the _ dilatation _ of the diffeomorphism . in general , the dilatation lies in a number field of degree at most twice the genus of the surface .
usually , this bound is achieved .
however , a special pseudo - anosov diffeomorphism has its dilatation in a field of `` unexpectedly '' lower degree .
in the even index @xmath30 case , arnoux - hubert @xcite note that the interchange of veech s two regular @xmath1-gons defines an involution on the surface @xmath8 .
the quotient is a translation surface that is equivalent to @xmath31 , the surface given by identifying by translation opposite sides of the regular @xmath1-gon gives a surface .
it is easily verified that this quotient surface has the same parabolic directions , and indeed has its veech group equal to that of @xmath8 . independently , earle and gardiner @xcite discussed the @xmath32 as triangle surfaces .
let @xmath33 be a relatively prime triple of positive integers and let @xmath1 be their sum .
let @xmath34 denote the translation surface arising from the unfolding of a euclidean triangle of angles @xmath35 . for odd @xmath1
, one has @xmath36 ; earle - gardiner show that @xmath37 .
recall that @xcite discusses the construction of triangle surfaces .
let our triangle @xmath38 be of angles @xmath39 as above , and such that the side opposite @xmath40 is of length one and agrees with the real line along @xmath41 $ ] .
thus the vertices of @xmath38 are @xmath42 and , say , @xmath43 . using the law of sines , the vertex @xmath44 has coordinates @xmath45 , with @xmath46 .
calta and smilie @xcite , proof of theorem 1.4 , prove the following : let @xmath33 be a relatively prime triple of positive integers , and let @xmath1 be their sum .
then the trace field of @xmath47 is @xmath48 . letting @xmath49 denote the triangle whose vertices are the complex conjugates of those of @xmath38 and @xmath50 be a primitive @xmath1th root of unity , using standard complex notation
, @xcite shows that up to appropriate side identifications the surface @xmath51 can be given as the union @xmath52 .
they also show that all vertices of this union are in @xmath53 .
there are two natural actions of matrix groups in our setting , each giving a normal form for a translation surface .
a translation surface is _ projectively normalized _ if all slopes of non - vertical homological directions lie in the trace field .
a translation surface given as a collection of euclidean polygons with sides identified by translations is _ linearly normalized _ if all of the vertices have coordinates in the trace field .
one easily verifies that a linearly normalized translation surface is projectively normalized .
( for this , see the argument in the proof of calta - smillie theorem 1.4 in @xcite . ) the theorem itself can be expressed in the following manner .
( calta - smillie @xcite ) any projectively normalized triangle surface has equality of its trace and periodic fields .
taking into account lemma [ ksimprts ] of the appendix , the aforementioned work of @xcite and @xcite implies the following .
let @xmath54 [ correctnormalization ] let @xmath33 be a relatively prime triple of positive integers and let @xmath1 be their sum .
let @xmath47 denote the translation surface resulting from the standard unfolding process of the triangle of angles @xmath55 and of vertices normalized as above .
let @xmath56 then @xmath57 is linearly normalized .
thus , the set of slopes of non - vertical directions of @xmath57 whose linear flow has vanishing saf - invariant is @xmath58 , the trace field of @xmath59 .
j. schmurr @xcite points out that the authors of @xcite forgot to apply the above normalization in the proof of their theorem 1.4 ( he gives an alternate proof of their theorem in his 2008 oregon state university ph.d .
dissertation ) .
when the index @xmath1 is odd , the veech group of @xmath0 is a hyperbolic triangle fuchsian group of signature @xmath60 meaning that ( by way of its action on the hyperbolic plane ) it uniformizes a once punctured sphere with singularities of torsion index @xmath61 and @xmath62 , respectively . when the index @xmath1 is even , the veech group is a subgroup of index two in a group of signature @xmath63 .
the parabolic directions on a translation surface are in two - to - one correspondence with the parabolic fixed points of the veech group of the surface ( directions @xmath4 and @xmath64 have identical images ) . for brevity
, one often refers to the set of the parabolic fixed points of a fuchsian group , acting on the ( extended ) reals , as its _ cusps_. perhaps unfortunately , it is also traditional to refer to each group orbit of these fixed points as a cusp .
context usually suffices to clarify this ambiguity .
we translate our geometric questions into questions about the cusps of a well - studied family of groups that are conjugate to our veech groups . in the first subsection
, we briefly recall this work .
a standard presentation of the hyperbolic triangle fuchsian group of signature @xmath63 , is as the hecke group @xmath65 with @xmath66 .
the first of these generators is parabolic , and for its mbius action on the extended reals , has @xmath67 as its fixed point . from this
it follows that every element of the @xmath68-orbit of @xmath67 is a parabolic fixed point .
as the signature indicates , @xmath68 has exactly one cusp : this orbit is the set of all parabolic fixed points .
now , the orbit of infinity is @xmath69 .
this is obviously a subset of @xmath70 , but in fact more can be said . here and below ,
notation such as @xmath71 denotes the set obtained upon multiplying all elements of the indicated field by the leading factor .
( leutbecher @xcite)[thmleut ] for each @xmath72 @xmath73 equality holds for @xmath12 . that @xmath74 is a subset of @xmath75 is easily shown by induction using the generators displayed above an ordered pair @xmath76 giving a column of any element of @xmath68 must be such that exactly one element of the pair is in @xmath77\,$ ] , and the other is in @xmath78\,$ ] .
( note also that only for @xmath1 even is @xmath79 unequal to @xmath80 , and then it is an index two subfield see lemma [ cos2tansqrd ] and lemma [ lemcos2x ] to verify this elementary observation . )
leutbecher s list of indices is exactly those for which the field @xmath81 , thus @xmath82 , is at most a quadratic extension of @xmath83 .
the case of @xmath84 is classical indeed , @xmath85is the classical modular group , its parabolic fixed points are certainly infinity and the rationals ! @xmath86 are easily derived from this . for the remaining cases , leutbecher s arguments strongly use both the quadratic nature of the fields , and
the fact that their rings of integers are principal ideal domains .
leutbecher s positive result is complemented by the following .
let @xmath87 denote the subset of @xmath71 whose elements may be written as @xmath88 or @xmath89 with @xmath90 $ ] where the pair @xmath91 ( resp .
@xmath92 ) is relatively prime in @xmath93\,$ ] .
( note that this subset can only be proper if @xmath93 $ ] is not a unique factorization domain . )
( wolfart @xcite ) for each @xmath72 @xmath94 equality holds for at most @xmath95 . the first part of our remark on leutbecher s proof , and the fact that any element of @xmath68 has determinant one , shows that the containment holds here . to prove that equality can not hold for all @xmath1
not listed , wolfart adapts an argument of borho and rosenberger @xcite .
informally , one replaces @xmath68 acting on @xmath96 by a homomorphic group that acts on a finite field . for @xmath1 not on wolfart s list , he uses counting arguments to show that this image group can not act transitively on the finite field .
however , this implies intransitivity of the action of @xmath97 .
the case of odd index was resolved by @xcite , who showed that equality does not hold when @xmath98 .
the examples from @xcite given in lemma [ lemcandidates ] are sufficient for our purposes one has that for odd @xmath1 , the orbit of infinity contains all of @xmath96 exactly for @xmath99 . in the even index case , we give explicit ( counter)examples for @xmath100 so as to conclude that leutbecher s list of indices is complete .
any two triangle groups of the same signature are @xmath101-conjugate .
arnoux and hubert @xcite give an explicit conjugation that sends the veech group of @xmath0 to the hecke group @xmath68 when @xmath1 is odd ; when @xmath1 is even , it sends the veech group to a subgroup of index 2 of @xmath97 .
the matrix @xmath102 acts so as to send the hyperbolic triangle of vertices @xmath103 to the triangle of vertices @xmath104 . from this
, one finds that the veech group of the translation surface @xmath105 is exactly the hecke group @xmath97 , or a subgroup of index two . in the remainder of this subsection
, we show that @xmath106 is projectively normalized exactly when @xmath1 is odd .
hence , one can directly use the leutbecher school s results to prove theorem [ main ] in this case .
when @xmath1 is even , we follow the path from @xmath106 to the projectively normalized @xmath107 or @xmath108 , and find that here also their results are germane .
[ propalgsensitiveodd ] suppose that @xmath1 is odd .
then the veech surface @xmath0 has the property that every direction of vanishing saf - invariant is parabolic if and only if the parabolic fixed points of @xmath68 are infinity and the field @xmath80 .
this follows , by a conjugation , from the following lemma .
[ lemheckeoddnormd ] suppose that @xmath1 is odd .
then the translation surface @xmath106 is projectively normalized .
the cotangents of its parabolic direction fill out all of the projective line of its trace field if and only if the orbit of infinity under @xmath68 is all of this projective line .
fix @xmath1 odd .
since @xmath36 lemma [ correctnormalization ] gives that @xmath109 is linearly normalized .
thus , the saf - invariant vanishes for all directions of flow on this surface with cotangent in @xmath110 .
consider now the matrix product @xmath111 since @xmath1 is odd , lemma [ lemcos2x ] certainly implies that @xmath112 .
we deduce that the matrix product is contained in @xmath113 . therefore
, acting as a mbius transformation , it sends @xmath110 to itself .
it thus follows that also on @xmath106 the slopes of directions with vanishing saf - invariant form @xmath110 .
that is , @xmath106 is projectively normalized .
now , the parabolic directions of @xmath106 have as their cotangents the parabolic fixed points of @xmath97 , and these form exactly the orbit of infinity under @xmath97 .
in fact , one can show that @xmath106 is linearly normalized .
[ propalgsensitiveeven ] suppose that @xmath1 is even .
the veech surface @xmath0 has the property that every direction of vanishing saf - invariant is parabolic if and only if the parabolic fixed points of @xmath68 form the set @xmath114 .
this follows from the following two lemmas ; note that although the veech group of @xmath3 is of index two in the conjugated copy of @xmath68 , it has the same set of cusp values ( there is simply a partition of these into two sets , one for each of the cusps of the uniformized hyperbolic surface ) . [ lemdiv4cusps ] suppose that the index @xmath1 is divisible by 4 .
then @xmath115 sends @xmath116 to @xmath117 . for any @xmath1
, lemma [ cos2tansqrd ] implies that @xmath118 since here @xmath1 is a multiple of 4 , lemma [ tangents ] gives that @xmath119 .
thus , @xmath115 sends @xmath120 to @xmath121 since multiplication by any non - zero element of a field defines a surjection . finally , by lemma [ lemcos2x ] , since 4 divides @xmath1 , this field is @xmath122 .
[ lem2mod4cusps ] suppose that the index @xmath1 is congruent to 2 modulo 4 .
then applying @xmath123 sends @xmath124 to @xmath117 .
the product @xmath123 is given in equation , and one easily finds that @xmath123 sends @xmath124 to @xmath125\,.\ ] ] since for any @xmath126 , we have @xmath127 , the result easily follows .
if @xmath1 is odd , then proposition [ propalgsensitiveodd ] and leutbecher s result ( theorem [ thmleut ] above ) give the first part of theorem [ main ] . the proposition combined with wolfart s and seibold s results
complete the proof in this odd index case .
when @xmath1 is even , an appropriate complement to wolfart s result is provided by examples in the next section , showing that leutbecher had indeed already listed all @xmath1 for which @xmath128 contains all of @xmath129 .
combining this with proposition [ propalgsensitiveeven ] gives theorem [ main ] in this even index case .
since @xmath130 when @xmath1 is odd , leutbecher s theorem here exactly addresses the question of when all elements of the field are parabolic fixed points of @xmath97 . for these odd @xmath1 , any element of the field @xmath131 that is not in @xmath128 corresponds to a non - parabolic direction that lies in the trace field of @xmath106 and thus has vanishing saf - invariant .
explicit examples include those given in subsection [ subodd ] by rosen and towse @xcite , and by towse _
et al _ @xcite .
rosen and towse only treat small index cases , and for even @xmath1 towse _ et al _ only discuss ( the numerous ) @xmath68-orbits of @xmath132 , whereas we are interested in the action when restricted to @xmath133 .
we give new examples of fixed points of special hyperbolic matrices , and thus also of special affine pseudo - anosov diffeomorphisms , in subsection [ subeven ] .
[ lemcandidates](rosen - towse @xcite , towse _ et al _ @xcite ) each of the following elements is not in the orbit of infinity under its respective hecke group . *
@xmath134 * @xmath135 [ corcandidates ] the above examples are fixed points of special hyperbolic matrices . in each of @xcite and @xcite ,
the examples referred to above are shown to have periodic ( rosen ) @xmath136-continued fraction expansion .
now , any real number of periodic expansion is easily shown to be the fixed point of a matrix in the corresponding hecke group .
rosen @xcite showed that ( 1 ) for each @xmath62 , each real number @xmath137 has a unique @xmath136-continued fraction expansion , and ( 2 ) the finite length @xmath136-continued fraction expansions are in one - to - one relationship with the parabolic fixed points of the corresponding hecke group .
therefore , any periodic expansion corresponds a real fixed point of a non - parabolic matrix .
thus , a real number of periodic expansion is indeed the fixed point of a hyperbolic element .
some of the examples of towse _ et al . _
were found by also using a variant of the borho - rosenberger arguments . in specific cases
, one can identify infinite classes of elements in @xmath96 that can not be in the orbit of infinity .
these elements can then be examined as to whether they are fixed points of ( special ) hyperbolic elements ( as all of the examples above ) , or not .
note also @xcite conjecture that for every @xmath138 there are elements of @xmath96 that are fixed by no elements of @xmath97 .
they also report having searched without success for ( in the present terminology ) hyperbolic fixed points in @xmath96 for odd @xmath1 in the range from @xmath139 to @xmath140 . based on this
, they indicate some doubt as to the existence of such fixed points for @xmath141 .
towse and his various co - authors focused on the question of expansions of elements of @xmath142 , but we are interested in properties of elements of @xmath143 .
thus , we further examine this even index setting .
we give explicit examples of special hyperbolic matrices for @xmath144 .
note that this also establishes that the corresponding @xmath0 have special affine pseudo - anosov diffeomorphisms .
we first give an example for the smallest even index for which this is possible .
[ q14candidate ] fix @xmath145 and @xmath146 .
let @xmath147 and @xmath148 .
let @xmath137 be the attractive fixed point of @xmath149 then @xmath150 and thus the conjugate of @xmath28 by @xmath151 is a special hyperbolic matrix . for any @xmath152
we find that a matrix of this form gives @xmath153 and is thus certainly hyperbolic , and fixes @xmath154 now , the minimal polynomial over @xmath11 of @xmath155 is @xmath156 and we find that @xmath157 .
thus , @xmath158 we conclude that @xmath159 and hence the result now follows from lemma [ lem2mod4cusps ] . in briefer form , we give examples for the three indices for which wolfart s result does not immediately imply the intransitivity of @xmath68 on @xmath160 .
[ q18candidate ] let @xmath161 , @xmath162 and let @xmath163 . then @xmath164 the matrix @xmath28 fixes the point @xmath165 the above is verified by direct calculation . [ q20candidate ]
let @xmath166 , and @xmath167 and let @xmath168 then @xmath169 and @xmath170 and @xmath28 fixes the point @xmath171 furthermore , @xmath172 and thus the conjugate of @xmath28 by @xmath115 is a special hyperbolic matrix . again , verification of the claimed equalities is straightforward .
lemma [ lemdiv4cusps ] finishes the proof .
whereas the special hyperbolics as above were found by searching amongst short words in our chosen generators , the next example was found by searching for elements of @xmath173 with periodic rosen continued fraction expansions .
indeed , @xmath174 is in the @xmath175-orbit of the fixed point of @xmath28 below .
thus , of course it too is fixed by a special hyperbolic matrix , and hence has eventually periodic expansion . however , as the preperiod is quite long , we prefer to give @xmath176 corresponding directly to the period .
[ q24candidate ] fix @xmath177 and @xmath178 .
let @xmath179 and @xmath180 then @xmath181 is a special hyperbolic matrix .
the minimal polynomial of @xmath182 is @xmath183 .
taking polynomial remainders with respect to the modulos @xmath184 , one thus finds that @xmath185 with @xmath186 one directly verifies that @xmath28 has trace ( much ! )
larger than 2 , and fixes @xmath187 and thus @xmath28 is special . our examples for @xmath188 complement wolfart s ( and seibold s ) results and thus complete the first part of the leutbecher program it is exactly for the list in theorem [ thmleut ] that the orbit of infinity under @xmath68 contains all of @xmath189 .
a second , yet more difficult , goal is to arithmetically characterize the cusp set for any @xmath62 .
note that this second goal is also of great interest in our setting one would like to precisely locate the set of parabolic directions amongst all those of vanishing saf - invariant .
it is also completely motivated in the initial setting : the rosen @xmath136-continued fractions define a numeration system ; one of the initial problems in the study of any enumeration system is to determine the set enumerated by finite length expansions ( for these continued fractions , this is indeed the orbit of infinity , as rosen showed ) .
leutbecher and his school were well aware that the orbit of infinity under @xmath68 can contain all of @xmath190 only if the ring of integers of @xmath191 is principal . with this as hindsight
, one can easily recognize that @xmath0 with non - parabolic directions with vanishing saf - invariant _ must _ arise .
for ease of discussion , let us consider the odd index case .
then @xmath68 is a subgroup ( of infinite index when @xmath192 ) of the hilbert modular group of the totally real field @xmath193 .
this hilbert modular group acts on the field @xmath194 union infinity , the number of orbits equals the class number of the ring of integers of the field , @xmath195 for this classical result , see say @xcite .
of course , no subgroup could act so as to give fewer orbits .
however , @xmath68 has exactly one cusp , corresponding to one orbit of field elements under this group .
thus , as soon as the class number of the field exceeds two there are certainly elements of the field that are not in the cuspidal orbit of this group . using the normalization as above
, one deduces that then @xmath0 has non - parabolic directions with vanishing saf - invariant .
now , although the class numbers of fields of the form @xmath196 are `` notoriously difficult to compute '' @xcite , one does know that for @xmath197 the above argument must give an example : van der linden @xcite shows that both @xmath198 , with equality assuming the generalized riemann hypothesis ; and for all primes @xmath199 , @xmath200 , under the same assumption .
tetal p. arnoux , _
changes dintervalles et flots sur les surfaces _ , pp .
538 in _ ergodic theory ( sem .
, les plans - sur - bex , 1980 ) _ , monograph .
, 29 , univ .
genve , geneva , 1981 .
, thse de 3@xmath201 cycle , universit de reims , 1981 .
p. arnoux and p. hubert , _
fractions continues sur les surfaces de veech _
, j. anal . math .
81 ( 2000 ) , 3564 .
p. arnoux and j .- c .
yoccoz , _ construction de diffomorphismes pseudo - anosov _ , c. r. acad .
paris sr .
i math . 292
( 1981 ) , no .
1 , 7578 .
w. borho and g. rosenberger , _ eine bemerkung zur hecke - gruppe @xmath202 _ , abh .
hamburg 39 ( 1973 ) , 8387 .
k. calta , _ veech surfaces and complete periodicity in genus 2 _ , j. amer . math .
soc . , 17 ( 2004 ) , no .
4 , 871908 .
k. calta and j. smillie , _ algebraically periodic translation surfaces _ , j. mod .
dyn . 2 ( 2008 ) , no .
2 , 209248 . c.j .
earle and f.p .
gardiner , _ teichmller disks and veech s @xmath203-structures _ , pp .
165 189 , in _ extremal riemann surfaces _ ,
201 , a. m. s. , providence , r.i . , 1997 .
g. van der geer , _ hilbert modular surfaces _ , ergebnisse math .
3 . folge bd . 16 , 1988 . e. gutkin and c. judge , _ affine mappings of translation surfaces : geometry and arithmetic _ , duke math .
j , 103 ( 2000 ) pp . 191213 .
e. hanson , a. merberg , c. towse , and e. yudovina , _ generalized continued fractions and orbits under the action of hecke triangle groups _
acta arith . 134
( 2008 ) , no .
4 , 337348 . p. hubert , e. lanneau and m. mller , _ the arnoux - yoccoz teichmller disc _ , geom .
18 ( 2009 ) , 19882016 .
p. hubert and t. a. schmidt , _ infinitely generated veech groups _
, duke math .
j. 123 ( 2004 ) , 49 69 .
r. kenyon and j. smillie , _ billiards in rational - angled triangles _
, comment .
* 75 * ( 2000 ) , 65108 .
d. h. lehmer , _ questions , discussions , and notes : a note on trigonometric algebraic numbers _ ,
monthly 40 ( 1933 ) , no .
3 , 165166 .
a. leutbecher , _ ber die heckeschen gruppen @xmath202 .
211 ( 1974 ) , 6386 .
d. long and a. reid , _ pseudomodular surfaces _ , j. reine angew .
( 2002 ) , 77100 .
f. j. van der linden , _ class number computations of real abelian number fields _ , math . comp .
39 ( 1982 ) , no .
160 , 693707 . c. mcmullen , _ billiards and teichmller curves on hilbert modular surfaces _ , j. amer
( 2003 ) , 857885 . , _ teichmller geodesics of infinite complexity _ , acta math . 191
( 2003 ) , no .
2 , 191223 .
d. rosen , _ a class of continued fractions associated with certain properly discontinuous groups _ , _ duke math .
j. _ 21 ( 1954 ) , 549 - 563 .
d. rosen , c. towse , _ continued fraction representations of units associated with certain hecke groups _
, arch . math .
( basel ) 77 ( 2001 ) , no .
4 , 294302 .
j. schmurr , _ translation covers of triangular billiards surfaces _ , preprint : arxiv:0810.3353 .
r. schoof _ class numbers of real cyclotomic fields of prime conductor _ , math . comp .
72 ( 2003 ) , no .
242 , 913937 f. seibold , _ zahlentheoretische eigenschaften der heckeschen gruppen @xmath204 und verwandter transformationsgruppen _ , dissertation , technische universitt mnchen , 1985 . w. thurston , _ on the geometry and dynamics of diffeomorphisms of surfaces
a.m.s . 19 ( 1988 ) , 417 431 . w.a
veech , _ teichmuller curves in modular space , eisenstein series , and an application to triangular billiards _ , inv . math .
97 ( 1989 ) , 553 583 .
j. wolfart , _ eine bemerkung ber heckes modulgruppen _ , arch . math .
( basel ) 29 ( 1977 ) , no .
1 , 7277 .
there is some minor confusion in the literature about containments amongst fields defined over @xmath11 by adjoining values of sines and cosines .
for example , even the redoubtable computational number theorist d. h. lehmer published inaccurate results on this matter @xcite .
( that there is a flaw in his work is highlighted by a displayed table indicating that there is no integer @xmath205 such that the algebraic number @xmath206 is of degree three : both @xmath207are of this type . ) we address this here , of course we acknowledge that these elementary results are well - known . @xmath217 = \begin{cases } 2 \cdot [ \ , \mathbb q(\cos \dfrac{2 \pi}{q})\,:\,\mathbb q\ , ] & \text{if}\ ; 4 \not{\vert } \ , q \,;\\ \\ \dfrac{1}{2}\cdot [ \ , \mathbb q(\cos \dfrac{2 \pi}{q})\,:\,\mathbb q\ , ] & \text{if}\ ; q \equiv 4 \bmod 8\,;\\ \\
\mathbb q(\cos \dfrac{2 \pi}{q})\,:\,\mathbb q\ , ] & \text{if}\ ; 8 \ ; \text{divides}\ ; q\ , . \end{cases}\ ] ] recall that for any @xmath219 letting @xmath220 gives @xmath221 = \phi(n)\,$ ] , where as usual @xmath222 denotes the euler totient function of @xmath223 .
but , @xmath224 and also @xmath225 is the fixed field of complex conjugation on @xmath226 .
the previous four lemmas combine to give the proof of proposition [ propcompdeg ] .
indeed , @xmath242 takes the values @xmath243 , @xmath244 , @xmath1 , @xmath245 and @xmath246 respectively in the five cases for @xmath1 of the previous lemma .
thus @xmath247 takes the value : @xmath248 when @xmath1 is either odd , or is @xmath249 ; @xmath250 when @xmath251 divides @xmath62 ; and @xmath252 in the final two cases , as there @xmath253 . for any field @xmath259 and any @xmath137 ,
we have that @xmath260 \in \{1 , 2\}\,$ ] .
thus , lemma [ cos2tansqrd ] shows that the bottom row of field extensions are correctly indicated .
now , clearly each of @xmath261 and @xmath262 is contained in @xmath263 .
using @xmath258 , and its quotients by @xmath264 and @xmath265 respectively , shows that also the top row of field extensions is valid .
recall that @xmath280 gives a @xmath11-vector space basis of @xmath226 .
thus , give @xmath281 , there are @xmath282 such that @xmath283 .
now , in our arguments above , we can replace @xmath284 by any of these @xmath285 .
thus , when @xmath238 divides @xmath1 , the first two parts of theorem [ thmtrigfldcont ] imply that @xmath286 , and the first part of this lemma follows .
+ recall that @xmath287 where @xmath288 is a chebyshev polynomial of the second type .
since the @xmath288 are polynomials with integer coefficients , the second part of the lemma also follows . | veech s original examples of translation surfaces @xmath0 enjoying what mcmullen has dubbed `` optimal dynamics '' arise from appropriately gluing sides of two copies of the regular @xmath1-gon , with @xmath2 .
we show that every @xmath0 whose trace field is of degree greater than 2 has non - periodic directions of vanishing saf - invariant .
( calta - smillie have shown that under appropriate normalization , the set of slopes of directions where this invariant vanishes agrees with the trace field . )
furthermore , we give explicit examples of pseudo - anosov diffeomorphisms whose contracting direction has zero saf - invariant . in an appendix , we prove various elementary results on the containment of trigonometric fields . |
ground - based spectroscopic observations allow one to study doppler shifts of scattered solar fraunhofer lines in the zodiacal light ( e.g. , james 1969 , hicks et al .
1974 , fried 1978 , and east & reay 1984 ) .
analysis of these shifts thus provide an opportunity to explore velocities of interplanetary dust in the inner solar system . using the wisconsin h - alpha mapper ( wham ) spectrometer ,
reynolds et al .
( 2004 ) were the first to obtain accurate measurements of the centroid velocities and line profiles of the scattered solar mg i @xmath05184 absorption line in the zodiacal light , both along the ecliptic equator and at high ecliptic longitudes .
the models for migration of dust particles are discussed in section 2 .
these results are based on integrations of the orbital evolution of dust particles developed by ipatov et al .
( 2004 ) and ipatov & mather ( 2006a - b ) ( all our recent papers can be found at astro - ph and at http://www.astro.umd.edu/$\sim$ipatov ) . in sections 3 and 4 , based on the results of our runs ,
we consider our model calculations of the radial velocity profile of the scattered mg i line .
we compare plots of centroid velocities and line widths of the mg i line versus solar elongation with wham observations obtained by reynolds et al .
studies of typical eccentricities and inclinations of zodiacal dust particles that fit the wham observations are presented in section 5 . in previous papers ,
observations have been compared with analytical models , but this is the first comparison with numerical integrations of the particles .
our studies of the mg i line shifts use the results of following the orbital evolution of about 15,000 asteroidal , cometary , and trans - neptunian dust particles under the gravitational influence of planets ( excluding pluto for asteroidal and cometary particles ) , the poynting - robertson drag , radiation pressure , and solar wind drag .
results of some of these integrations were presented by ipatov et al .
( 2004 ) and ipatov & mather ( 2006a - b ) . as in liou et al .
( 1999 ) and moro - martin & malhotra ( 2002 ) , we assume the ratio of solar wind drag to poynting robertson drag to be 0.35 . in this section
we describe the model used for our studies of the migration of dust particles . in our runs for asteroidal and cometary particles ,
the values of @xmath1 , the ratio of the sun s radiation pressure force to gravitational force , varied from @xmath20.0004 to 0.4 .
burns et al .
( 1979 ) obtained @xmath1=@xmath3 , where @xmath4 is the particle s density in grams per cubic centimeter , @xmath5 is its radius in micrometers , and @xmath6 is the radiation pressure coefficient , which is close to unity for particles larger than 1 @xmath7 m . for silicates at density of 2.5 g/@xmath8 ,
the @xmath1 values 0.004 , 0.01 , 0.05 , 0.1 , and 0.4 correspond to particle diameters @xmath9 of about 120 , 47 , 9.4 , 4.7 , and 1 microns , respectively . for water ice
, @xmath9 is greater by a factor of 2.5 than that for silicate particles .
the orbital evolution of dust particles was studied for a wider range of masses ( including particles up to several milimeters ) of asteroidal and cometary particles than in the papers by other authors ( e.g. , dermott et al .
2001 , 2002 ; gorkavyi et al .
2000 ; grogan et al . 2001 ; kortenkamp & dermott 1998 ; liou et al .
1995 , 1996 , 1999 ; liou & zook 1999 ; moro - martin & malhotra 2002 , 2003 ; ozernoy 2001 ; reach et al .
smaller particles have a larger surface per unit of mass of particles .
it has been proposed ( grn et al . 1985 ) that the main contribution to the zodiacal light is from particles that range from 20 to 200 @xmath7 m in diameter ( for silicate particles , this range corresponds to @xmath1 about 0.002 - 0.02 ) . according to grn et al .
( 2000 ) , the lorentz force is comparable to solar gravitational interaction for particles of @[email protected] @xmath7 m at 1 au and of @xmath9@xmath101 @xmath7 m at 50 au .
interstellar particles dominate among such small particles , and are not believed to be significant contributers to the zodiacal light .
we considered mainly larger interplanetary particles , so we did not include the lorentz force in our runs .
migration of dust particles was integrated using the bulirsh - stoer method ( bulsto ) with the relative error per integration step less than @xmath11 .
the swift integration package by levison & duncan ( 1994 ) was modified to include the additional forces of radiation pressure , poynting - robertson drag , and solar wind drag .
the integration continued until all of the particles either collided with the sun or reached 2000 au from the sun . for small @xmath1 ,
lifetimes exceeded 50 - 80 myr ( 240 myr for trans - neptunian particles ) . in each run ( with a fixed source of particles and
@xmath1=const ) we took @xmath12@xmath2250 particles , because for @xmath12@xmath13500 the computer time per calculation for one particle was several times greater than for @xmath12=250 . the total number of particles in several tens of runs ( with different @xmath1 and different sources of particles ) was about 15,000 . in our runs ,
orbital elements were stored with a step of @xmath14 of 20 yr for asteroidal and cometary particles and 100 yr for trans - neptunian particles during all considered time intervals .
the stored orbital elements of all particles during their dynamical lifetimes were then used in our studies of the velocities of particles presented in the next section .
the planets were assumed to be material points .
however , using orbital elements obtained with a step @xmath14 , ipatov and mather ( 2006a ) calculated the mean probability of a collision of a particle with the planet during the particle lifetime .
the total probability of collisions of all particles with planets in one run is small .
therefore , in most runs the results obtained will be exactly the same whether or not we consider collisions of particles with planets during integration .
even if there were a particle that actually collides with a planet , the sum of dynamical lifetimes of all particles and the distribution of particles over their orbital elements during their dynamical lifetimes will be practically the same as in the model for which planets are considered as material points .
destruction of particles due to their collisions with other particles is also not considered in the present paper .
the destruction affects mainly lifetimes of particles and their size distributions at different distances from the sun , but not the distribution of the particles orbital elements during their migration in the zodiacal cloud . in future models , which will consider the size distributions of particles , we plan to take into account the destruction of particles .
the initial positions and velocities of asteroidal particles used in our models were the same as those of the first @xmath12 numbered main - belt asteroids ( jdt 2452500.5 ) , i.e. , dust particles are assumed to leave the asteroids with zero relative velocity .
the initial positions and velocities of the trans - neptunian particles were the same as those of the first trans - neptunian objects ( jdt 2452600.5 ) .
the initial positions and velocities of cometary particles were the same as those of comet 2p / encke ( @[email protected] au , @[email protected] , @xmath18@xmath1612@xmath19 ) , or comet 10p / tempel 2 ( @[email protected] au , @[email protected] , @xmath18@xmath1612@xmath19 ) , or comet 39p / oterma ( @[email protected] au , @[email protected] , @xmath18@xmath162@xmath19 ) , or long - period comets ( @xmath17=0.995 , @xmath20=@xmath15@xmath21(1-@xmath17)=0.9 au or @xmath20=0.1 au , @xmath18 varied from 0 to 180@xmath19 in each run , particles started at perihelion ; these runs are denoted as ` lp ' runs ) .
we considered encke particles starting near perihelion ( runs denoted as 2p ) , near aphelion ( runs denoted as ` 2p 0.5 t ' ) , and when the comet had orbited for @xmath22 after perihelion passage , where @xmath23 is the period of the comet ( runs denoted as ` 2p 0.25 t ' ) . note that semimajor axes and eccentricities of dust particles differed from those of parent bodies , but inclinations were the same . for cometary particles ,
the initial value of time @xmath24 when perihelion was passed was varied for different particles with a step @xmath25=1 day or @xmath25=0.1 day near the actual value of @xmath24 for the comet .
comet 10p is an example of a typical jupiter - family comet moving inside jupiter s orbit ; comet 39p moves outside jupiter s orbit .
comet 2p is the only known high - eccentricity comet with aphelion distance @[email protected] au .
it comes close to the sun and produces a lot of dust .
our initial data for dust particles are different from those in previous papers , and the parent comets producing dust are different ( exclusive of comet 2p ) from those considered earlier .
we calculated how the solar spectrum was changed after the light was scattered by the dust particles and observed at earth .
this was carried out by first considering all orbital elements of dust particles during a single run , which were stored in computer memory with a step @xmath28 ( see the previous section ) .
based on these stored orbital elements , we calculated velocities and positions of particles and the earth during the dynamical lifetimes of the particles . for each pair of positions of a particle and the earth ,
we then calculated many ( @xmath1010@xmath29 - 10@xmath30 , depending on a run ) different positions of the particle and the earth during the period @xmath31 of revolution of the particle around the sun , assuming that orbital elements do not vary during @xmath31 .
this number is smaller for runs with larger maximum dynamical lifetimes of particles .
the model , which is based on all positions and velocities of dust particles during their dynamical lifetimes , represents the zodiacal cloud for the case when small bodies continuously produce dust at a constant rate along their orbits ; seasonal effects and jumps in production of dust are not considered in this model .
the model allows one to study the principal differences between spectra corresponding to particles produced by asteroids , comets , and trans - neptunian objects .
more complicated models to study the fractions of particles of different origin can be considered in the future .
the choice of a scattering function was based on analyses of dependences of scattering functions on angles @xmath32 and @xmath33 ( see below ) and wavelength presented in several papers ( e.g. , giese 1963 ; giese & dziembowski 1969 ; leinert 1975 ; leinbert et al .
1976 ; weiss - wrana 1983 ; hong 1985 ; lamy & perrin 1986 ) , which mainly followed the mie theory for scattering .
the scattering function depends on the composition of particles , their sizes , and other factors . however , we considered three simple scattering functions : ( 1 ) @xmath34 for @xmath35 , and @xmath36(@xmath37)@xmath29 for @xmath38 , where @xmath32 is the angle between the earth and the sun , as viewed from the particle in radians and @xmath39/3 radian ; ( 2 ) @xmath40 having the same functional form as above , but with @xmath32 replaced by elongation @xmath33 , the angle between the particle and the sun , as viewed from the earth ( eastward from the sun ) ; ( 3 ) isotropic scattering .
for all three functions , the intensity @xmath41 of light that reaches the earth was considered to be proportional to @xmath42@xmath21@xmath43 , where @xmath44 is the distance between the particle and the earth , @xmath45 is the distance between the particle and the sun , and @xmath0 is the wavelength of light .
since we considered the scattering near a single spectral line , wavelength @xmath0 in our runs was essentially a constant .
except for lines of sight close to the sun , these three scattering functions give virtually the same results ( see below ) .
since the differences between the functions that we considered were much greater than the differences between scattering functions presented in the publications cited above , we conclude that we need not worry about the precise form of the scattering . for each particle position , we calculated the velocity relative to the sun , the velocity relative to the earth , and its respective distance to the earth and to the sun , @xmath44 and @xmath45 . these parameters and the scattering function
were then used to construct the solar spectrum received at the earth after light had been scattered by all the particles located within some limited solid angle ( beam ) along a line of sight from the earth .
the direction of the beam is characterized by its eastern elongation from the sun @xmath33 , and by its inclination above the ecliptic plane .
particles in beams of diameter 2@xmath19 ( fig . 1 ) and 2.5@xmath19 ( other figures ) were considered .
projections of relative velocities of particles on the lines of sight from the earth and the sun to particles were used for analysis of the doppler shift . in each run , particles all had the same size , the same @xmath1 , and the same source ( e.g. , asteroidal ) .
figure 1 shows sample spectra of scattered mg i @xmath05184 obtained from these calculations toward sightlines in the antisolar direction ( fig .
1a , c ) and toward the ecliptic pole ( fig .
the spectra consist of intensity vs. wavelength shift @xmath46 with respect to 5183.62 angstrom .
the thinnest line in figure 1 denotes the initial ( unscattered ) solar spectrum . for asteroidal ( ast ) and comet encke ( ` 2p ' ) particles , the solar spectrum received at the earth is found to be nearly independent of the scattering function , while for trans - neptunian ( ` tn ' ) particles , the results for scattering function ( 3 ) , denoted as ` tn3 ' , differ slightly from ( 1 ) and ( 2 ) , denoted as ` tn 1 ' and tn 2 , respectively . in the legend in the figure , the first number ( 0.2 in fig .
1a , c ) denotes @xmath1 , and the next number in figures 1a , c denotes elongation @xmath33 ( in degrees ) .
the wham observations are also presented for comparison . these observations and all other plots in figure 1c ,
d were stretched so that the minimum in the line was at approximately the same depth as that for the initial solar spectrum .
the continuum levels were also set to be the same ( equal to 1 ) for all plots .
similar plots at @xmath33=90@xmath19 and @xmath33=270@xmath19 for zero inclination above the ecliptic plane were presented by ipatov et al .
( 2005 ) . unlike results by clarke et al .
( 1996 ) who considered spectrum near 4861 angstrom ( @xmath47 line of hydrogen ) , our modeled spectra do nt exhibit strong asymmetry .
we similary found that minima in the plots of dependencies of the intensity of light on its wavelength near 5184 angstrom are not as deep as those for the initial solar spectrum .
note that the value of @xmath33 in the paper by clarke et al .
( 1996 ) is measured in the opposite clockwise direction than in our present paper , so our 90@xmath19 corresponds to 270@xmath19 in their paper . at the north
ecliptic pole , the calculated spectrum was shifted slightly to the left relative to the solar spectrum for asteroidal particles and slightly to the right ( to the red ) for particles originating from comet 2p ( fig .
these shifts may be due to small asymmetries in the model particle distributions with respect to the ecliptic plane .
the spectra of 10p and 39p particles and those from long - period comets were very similar to each other .
for cometary particles , the line profile has a flatter bottom than that for asteroidal particles , but it was not so wide as the observed spectrum presented in ( reynolds et al .
none of our model runs matched the large width of the observation toward the ecliptic pole .
this issue will not be addressed in this paper , but will be a topic for future investigation .
the calculated model spectra along the ecliptic plane are in general agreement with the observations . for different values of @xmath33 , based on the model spectra , which use the distribution of velocities and positions of dust particles in our run
, we determined the doppler shift of the scattered mg i line with respect to the centroid wavelength of the unscattered solar profile .
the plot of this characteristic velocity vs. the solar elongation @xmath33 along the ecliptic plane is called the ` velocity - elongation ' plot . for ` velocity - elongation ' plots marked by @xmath48 , we considered the shift of the centroid ( the ` center of mass ' of the line ) , while ` velocity - elongation ' plots marked by @xmath49 denote the shift of the minimum of the line . `
velocity - elongation ' plots for different scattering functions are denoted as @xmath50 and @xmath51 for the scattering function 1 , as @xmath52 and @xmath53 for the function 2 , and as @xmath54 and @xmath55 for the function 3 .
the lines in figure 1 are nearly symmetric , so the results for ` c ' and ` m ' in figure 2 differ only a little .
brief discussions of ` velocity - elongation ' plots obtained in our runs with observations have been made by ipatov et al .
( 2005 , 2006 ) , madsen et al .
( 2006 ) , and ipatov and mather ( 2006a ) .
below we compare the ` velocity - elongation ' plots in more details .
the details of the model spectra depend on @xmath33 , @xmath1 , @xmath18 , and the source of particles .
different particles populations produce clearly distinct model spectra of the zodiacal light .
the ` velocity - elongation ' plots obtained for different scattering functions were similar at 30@xmath19@xmath27@xmath33@xmath27330@xmath19 ( fig .
2 ) ; though the difference was greater for directions close to the sun .
some jumps in the plots are caused by small number statistics for these runs . in figures 3 - 8 the results
were obtained using only the second scattering function . a comparison of the observed ` velocity - elongation ' plot with those obtained from our model for dust particles of different sizes ( i.e. , different values of @xmath1 ) produced by asteroids , comets ( 2p / encke , 10p / tempel 2 , 39p / oterma , long - period ) , and trans - neptunian objects allowed us to draw some conclusions about the sources of zodiacal dust particles .
asteroidal , trans - neptunian , and 2p particles populations produce clearly distinct model spectra of the zodiacal light .
the differences between ` velocity - elongation ' plots for several sources of dust reached its maximum at @xmath33 between 90@xmath19 and 120@xmath19 ( figs .
36 ) . for future observations of velocity shifts in of the zodiacal spectrum
, it will be important to pay particular attention to these elongations .
we consider the amplitude of ` velocity - elongation ' curves as @xmath56=(@xmath57-@xmath58)/2 , where @xmath58 and @xmath57 are the minimum and maximum values of velocities at 90@xmath19@xmath27@xmath33@xmath27270@xmath19 .
the observational value of @xmath56 is about 12 km / s ( if we smooth the plot ) . for several dust sources , the characteristic values of @xmath56 , @xmath58 , and @xmath57
are presented in table 1 .
mean eccentricities @xmath59 and mean inclinations @xmath60 at distance from the sun 1@xmath2@xmath45@xmath23 au are also included in the table . for asteroidal dust ,
the ` velocity - elongation ' plots had lower amplitudes than the observations ( fig .
b ) . the plots obtained at different @xmath1 differed little from each other , especially at @[email protected] . for 10p particles , the ` velocity - elongation ' amplitudes
were also lower than that of the observations ( fig .
the difference between the plots obtained for 10p particles at different @xmath1 was greater than that for asteroidal particles , but was usually less than that for other sources of particles considered .
on the other hand , the ` velocity - elongation ' curve corresponding to particles produced by comet 2p have slightly larger amplituides than the observed curve ( fig .
so perhaps a combination of cometary 2p dust particles and asteroidal particles could provide a result that is close to the observational ` velocity - elongation ' curve .
the velocity amplitudes @xmath56 for particles originating from long - period comets are much greater than those for the observational curve ( fig .
the orbit of comet 39p is located outside jupiter s orbit .
studies of the migration of 39p particles thus give some information about the the migration of particles originating beyond neptune s orbit that have reached the orbit of jupiter . for 39p particles and 0.01@xmath2@[email protected] at 60@xmath19@xmath27@xmath33@xmath27@xmath61 , ` velocity - elongation ' amplitudes were smaller than those for the observations ( fig .
5a ) , while for @[email protected] , the curves more closely match the observations ( fig .
5b ) . for such small @xmath1 ,
only a small number of particles entered inside jupiter s orbit , and statistics were poor .
the distribution of particles over their orbital elements could be somewhat different if we had considered a greater number @xmath12 of particles in one run , but the difference between the curves obtained at different @xmath12 will not be more than the difference between the curves obtained at adjacent values of @xmath1 presented in the figure . at @xmath1=0.0001 , particles spent on average more time in the zodiacal cloud than for other runs , but it was due mainly only to one particle , which during about 9 myr moved inside jupiter s orbit before its collision with the sun .
in reality such a particle could sublimate or be destroyed in collisions during its dynamical lifetime .
small number statistics may also be responsible for the spread of results for the trans - neptunian dust in figure 6a .
if there had been a greater number of particles in the runs , the differences between the corresponding plots would probably have been smaller .
there was a zero vertical shift for the observational plot .
plots for asteroidal , 10p , and 39p particles are shifted a little down , and that for 2p particles was shifted up relative to the observational plot .
so a combination of different sources of particles could give a zero vertical shift .
for trans - neptunian dust particles , velocity amplitudes @xmath56 were greater than the observational values at @[email protected] and were about the same at @xmath1=0.01 ( fig .
so together with cometary particles , trans - neptunian particles can compensate small values of @xmath56 for asteroidal particles . the difference between different plots in fig . 6a
may be caused partly by small statistics , as only a few trans - neptunian particles in each run ( at fixed @xmath1 ) entered inside jupiter s orbit .
the observational plot was mainly inside the region covered by trans - neptunian plots obtained for different @xmath1 , but at 180@xmath19@xmath27@xmath33@xmath27270@xmath19 it was mainly above the trans - neptunian curves .
in the future we plan to explore the fractions of particles of different origin in the overall dust population based on various observations and taking into account a model for the size distribution of particles . here
we present estimates based on a much simpler , two - component zodiacal dust cloud that fits the observations .
for example , with a velocity amplitude @xmath56=9 km / s for asteroidal dust and at @xmath56=14 km / s for 2p particles , the fraction @xmath62 of asteroidal dust would need to be 40% . if all of the cometary particles in the zodiacal dust were from long - period comets ( @xmath56=33 km / s ) , then @xmath62 = 88% .
the contribution of _ lp _ particles to the zodiacal light can not be large also because their inclinations are large and iras observations showed ( liou et al .
1995 ) that most of the zodiacal light is due to particles with inclinations @xmath63@xmath2730@xmath19 . at @[email protected] ,
_ lp _ particles are quickly ejected from the solar system , so among zodiacal dust we can find only _ lp _ particles greater than 100 @xmath7 m .
the contribution of _ lp _ particles to the total mass of the zodiacal cloud is greater than their contribution to the brightness @xmath41 , as a surface of a particle of radius @xmath64 is proportional to @xmath65 , and its mass @xmath66 is proportional to @xmath67 ( i.e. , @xmath68@xmath69@xmath64 ) .
dynamical lifetimes of _ lp _ particles at @[email protected] ( i.e. , at diameters greater than 200 @xmath7 m ) can exceed several myrs ( i.e. , exceed mean lifetimes of asteroidal and 2p particles ) , so the fraction of _ lp _ particles in the zodiacal cloud can be greater than their fraction in the new particles that were produced by small bodies or came from other regions of the solar system . dynamical lifetimes of dust particles are greater for smaller @xmath64 ( ipatov and mather 2006a ) , and some particles can be destroyed in collisions with other particles .
so the mass distributions of particles produced by small bodies are different from the mass distributions of particles located at different distances from the sun .
a significant fraction of cometary dust in the near - earth space was proposed by liou et al .
( 1995 ) and zook ( 2001 ) .
liou et al .
( 1995 ) based his estimates of the fraction of cometary dust ( 3/4 to 2/3 ) on comparison of the migration of 9 @xmath7 m diameter dust particles that produced by comet encke - type parent bodies with the iras observations of the shape of the zodiacal cloud .
based on cratering rates from an ensemble of earth- and lunar - orbiting satellites , zook ( 2001 ) estimated that the cometary contribution to the near - earth flux of particles is @xmath1075% .
ozernoy ( 2001 ) compared the computer simulation results of migration of 1 @xmath7 m and 5 @xmath7 m dust particles with the cobe / dirbe data and showed that the trans - neptunian dust contributes as much as @xmath101/3 of the total number density near the earth .
grogan et al . ( 2001 ) suggested that at least 30% of zodiacal dust comes from the break - up of asteroids in order to explain formation of dust bands .
the conclusions on a considerable fraction of cometary dust are also in agreement with earlier studies of the dynamics of jupiter - family comets ( ipatov & mather 2003 , 2004a - b , 2006a ) , which showed that some former cometary objects could get high eccentric orbits located entirely inside jupiter s orbit and move in such orbits for a long time
. some of these objects could disintegrate producing a substantial amount of dust .
comparison of the number density of migrating dust particles with the observation that number density is constant at 3 - 18 au from the sun ( ozernoy 2001 , landgraf 2002 , ipatov & mather 2005 , 2006a ) is also consistent with that a fraction of cometary dust particles is significant ( perhaps dominant ) inside saturn s orbit .
observations by reynolds et al .
( 2004 ) also provided the fwhm ( full width at half maximum ) of the mg i line in the zodiacal light . in figure 7 , their values are marked by crosses .
most of the observed values were between 70 and 80 km / s , with a mean value of 76.6 . in figure 7 , we compare these observations with the fwhm values obtained from our model line profiles .
the values are different for different @xmath1 and different sources of the dust . for particles started from asteroids , comets 2p , 10p and 39p , and long - period comets ,
the mean ( at 30@xmath19@xmath27@xmath33@xmath27330@xmath19 ) values of fwhm are about 74 , 81 - 88 , 76 - 77 , 76 - 77 , 73 - 86 km / s , respectively . for 2p particles ,
the values of fwhm depend on @xmath1 and the place of origin from the orbit .
the mean values of fwhm obtained for particles started from long - period comets can differ considerably in different runs with different @xmath1 .
discrete values of the width in figure 7 are caused by a discrete model of spectra .
as observational values of the width can differ by 10 km / s for close values of @xmath33 , there is no need to calculate accurately the width in our models . at @xmath1@xmath2@xmath70 for particles started from comet 2p , the width is greater at @xmath33@xmath16220@xmath19 than at @xmath33@xmath1645@xmath19 ( fig .
7b - d ) . for other sources of dust ,
the width is approximately independent of @xmath33 at 30@xmath19@xmath27@xmath33@xmath27@xmath71 .
the widths all become relatively large at @xmath72@xmath27@xmath33@xmath2715@xmath19 , where there are no observations . at @[email protected] for particles started from comet 2p , the widths are generally narrower than that at smaller @xmath1 . for asteroidal particles , on the contrary , the width is maximum at @xmath1=0.4 . for 10p and 39p particles , the mean
width is slightly greater for smaller @xmath1 at 0.0001@xmath2@[email protected] ( fig .
8) . for particles started from long - period comets ( at @xmath59=0.995 , @xmath73=@xmath74=0.9 au , and orbital inclinations distributed between 0 and 180@xmath19 )
the mean width usually is even less than that for 2p particles at the same @xmath1 .
as it is summarized in figure 8 , the fwhm for asteroidal dust is significantly less than the 77 km / s fwhm of the observations . to fit the observations , we need to consider a significant fraction of cometary particles .
2p particles and particles started from long - period comets provided large values of the width in most runs , but the width varied considerably in such runs depending on @xmath1 and the place of start ( i.e. , the value of true anomaly ) from the comet .
in order to understand the variations in the model line profiles with the source and size of particles , we examined the values of mean eccentricities and mean orbital inclinations of particles at different distances @xmath45 from the sun ( figs .
these data show that , in general , the velocity amplitudes of the mg i line are greater for greater mean eccentricities and inclinations , but they depend also on distributions of particles over their orbital elements . the main contribution to the brightness of a dust cloud observed at the earth is from particles located at @xmath45@xmath273 au . therefore , because only positions of particles at @xmath45@xmath131 au are used for calculation of the brightness of particles at 90@xmath19@xmath2@xmath33@xmath2@xmath75 , if it is not mentioned specially , the mean eccentricities @xmath59 and orbital inclinations @xmath60 refer to 1@xmath2@xmath45@xmath23 au . at @xmath33=270@xmath19 prograde particles
move in the same direction as the earth , but usually slower because they are located father from the sun than the earth .
therefore , velocities of mg i line are usually positive at @xmath33=90@xmath19 and negative at @xmath33=270@xmath19 .
the values of @xmath59 , @xmath60 , @xmath56 , @xmath58 , and @xmath57 for particles from different sources are presented in table 1 for several values of @xmath1 .
this table and figures 3 - 6 , 9 - 12 show that at @[email protected] ( e.g. , for particles started from asteroids and comet 10p ) the velocity amplitudes @xmath56 are usually smaller than the observed amplitude ( 12 km / s ) , while for some runs at @[email protected] ( e.g. , for runs 2p and _ lp _ ) , @xmath56 is greater than the observed amplitude .
thus the wham observations correspond to a mean orbital eccentricity @xmath59 of about 0.5 .
however , the velocity amplitudes of the line depend not only on @xmath59 , but also on the distribution of all orbital elements of dust particles . for example , for trans - neptunian particles at 0.05@xmath2@[email protected]
, we have relatively large amplitudes ( @xmath56@xmath1616 km / s ) , even though @[email protected] ; the values of @xmath60 in this case were a little greater than for other sources of particles at @[email protected] .
also , for trans - neptunian particles , the values of @xmath56 were smaller at @xmath1=0.01 than at 0.05@xmath2@[email protected] , though @xmath59 was greater at @xmath1=0.01 .
the velocity amplitudes also depend on inclinations , because high inclination orbits have smaller projections of their orbital velocities on the lines of sight from the earth and the sun to particles near the ecliptic plane . for particles originating from long - period comets ( _ lp _ particles ) , the values of @xmath56 are significantly greater than those for dust particles from other sources , and the values of @xmath60 ( @xmath76105@xmath19 ) were much greater than for other runs . for _ lp _ runs at @xmath1=0.002 and @xmath73=0.9 au , @xmath59 was smaller ( @xmath270.25 ) than in most other runs , but @xmath56 was large ( 41 km / s ) , because @xmath60 was large ( 154 - 160@xmath19 ) .
note that mean initial inclinations for _ lp _ runs were about 90@xmath19 ( initial orbital inclinations were distributed uniformly between 0 and 180@xmath19 ) , but the mean inclinations of migrating particles were greater than 90@xmath19 ( see fig .
the velocities of mg i line depend not only on mean eccentricities and inclinations of particles , but also on the distributions of the orbital elements .
the distribution of orbital parameters and the resulting scattered line profile is dependent upon @xmath1 , because @xmath1 influences the lifetime of the particle . dynamical lifetimes of particles are greater for smaller @xmath1 . in the case of particles started from comet 2p , for runs presented in figures 13a , c , d , the maximum lifetimes were less than 5 kyr , and for runs presented in fig .
13b , f they were about 50 - 60 kyr .
particles that migrate more slowly into the sun ( smaller @xmath1 ) interact with planets for a longer time , and therefore they exhibit a wider range of parameter values , even though they have the same origin .
this explains the difference between the plots in figure 13 obtained at different @xmath1 ( wider range of @xmath77 was obtained at smaller @xmath1 ) .
as the motion is stochastic and the number of particles in our runs is not large , there may be no strict dependence on @xmath1 .
the plots of the velocities of mgi line for ` 2p 0.25 t ' and ` 2p 0.5 t ' runs presented in fig .
4c - d are higher than the observational line at large @xmath1 . in this case , all eccentricities are large at @xmath78@xmath761 au ( fig .
13a , c , d ) , lifetimes of all particles are close to each other and are very short ( @xmath275 kyr ) , and all particles moved practically in the same way . for @xmath1=0.05 ,
the maximum lifetime of 2p - particles started at aphelion ( ` 2p 0.5 t ' run ) was greater by a factor of 4 than that for ` 2p 0.25 t ' run .
therefore for @xmath1=0.05 , variations in @xmath77 at the same @xmath78 were greater for ` 2p 0.5 t ' run than for ` 2p 0.25 t ' run .
for many particles other than 2p particles , lifetimes are several tens or several hundreds of thousands of years and can reach tens of millions of years ( ipatov and mather 2006a ) . besides
, more particles start from comet 2p in its perihelion ( in this case , the plots do not differ much from the observational plot even at large @xmath1 ) than in aphelion .
therefore the zodiacal light contribution is very small for particles with @[email protected] started from comet 2p at aphelion or in the middle of the orbit .
finally , we examine the number density @xmath79 of particles with distance @xmath45 from the sun . in figures 14a - c
we present the values of @xmath80 in @xmath81 for @xmath45 equal to 0.3 and 1 au ( a ) , at @xmath45=0.8 and @xmath45=1.2 au ( b ) , and at @xmath45 equal to 1 and 3 au ( c ) .
these results can then be compared to values of @xmath80 deduced from observations of the actual zodiacal cloud .
the micrometeoroid flux ( @xmath82 g - @xmath83 g , i.e. , at @xmath9@xmath275 @xmath7 m for @xmath4=2.5 g/@xmath8 , or at @[email protected] ) measured on board helios 1 during 1975 is compatible with a number density @xmath84 at distance @xmath45 from the sun between 0.3 and 1 au ( grn et al .
1977 , leinert et al .
pioneer 10 observations between the earth s orbit and the asteroid belt yielded @xmath85 for particles of mass @xmath10@xmath83 g ( reach 1992 ) .
note that in our models , at 0.3@xmath2@xmath45@xmath21 au and 0.001@xmath2@[email protected] all values of @xmath80 exceed 1.9 for 2p - particles and are smaller than 1.1 for asteroidal particles ( fig .
at @[email protected] , the values of @xmath80 for particles started from other considered comets were less than 1.5 , but were mainly greater than those for asteroidal particles .
we can get @xmath80=1.3 for two component dust cloud if we consider 86% of particles with @xmath80=1.1 and 14% of particles with @xmath80=2 .
so the fraction of 2p particles needed to fit the helios observations can be less than 15% . at @[email protected] and 0.8@xmath2@[email protected] au ,
the mean value of @xmath80 for all points in figure 14b was a little smaller than 1.5 .
for cometary dust , @xmath80 was mainly greater than for asteroidal dust ; this difference was greater at @[email protected] than for @[email protected] . at @[email protected] ,
the values of @xmath80 for 2p - particles were greater than for other sources of dust .
at 1@xmath2@xmath45@xmath23 au for most of the dust sources , the values of @xmath80 were mainly greater than 1.5 ( fig .
for 0.1@xmath2@[email protected] , the values of @xmath80 for particles started from trans - neptunian objects and comet 39p better fit the observational value of 1.5 than those for particles from other sources ( including asteroidal dust ) .
so our studies of the distributions of particles over @xmath45 and of the velocities of mg i line testify that a fraction of trans - neptunian particles among zodiacal dust particles can be considerable .
our study of velocities and widths of the scattered mg i line in the zodiacal light is based on the distributions of positions and velocities of migrating dust particles originating from various solar system sources .
these distributions were obtained from our integrations of the orbital evolution of particles started from asteroids , comets , and trans - neptunian objects . a comparison of our models with the observations of velocities and widths of the zodiacal mg i line made by reynolds et al .
( 2004 ) shows that asteroidal dust particles alone can not explain these observations , and that particles produced by comets , including high - eccentricity comets ( such as comet 2p / encke and long - period comets ) , are needed .
a considerable fraction of trans - neptunian dust particles among zodiacal dust particles also fits different observations .
the mean eccentricity of zodiacal dust particles that best fit the wham observations is estimated to be about 0.5 .
this work was supported by nasa ( nag5 - 12265 ) and by the national science foundation through ast-0204973 .
ipatov , s. i. , mather , j. c. , & taylor , p. a. 2004 , in annals of the new york acad . of sci . ,
v. 1017 , astrodynamics , space missions , and chaos , ed .
b. belbruno , d. folta , & p. gurfil ( new york : nyas ) , 66 ( in angstrom ) at @xmath1=0.2 , @xmath33=180@xmath19 , in the ecliptic plane ( a , c ) and at @xmath1=0.05 ( exclusive for lp particles considered at @xmath1=0.002 ) toward the north ecliptic pole ( b , d ) .
zero of @xmath46=@xmath0-@xmath86 corresponds to @xmath0=@xmath86=5183.62 angstrom .
the plots for dust particles started from asteroids , trans - neptunian objects , comet 2p at perihelion , comets 10p and 39p are denoted by ` ast ' , ` tn ' , ` 2p ' , ` 10p ' , and ` 39p ' , respectively .
data for particles started from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 , are marked as ` lp ' .
the curves marked by ` ast 1 ' , ` tn 1 ' , and ` 2p 1 ' are for scattering function 1 , ` ast 2 ' , ` tn 2 ' , and ` 2p 2 ' are for function 2 , and ` ast3 ' , ` tn3 ' , and ` 2p 3 ' are for function 3 .
the curves are practically the same for all three functions .
marks in fig .
( a , c ) are for average intensity for observations at 174@xmath19@xmath2@xmath33@xmath2188@xmath19 ) , and marks in fig .
( b , d ) are for average observations toward the north ecliptic pole .
coordinates of marks for ` observ / sol spectr ' were obtained from observational data by making it have the same minimum value as the solar spectrum . in fig .
( c , d ) the minima of plots obtained in all runs were made the same as that for the solar spectrum .
, title="fig:",width=306 ] ( in angstrom ) at @xmath1=0.2 , @xmath33=180@xmath19 , in the ecliptic plane ( a , c ) and at @xmath1=0.05 ( exclusive for lp particles considered at @xmath1=0.002 ) toward the north ecliptic pole ( b , d ) .
zero of @xmath46=@xmath0-@xmath86 corresponds to @xmath0=@xmath86=5183.62 angstrom .
the plots for dust particles started from asteroids , trans - neptunian objects , comet 2p at perihelion , comets 10p and 39p are denoted by ` ast ' , ` tn ' , ` 2p ' , ` 10p ' , and ` 39p ' , respectively .
data for particles started from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 , are marked as ` lp ' .
the curves marked by ` ast 1 ' , ` tn 1 ' , and ` 2p 1 ' are for scattering function 1 , ` ast 2 ' , ` tn 2 ' , and ` 2p 2 ' are for function 2 , and ` ast3 ' , ` tn3 ' , and ` 2p 3 ' are for function 3 .
the curves are practically the same for all three functions .
marks in fig .
( a , c ) are for average intensity for observations at 174@xmath19@xmath2@xmath33@xmath2188@xmath19 ) , and marks in fig .
( b , d ) are for average observations toward the north ecliptic pole .
coordinates of marks for ` observ / sol spectr ' were obtained from observational data by making it have the same minimum value as the solar spectrum . in fig .
( c , d ) the minima of plots obtained in all runs were made the same as that for the solar spectrum .
, title="fig:",width=306 ] ( in angstrom ) at @xmath1=0.2 , @xmath33=180@xmath19 , in the ecliptic plane ( a , c ) and at @xmath1=0.05 ( exclusive for lp particles considered at @xmath1=0.002 ) toward the north ecliptic pole ( b , d ) .
zero of @xmath46=@xmath0-@xmath86 corresponds to @xmath0=@xmath86=5183.62 angstrom .
the plots for dust particles started from asteroids , trans - neptunian objects , comet 2p at perihelion , comets 10p and 39p are denoted by ` ast ' , ` tn ' , ` 2p ' , ` 10p ' , and ` 39p ' , respectively .
data for particles started from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 , are marked as ` lp ' .
the curves marked by ` ast 1 ' , ` tn 1 ' , and ` 2p 1 ' are for scattering function 1 , ` ast 2 ' , ` tn 2 ' , and ` 2p 2 ' are for function 2 , and ` ast3 ' , ` tn3 ' , and ` 2p 3 ' are for function 3 .
the curves are practically the same for all three functions .
marks in fig .
( a , c ) are for average intensity for observations at 174@xmath19@xmath2@xmath33@xmath2188@xmath19 ) , and marks in fig .
( b , d ) are for average observations toward the north ecliptic pole .
coordinates of marks for ` observ / sol spectr ' were obtained from observational data by making it have the same minimum value as the solar spectrum . in fig .
( c , d ) the minima of plots obtained in all runs were made the same as that for the solar spectrum .
, title="fig:",width=306 ] ( in angstrom ) at @xmath1=0.2 , @xmath33=180@xmath19 , in the ecliptic plane ( a , c ) and at @xmath1=0.05 ( exclusive for lp particles considered at @xmath1=0.002 ) toward the north ecliptic pole ( b , d ) .
zero of @xmath46=@xmath0-@xmath86 corresponds to @xmath0=@xmath86=5183.62 angstrom .
the plots for dust particles started from asteroids , trans - neptunian objects , comet 2p at perihelion , comets 10p and 39p are denoted by ` ast ' , ` tn ' , ` 2p ' , ` 10p ' , and ` 39p ' , respectively .
data for particles started from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 , are marked as ` lp ' .
the curves marked by ` ast 1 ' , ` tn 1 ' , and ` 2p 1 ' are for scattering function 1 , ` ast 2 ' , ` tn 2 ' , and ` 2p 2 ' are for function 2 , and ` ast3 ' , ` tn3 ' , and ` 2p 3 ' are for function 3 .
the curves are practically the same for all three functions .
marks in fig .
( a , c ) are for average intensity for observations at 174@xmath19@xmath2@xmath33@xmath2188@xmath19 ) , and marks in fig .
( b , d ) are for average observations toward the north ecliptic pole .
coordinates of marks for ` observ / sol spectr ' were obtained from observational data by making it have the same minimum value as the solar spectrum . in fig .
( c , d ) the minima of plots obtained in all runs were made the same as that for the solar spectrum .
, title="fig:",width=306 ] ( measured eastward from the sun ) at @xmath1=0.05 for dust particles started from asteroids ( a ) , comet 2p at perihelion ( b ) , comet 39p ( c ) , and trans - neptunian particles ( d ) .
letter ` c ' denotes the model for which the shift of the plot of @xmath41 vs. @xmath33 is calculated as a shift of centroid , and letter ` m ' denotes the model for which the shift of the plot is calculated as a shift of the minimum of the plot .
the number after ` m ' or ` c ' characterizes the number of a scattering function used.,title="fig:",width=306 ] ( measured eastward from the sun ) at @xmath1=0.05 for dust particles started from asteroids ( a ) , comet 2p at perihelion ( b ) , comet 39p ( c ) , and trans - neptunian particles ( d ) .
letter ` c ' denotes the model for which the shift of the plot of @xmath41 vs. @xmath33 is calculated as a shift of centroid , and letter ` m ' denotes the model for which the shift of the plot is calculated as a shift of the minimum of the plot . the number after ` m ' or ` c ' characterizes the number of a scattering function used.,title="fig:",width=306 ] ( measured eastward from the sun ) at @xmath1=0.05 for dust particles started from asteroids ( a ) , comet 2p at perihelion ( b ) , comet 39p ( c ) , and trans - neptunian particles ( d ) . letter ` c ' denotes the model for which the shift of the plot of @xmath41 vs. @xmath33 is calculated as a shift of centroid , and letter ` m ' denotes the model for which the shift of the plot is calculated as a shift of the minimum of the plot .
the number after ` m ' or ` c ' characterizes the number of a scattering function used.,title="fig:",width=306 ] ( measured eastward from the sun ) at @xmath1=0.05 for dust particles started from asteroids ( a ) , comet 2p at perihelion ( b ) , comet 39p ( c ) , and trans - neptunian particles ( d ) .
letter ` c ' denotes the model for which the shift of the plot of @xmath41 vs. @xmath33 is calculated as a shift of centroid , and letter ` m ' denotes the model for which the shift of the plot is calculated as a shift of the minimum of the plot . the number after ` m ' or ` c ' characterizes the number of a scattering function used.,title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 ( see the last number in the legend ) for particles started from asteroids ( a - b ) and comet 10p ( c - d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . , title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 ( see the last number in the legend ) for particles started from asteroids ( a - b ) and comet 10p ( c - d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . , title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 ( see the last number in the legend ) for particles started from asteroids ( a - b ) and comet 10p ( c - d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . ,
title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 ( see the last number in the legend ) for particles started from asteroids ( a - b ) and comet 10p ( c - d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . , title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 for particles started from comet 2p at perihelion ( a - b ) , at the middle of the orbit ( c ) , and at aphelion ( d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . ,
title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 for particles started from comet 2p at perihelion ( a - b ) , at the middle of the orbit ( c ) , and at aphelion ( d ) .
the line corresponds to the observations made by reynolds et al .
( 2004 ) . , title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 for particles started from comet 2p at perihelion ( a - b ) , at the middle of the orbit ( c ) , and at aphelion ( d ) .
the line corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] ( measured eastward from the sun ) at several values of @xmath1 for particles started from comet 2p at perihelion ( a - b ) , at the middle of the orbit ( c ) , and at aphelion ( d )
. the line corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] at several values of @xmath1 for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) , from comet 10p ( 10p ) , and comet 39p ( 39p ) .
the first designation ( + ) corresponds to the observations made by reynolds et al .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) . , title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) . , title="fig:",width=306 ]
( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 . ,
title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) . , title="fig:",width=306 ] ( see the last number in the legend ) for particles started from asteroids ( ast ) , from comet 2p at perihelion ( 2p ) , at the middle of the orbit ( 2p 0.25 t ) , and at aphelion ( 2p 0.5 t ) . , title="fig:",width=306 ]
( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 . , title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 . ,
title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
, title="fig:",width=306 ] ( see the last number in the legend ) for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
, title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . , title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . , title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . , title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . , title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . ,
title="fig:",width=306 ] ( 0.05 for ( a),(e ) , 0.01 for ( b),(f ) , 0.2 for ( c ) , and 0.1 for ( d ) ) for particles started from comet 2p at the middle of the orbit ( 2p 0.25 t ) and at aphelion ( 2p 0.5 t ) .
@xmath87 is the number of orbits with @xmath77 and @xmath78 in the interval of width of 0.02 and 0.05 au , respectively .
orbital elements were stored with a step of 20 yr , and the number of particles in each run was 100 or 250 . , title="fig:",width=306 ] in @xmath88 obtained by comparison of the values of the number density @xmath79 at distance @xmath45 from the sun equal to 0.3 and 1 au ( a ) , at @xmath45=0.8 and @xmath45=1.2 au ( b ) , and at @xmath45 equal to 1 and 3 au ( c ) .
the values are presented at several values of @xmath1 for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
horizontal bars correspond to observations .
, title="fig:",width=204 ] in @xmath88 obtained by comparison of the values of the number density @xmath79 at distance @xmath45 from the sun equal to 0.3 and 1 au ( a ) , at @xmath45=0.8 and @xmath45=1.2 au ( b ) , and at @xmath45 equal to 1 and 3 au ( c ) .
the values are presented at several values of @xmath1 for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
horizontal bars correspond to observations .
, title="fig:",width=204 ] in @xmath88 obtained by comparison of the values of the number density @xmath79 at distance @xmath45 from the sun equal to 0.3 and 1 au ( a ) , at @xmath45=0.8 and @xmath45=1.2 au ( b ) , and at @xmath45 equal to 1 and 3 au ( c ) .
the values are presented at several values of @xmath1 for particles started from comets 10p and 39p ( 10p and 39p ) , from trans - neptunian objects ( tno ) , and from long - period comets ( lp ) at @xmath17=0.995 , @xmath20=0.9 au , and @xmath18 distributed between 0 and 180@xmath19 .
horizontal bars correspond to observations .
, title="fig:",width=204 ] lllllll particles & @xmath1 & @xmath59 & @xmath60 & @xmath56 & @xmath58 & @xmath57 + & & & ( deg . ) & ( km s@xmath89 ) & ( km s@xmath89 ) & ( km s@xmath89 ) + asteroidal & 0.0004 - 0.1 & @xmath270.3 & 8 - 13 & 9 & ( -13)-(-11 ) & 5 - 7 + 10p & 0.0001 - 0.2 & 0.2 - 0.5 & 7 - 18 & 8 & ( -11)-(-9 ) & 3 - 9 + 39p & 0.01 - 0.2 & 0.2 - 0.4 & 12 - 15 & 8 - 9 & ( -13)-(-9 ) & 4 - 8 + 39p & 0.0001 - 0.004 & 0.28 - 0.82 & 5 - 25 & 11 - 17 & ( -20)-(-10 ) & 8 - 16 + 2p & 0.01 - 0.1 & 0.45 - 0.85 & 0.45 - 0.8 & 13 & ( -12)-(-7)&12 - 22 + 2p & 0.0001 - 0.004 & 0.6 - 0.9 & 0.6 - 0.9 & 14 & ( -16)-(-12)&15 - 20 + 2p 0.25 t & 0.002 - 0.2 & 0.6 - 0.85 & 4 - 11 & 16 & ( -14)-4 & 12 - 40 + 2p 0.5 t & 0.002 - 0.4 & 0.55 - 0.85 & 4 - 11 & 14 & ( -22)-(-2 ) & 2 - 29 + tno & 0.05 - 0.4 & 0.15 - 0.45 & 13 - 22 & 16 & ( -20)-(-14)&12 - 18 + tno & 0.01 & 0.48 - 0.76 & 14 - 17 & 12 & -16 & 8 + lp , @xmath73=0.9 au & 0.002 & 0.1 - 0.25 & 154 - 160 & 41 & -44 & 38 + lp , @xmath73=0.9 au & 0.0001 - 0.001 & 0.28 - 0.76 & 105 - 132 & 33 & ( -38)-(-32)&28 - 34 + lp , @xmath73=0.1 au & 0.0004 & 0.44 - 0.55 & 104 - 113 & 34 & -34 & 34 + | we present simulated observations of the doppler shifts of the solar mg i fraunhofer line scattered by asteroidal , cometary , and trans - neptunian dust particles .
the studies are based on the results of integrations of orbital evolution of particles under the gravitational influence of planets , the poynting - robertson drag , radiation pressure , and solar wind drag .
the derived shifts in the centroid and profile of the line with solar elongation are different for different sources of dust .
a comparison of the velocities of zodiacal dust particles based on these numerical integrations with the velocities obtained from wham observations shows that the fraction of cometary dust particles among zodiacal dust particles is significant and can be dominant .
a considerable fraction of trans - neptunian dust particles among zodiacal dust particles also fits different observations .
the mean eccentricity of zodiacal dust particles is estimated to be about 0.5 . |
an elongated current sheet attached above the flare loop top @xcite is usually observed in an eruptive solar flare .
magnetic reconnection inside current sheets(css ) plays an important role to release and convert the magnetic energy to the plasma thermal and kinetic energy .
reconnection inflows and high speed outflows in css have been recognized by lots of observations @xcite .
the speed of outflows ranges from @xmath12 kms@xmath11 .
many studies show that the outflow speed is around @xmath13 kms@xmath11 .
the maximum outflow velocity can reach @xmath14 kms@xmath11 @xcite , which is close to the alfvn velocity in solar corona .
the temperature of the plasmas relating with a magnetic reconnection process can be heated to around 3 mk @xmath15 20 mk .
the 3 mk plasma was observed above the post - flare loops by @xcite in euv with hinode / extreme - ultraviolet imaging spectrometer spectra ; using the solar and heliospheric observatory(soho ) ultraviolet coronagraph spectrometer ( uvcs ) data , @xcite observed plasmas with the temperature around 6 mk inside the current sheet region ; the bright blob with hot plasma in the atmospheric imaging assembly ( aia ) @xmath16 passband has been seen with the peak temperature @xmath1511 mk @xcite ; reuven ramaty high energy solar spectroscopic imager(rhessi ) x - ray spectra and images simultaneously show that the plasma has been heated to @xmath17 10 mk in a solar flare in the paper by @xcite ; @xcite even observed the hot loop plasmas with the peak temperature reaching @xmath1522 mk by rhessi hard x - ray spectra . in the paper by @xcite , the rhessi hard x - ray emissions in figure 6(b ) indicate that some local heating is located at the reconnection site .
the local heating regions possibly correspond to some fine structures like magnetic islands within the current sheet .
they also predict that the sharp change of temperature across the current sheet in figure 7(f ) in their paper is probably a signature of a slow shock .
the generation mechanism of the high temperature emission observed by solar telescopes is still not understood well .
@xcite proposed that both the petschek - like reconnection and turbulent reconnection are possible to explain the hot cs plasma .
the petschek - like reconnection model can be used to explain some observation evidences of white - light , uv and x - ray emissions@xcite . however
, a nonuniform resistivity which should be enhanced at the x - point is required to produce steady - state petschek reconnection(see kulsrud 2001 and references ) .
the physical mechanism about the origin of the enhanced resistivity is still unknown .
@xcite studied the energy conversion mechanism by three - dimensional mhd magnetic reconnection model , they concluded that petschek reconnection accelerates plasma to convert magnetic energy to bulk kinetic energy , then the accelerated plasma becomes slowed down and the bulk kinetic energy is transformed to heat .
the hottest plasma can reach around 40 mk in their simulation .
however , they assumed a non - uniform resistivity , and the corresponding lundquist number at the x - point is only 200 , which is much smaller than that in solar corona ( @xmath18 ) . both radiative cooling and heat conduction effect
are not included in their model .
fast magnetic reconnection develops in a turbulent plasma is expected to produce hot plasma as observed by ucvs and x - ray telescope images@xcite .
several types of reconnection models in turbulent plasma are proposed , e.g. stochastic reconnection@xcite , fractal reconnection or plasmoid instabilities@xcite .
these models have been studied analytically and numerically in the past few years , and some common charicteristics were found . in a high lundquist number environment , plenty of observations , numerical simulations@xcite and even laboratory experiments@xcite show that multiple levels of plasmoids can always occur in an unstable magnetic reconnection process .
numerical simulations have demonstrated that the reconnection rate @xmath19 can be strongly increased to a high value(@xmath20)@xcite after multiple levels of plasmoid instabilities appear , and @xmath19 weakly depends on lundquist number .
this value is very close to the reconnection rate measured by observations@xcite .
however , few numerical simulations focus on the generation of hot plasma inside the current sheet during plasmoid instabilities . in the paper by @xcite , the plasma densityis ( @xmath21 )
is set much higher than that in the corona environment , and the temperature can only be increased from 1 mk to around 2 mk .
the spectrum studies are important for understanding the energy conversion mechanism and dynamics of plasmoids during the plasmoid instabilities . in the papers by @xcite , one dimensional magnetic and kinetic energy spectra along the current sheet
have been studied detailedly .
the results show that the spectrum is not a single power law form for both magnetic and kinetic energy , the spectral index increases with the wave number after secondary islands appear .
the average spectral index for magnetic energy is around 2.0 and it is around 3.0 for kinetic energy in the previous studies @xcite .
these values are larger than the kolmogorov spectral index ( @xmath22 ) .
the plasmoid distribution function has also been studied detailedly in some previous papers , e.g. @xcite .
their results also demonstrate that the spectrum of the distribution function does not behave as a single power law . in this work
, we study the energy conversion during plasmoid instabilities in the solar corona current sheet .
the physical parameters such as the initial temperature , density and strength of magnetic field in the simulations are close to those in real corona environment .
the evolutions of the reconnection rate , temperature and velocity of the plasma inside the current sheet region have been analyzed .
though , the energy conversion process in a flare current sheet has already been studied detailedly in some previous papers ( e.g. , reeves et al .
compared with these previous studies , a more realistic temperature dependent high lundquist number ( @xmath23 ) and much higher resolutions are used in our models .
therefore , we can discover many fine structures related to the heating mechanism inside the current sheet region .
both the one and two dimensional spectra have also been studied carefully in this work . in section
ii , we present the numerical approach and initial states of our simulations . the results are presented in section iii .
we summarize our results and give discussions in section iv .
the 2.5-dimensional one fluid mhd model is used in this work .
we only consider the fully ionized hydrogen gas and the plasma are composed with electrons and ions .
the temperature of the two species are considered the same ( @xmath24 ) .
the mhd equations in our simulations are given by : @xmath25 @xmath26 \nonumber\\ & & + \nabla \cdot [ \textbf{v}\vec{\vec{\tau}}+\frac{\eta}{\mu_0 } \textbf{b } \times ( \nabla \times \textbf{b } ) ] + l_{rad}+h,\end{aligned}\ ] ] @xmath27+\nabla \cdot \vec{\vec{\tau}},\ ] ] @xmath28 @xmath29 @xmath30 @xmath31 where @xmath32 is the plasma mass density , @xmath33 is the centre of mass velocity , @xmath34 is the total energy density , @xmath35 is the magnetic field , @xmath36 is the stress tensor , @xmath37 is the mass of a hydrogen ion , @xmath38 is the magnetic diffusivity , @xmath39 is the dynamic viscosity coefficient and @xmath40 is plasma thermal pressure .
the ratio of specific heats @xmath41 is set to @xmath42 ( ideal gas ) .
the magnetic permeability coefficient @xmath43 is set to @xmath44 .
the radiative cooling @xmath45 and heating @xmath46 for fully ionized high temperature plasma are analytically assumed as@xcite : @xmath47 @xmath48 where @xmath49 k is the initial temperature in the whole simulation domain at @xmath50 .
the simulation domain extends from @xmath51 to @xmath52 in @xmath53 direction and from @xmath54 to @xmath55 in @xmath56 direction , with @xmath57 m. open boundary conditions are used in both @xmath53 and @xmath56 directions .
we use a force - free current sheet with a strong guide field in the center as the initial equilibrium distributions of magnetic fields : @xmath58 where @xmath59 t. the initial current sheet width thus is @xmath60 . due to the force - freeness and neglect of gravity ,
the initial equilibrium thermal pressure and plasma @xmath2 is uniform . in this work ,
we have simulated two cases , @xmath61 in case a and @xmath62 in case b. these yield the initial plasma thermal pressure @xmath63 pa in case a and @xmath64 pa in case b. since the initial equilibrium temperature is @xmath49 k in both of the two cases , we can find that the initial plasma density @xmath65 kgm@xmath66 and alfvn velocity @xmath67 kms@xmath11 in case a , @xmath68 kgm@xmath66 and @xmath69 kms@xmath11 in case b. therefore , case a could represent a lower height in solar corona than that for case b. the same magnitude ( @xmath70 ) of initial perturbations of both magnetic field and velocity are applied at @xmath50 in the two cases to trigger the reconnection process .
the forms of perturbations are listed as below : @xmath71 \\
b_{ypert } & = & pert\cdot b_0 \cdot sin(\frac{2\pi x}{l_0 } ) \cdot cos[\frac{2\pi ( y-0.5l_0)}{l_0}]\\ v_{xpert } & = & pert\cdot v_{a0 } \cdot sin(\frac{2\pi x}{l_0 } ) \cdot sin(\frac{\pi y}{2l_0 } ) \\
v_{ypert } & = & -pert\cdot v_{a0 } \cdot sin(\frac{8\pi y}{l_0})\cdot \frac{random_n } { max(\lvert random_n\rvert ) } , \end{aligned}\ ] ] where @xmath72 , @xmath73 is the initial alfvn velocity as presented in the above paragraph , @xmath74 is the random noise function in our code , and @xmath75 is the maximum of the absolute value of the random noise function .
the perturbations result in a thinning of the current sheet in two sections between a set of three primary islands , whose midpoints are located at @xmath54 , @xmath76 , and @xmath77 ( see figure 1 ) . in this paper , we will focus only on the section in the domain @xmath78 , i.e. , the bottom half of the box is used as an auxiliary part of the computation only .
its function is to generate a stationary primary plasmoid at the bottom of the height range of interest , which is not influenced by any effects of a numerical boundary ; the primary plasmoid acts like a line - tied bottom of the current sheet of interest .
the magnetic diffusivity in the two cases are both assumed as @xmath79 .
since the initial temperature is @xmath80 k , the initial magnetic diffusion is calculated as @xmath81 m@xmath10s@xmath11 . the lundquist number based on this magnetic diffusivity , @xmath73 , and @xmath76 , which corresponds to the global scale of the current sheet in the upper part of the box , is @xmath82 in case a and @xmath83 in case b. as the plasmoid instabilities develop
, the highest temperature at the main @xmath84-point can reach around @xmath85 mk in our simulations in case b , and the corresponding lunduist number is around @xmath86 .
such a lundquist number is already very high comparing with all the previous solar corona magnetic reconnection simulations , even though it is still lower than the lundquist number in real solar corona ( @xmath18 ) .
the dynamic viscosity coefficient is assumed as a constant , @xmath87kgm@xmath11s@xmath11 .
the initial @xmath88 is around @xmath89 kgm@xmath11s@xmath11 .
the computations are performed by using the mhd code nirvana ( version 3.6 ; * ? ? ?
* ) . adaptive mesh refinement ( amr ) is applied .
the derivatives - based mesh refinement criterion is used for the mesh refinement .
the gradient - based / second - derivatives - based criterion is given by : @xmath90
\left(\frac{\delta x^{(\ell)}}{\delta x^{(0 ) } } \right)^{\xi } \left\ { \begin{array } { ll } > \varepsilon_u \qquad \exists u & \textrm { refinement } \\
< 0.8\varepsilon _ u \quad \forall u & \textrm{derefinement } \end{array } \right.\ ] ] where @xmath91 is a set of primary variables that the criterion is applied to . the mass density , momentum densities , energy density and magnetic field can all be chosen as the variable @xmath91 to set the criterion in the nirvana code . undivided first ( @xmath92 ) and second ( @xmath93 ) differences of @xmath91 are computed is some sort of 2-norm . @xmath94 $ ] ( switch between a purely gradient - based criterion when @xmath95 and second - derivatives - based criterion when @xmath96 ) , @xmath97 ( reference values ) , @xmath98 ( thresholds , the typical range is @xmath99 $ ] ) and @xmath100 ( level dependence ) are user - controllable parameters . the criterion is checked on a generic block octant - wise including a 2-cell wide buffer zone around the octant .
fil(preset to @xmath101 ) is a filter to suppress refinement at small - scale wiggles . for the results presented in this manuscript
, we choose the magnetic field to set the criterion .
the threshold parameter @xmath98 for the magnetic field is set to equal @xmath102 , the reference value @xmath97 is @xmath103 t , @xmath104 .
the time integrator for mhd equations we have used in this code is the third - order accurate runge - kutta method .
the second - order version of the central - upwind scheme is applied to the euler equations with lorentz - force term combined with a ct scheme for the induction equation .
the electric field is computed from a genuinely 2d central upwind procedure ( cct ) based on the evolution - projection method .
the divergence - free condition of the magnetic field is built - in property of the scheme by virtue of a constrained - transport ansatz for the induction function .
the relative divergence of the magnetic field which has been tested is normally smaller than @xmath105 .
the detailed descriptions of this kind of scheme are presented in the paper by @xcite . in that paper ,
numerical experiments illustrate the overall robustness and performance of the scheme for some tests .
we start our simulations from a base - level grid of @xmath106 .
the highest refinement level is 13 , which corresponds to a grid resolution @xmath107 m. it is around one magnitude higher than the ion inertial length in solar corona .
convergence studies have been carried out by repeating some simulations with a higher resolution for case b , with the highest refinement level limited to 14 .
the numerical results in the higher resolution case are very similar to the results presented in the next section . as we all know
, the numerical diffusion is inevitable in numerical experiments . to evaluate the numerical noise in our simulations
, we use the similar method as in the paper by @xcite to perform an estimate for case b as below . the magnetic induction equation ( 4 )
can be wrote as : @xmath108 here the flux function @xmath7 is defined through the relations @xmath109 , @xmath110 . in the absence of numerical diffusion , both sides of equation(16 ) should ideally balance each other .
however , the two sides can not exactly balance each other in realistic numerical simulations .
we have estimated the numerical diffusivity around the main x - point within a short time for case b by using the following method .
the main reconnection @xmath84-point is determined as the @xmath84-point which has the highest @xmath7 value of all @xmath84-points in the box .
suppose that the main x - point at @xmath111 s is at position @xmath112 .
we define @xmath113 , @xmath114_z$ ] , and @xmath115_z$ ] .
the values of @xmath116 and @xmath117 are calculated and presented in figure . 2 .
since the plasma velocity and magnetic field in @xmath53 and @xmath56 direction at the main x - point is near zero , the value of @xmath118 is much smaller than the value of @xmath119 in figure . 2 .
therefore , @xmath120 represents the ratio of @xmath121 , @xmath122 represents @xmath123 . in figure . 2 , one can find that the numerical diffusivity should be smaller than @xmath124 of the physical one within such a time interval .
we only choose a short time interval in figure 2 , the reason is that the position of the main x - point varies with time .
the main x - point is no longer at around position @xmath112 after @xmath125 s. therefore , the value of @xmath126 gradually becomes larger than @xmath127 when the main x - point leaves away from position @xmath128 . by using the same method
, we can not prove that the numerical diffusivity in the regions away from the reconnection x - point is also smaller than the the physical diffusivity . because the numerical diffusion is mostly caused by the @xmath129 term in this situation
, such kind of numerical diffusion could be larger than the physical @xmath130 term .
disturbed by initial perturbations , the current sheet section between two primary islands as shown in figure . 1 starts to develop towards a thinner and thinner sweet - parker like long current sheet . as the aspect ratio of the long current sheet exceeds a critical value , the current sheet
is broken to multiple secondary fragments and many small islands start to appear .
these secondary magnetic islands start to grow bigger and move along the current sheet after they appear .
many newer and higher order islands also begin to develop .
when we zoom in to small scales with higher resolutions as we did in figure . 2 in our previous paper @xcite , we find that some secondary current sheet fragments are broken to thinner filaments and the third order smaller plasmoids are formed . as we continue to zoom in to smaller scales , the thinnest current sheet width
is then found around @xmath14 m and the highest order of plasmoids in our simulations is the fourth - order .
since there are multiple reconnection x - points , some of the islands move upward and some of them move downward .
the fast moving island can catch up with the slow island and the two islands close to each other moving with opposite directions will collide eventually .
after the two islands collide , they can be coalesced to form one bigger island .
the above phenomena can be seen clearly in figures . 3 , 5 and 7 . from figure . 3 and 4
, we can also find that the main x - point moves up with time in our simulations , the red crosses in figures 4(a ) , ( b ) and ( c ) stand for the main x - point .
the main reconnection x - points at @xmath131 s , @xmath132 s , @xmath133 s , @xmath134 s and @xmath135 s are detected separately at @xmath136 , @xmath137 , @xmath138 , @xmath139 and @xmath140 .
we use the same method as that in our previous papers @xcite to detect the main x - point . figure . 3 and 5
demonstrate that the plasmoids above the main x - point eventually move out the simulation domain and those below the main x - point collide with the primary big island at the bottom . figure .
3 also shows that the maximum outflow velocity can reach around @xmath14 kms@xmath11 even before secondary islands appear .
this value is the same as the observed maximum outflow velocity .
after secondary islands appear , the outflow velocities above and below the main x - point acutely fluctuate . figure .
3 , 5 and 6 indicate that a termination shock is formed above the primary island .
figure 6(a ) shows that the plasma velocity along y direction starts to decrease to a value smaller than the sound speed @xmath141 at around @xmath142 at @xmath135 s. the entropy s in figure .
6(b ) and the magnetic field parallel to the shock front in figure .
6(c ) suddenly jump to a much a higher value at around @xmath142 .
this is exactly the behavior of a fast - mode shock .
the termination shocks have also been found in the out - flow regions of the multiple reconnection x - points in the plasmoid dominated regime .
these fast - mode shocks may be related to the hard x - ray non - thermal emission above the soft x - ray flare arcades ( e.g. masuda et al .
1994 , tsuneta & naito 1998 , krucker et al .
2010 ) . however , the particle acceleration process by fast - mode shocks in kinetic scale is still not clear and beyond the scope of our present mhd work . in the mhd scale ,
the hot plasmas generated by fast - mode shocks in solar corona are expected to be observed @xcite .
the slow - mode shocks are usually formed at the outflow regions of a petschek - like current sheet or behind the moving magnetic islands @xcite . in our simulations
, a lot of slow - mode shocks also appear at the edges of the plasmoids .
the fifth contour plot for @xmath135 in both figure .
3 and 5 clearly indicate that there is a pair of slow - mode shocks at the edges of the primary island in the down flow region below the main - x point .
the slow - mode shocks are also formed at the edges of secondary islands .
the bottom right panel of figure .
7(a ) and bottom left panel of figure .
7(b ) show a up - moving magnetic island which is formed by two coalescent islands , a pair of nearly symmetric slow - mode shocks are formed in front of such a island , another pair of disturbed slow - mode shocks are behind it . except for such special cases ,
the slow - mode shocks are usually formed behind the moving magnetic islands in our simulations .
the shock angle of the two nearly symmetric shocks is around @xmath143 .
7(c ) presents the magnetic field and the current density along a cut in the @xmath53-direction at @xmath144 . around @xmath145 , the field component tangential to the shock , @xmath146 , decreases rapidly toward the downstream side and that the current density @xmath147 has a peak , but the component normal to the shock , @xmath148 , stays nearly uniform along the cut .
these shock structures can be distorted as the reconnection outflow plasma becomes turbulent and the plasmoids collide with each other .
the distorted slow - mode shock fronts which are formed behind a up - moving magnetic island are presented in the right up panels of figure .
7(a ) and ( b ) .
the black arrows in the right up panel of figure .
7(b ) represent plasma velocity , one can see that the parallel shear flows with different velocities appear around the contact surface between the reconnection outflows and the ambient plasmas .
however , the kelvin - helmholtz instabilities with multiple vorticities do not appear because of the strong aligned magnetic fields@xcite .
these dynamic structures are only the secondary fragments of the current sheet which are fluctuated along the shock fronts . figure .
5 shows the temperature distributions at five different times for case b with @xmath62 . during the reconnection process
, one can see that the temperature gradually increases with time by ohmic heating inside the current sheet region , which is similar as those previous work ( e.g. , ugai 1992 ) . however , as shown in figure . 4 and 5 , the temperature distribution is not uniform at the current sheet region especially after secondary islands appear .
since the significant heating takes place at the slow shocks attached to the plasmoids , not at the x - points , the temperature inside the plasmoids is usually higher than that at the reconnection x - points as shown in figure
the non - uniform behaviors of the temperature at the current regions in our simulations are in accordance with the observation results about the current sheet above the cusp - shaped structure in the gradual phase by @xcite . in figure . 7 in their paper
, they also show the nonuniform temperature distribution along and vertical to the observed current sheet .
they predict that the sharp change in both the temperature and the emission measure distribution curves could be the evidence of a slow mode shock produced by magnetic reconnection .
we have used the same method as those in our previous papers to calculate the reconnection rate in both case a and case b. the reconnection rate is computed as the rate of change of the magnetic flux accumulated between the @xmath149-point in the primary island at @xmath150 and the main reconnection @xmath84-point ( see ni et al .
2012 ; ni et al . 2013 ) , @xmath151 where @xmath152 and @xmath73 are the initial magnetic field and alfvn velocity at the inflow boundary presented in section ii , one can note that the values of @xmath73 are different in case .
a and case .
b. the magnetic reconnection rates varying with the normalized timescale in figure .
8(a ) are very similar in the two cases .
secondary islands start to appear at around @xmath153@xmath154 in case a and @xmath155@xmath156 in case b , where @xmath157 s and @xmath158 s are the initial alfvn time in case a and case b respectively .
the reconnection rate can reach around @xmath159 in both of the two cases .
however , as shown in figure .
8(b ) , the maximum temperature in case b can reach more than 2 times higher than that in case a. therefore , the plasma can be heated to a higher temperature in a reconnection layer with a lower plasma density .
the hottest plasma in case b with @xmath62 is around @xmath160 mk , and such high temperature plasmas can be observed by hard x ray telescopes .
@xcite have measured the effective magnetic diffusivity @xmath161 by observations , where @xmath162 is the reconnection inflow velocity and @xmath163 is the half - thickness of the current sheet .
9 presents the inflow velocity @xmath164 in the @xmath53-direction through the main x point at @xmath131 s and @xmath135 s in our simulations , here @xmath164 represents the inflow velocity @xmath162 . in order to compare with observations
, we measure the half - thickness of the current sheet @xmath163 from the position where @xmath164 starts to decrease to the position where @xmath165 as indicated in figure . 9 . at @xmath131
s , the half - thickness is found to be @xmath166 m and the inflow velocity is around @xmath167 ms@xmath11 . at @xmath135 s , @xmath163 decreases to @xmath168 m and @xmath162 decrease to @xmath169 ms@xmath11 .
the effective magnetic diffusivity is then obtained as @xmath170 m@xmath10s@xmath11 at @xmath131 s , and @xmath171 m@xmath10s@xmath11 at @xmath135 s. these values are close to those deduced by @xcite on the basis of observations . though the resolution used in our simulations is much higher than that of the observational instruments .
we should note that the refinement level for calculating the half - thickness of the current sheet in figure .
9 is zero and the resolution is around @xmath172 km , which is close to the highest resolution of the solar space telescopes .
the effective magnetic diffusivities at @xmath135 s have also been measured through other x - points by using the same method as above .
the measured values of @xmath173 through these x - points are all on the order of @xmath174 m@xmath10s@xmath11 .
10 presents the time dependent energy conversion for @xmath61 and @xmath62 in a fixed region ( @xmath175 and @xmath176 ) . since energy fluxes through the boundaries at @xmath177 , @xmath178 , @xmath179 and @xmath180 can always exist . the magnetic , thermal , and kinetic energy flowing into this region through these boundaries from the beginning of the simulation ( @xmath50 ) to time
@xmath181 is denoted as @xmath182 , @xmath183 , and @xmath184 , respectively .
( note that these quantities may have negative signs if energy flows out of the region . )
the magnetic , thermal , and kinetic energy confined to this region at time @xmath181 is denoted as @xmath185 , @xmath186 , and @xmath187 , respectively .
the initial magnetic , thermal , and kinetic energy at @xmath50 is denoted as @xmath188 , @xmath189 , and @xmath190 , respectively .
the total radiated thermal energy from beginning to time @xmath181 is denoted as @xmath191 . in these notations , the dissipated magnetic energy in the region defined by @xmath175 and @xmath176 ,
is given by : @xmath192 .
in the same region , the generated thermal energy is @xmath193 , and the generated kinetic energy is @xmath194 .
the detailed calculations about the above variables are similar as those in our previous paper @xcite . however , there are some improvements in this work comparing with our previous paper .
we have used 2.5-dimensional mhd instead of the 2-dimensional mhd equations , open instead of periodic boundaries are used at the top and bottom , radiation cooling and heating terms are also included in this work .
therefore , the @xmath195 components of the magnetic field and velocity should be included to calculate the above variables , the energy flux @xmath196 , @xmath197 and @xmath198 at the boundaries and the radiated energy @xmath191 have to be included . figure .
10 shows that the corresponding generated thermal and kinetic energy behave similarly as the dissipated magnetic energy . from beginning to time @xmath181 , the total dissipated magnetic energy exactly equal the generated thermal energy plus kinetic energy , the errors are under @xmath199 .
the magnetic energy is mostly converted to kinetic energy before secondary islands appear , e.g. , the generated kinetic energy is around four times higher than the generated thermal energy from beginning to @xmath200 s@xmath201@xmath156 in case b with @xmath62 .
after secondary islands appear , the generated thermal energy grows much faster with time than the kinetic energy , and there are more generated thermal energy than kinetic energy eventually .
the small scale slow - mode shocks attached to the edges of the multiple magnetic islands play important roles in generating thermal energy .
5 , and 7 clearly show that the highest temperature structures always appear at the shock front regions . in the previous papers by @xcite
, they also pointed out that the energy dissipation is accomplished via many concurrent small - scale events appearing in multiple sites distributed in space . in order to inspect the effects of joule heating ,
the generated thermal energy and joule heating have been calculated in several fixed regions during a period . as long as the slow - mode shocks are included in these regions
, we find that the generated thermal energy is always much larger than the joule heating in a fixed period . therefore , joule heating is not the main reason to cause the sharply increasing thermal energy .
though the generated kinetic energy increases slower than thermal energy after secondary islands appear , there is still around @xmath202 of the dissipated magnetic energy which has been converted to kinetic energy from beginning to @xmath203 s@xmath204@xmath156 in case b. this is different from the result in our previous paper @xcite , after secondary islands appear , the generated kinetic energy decreases fast to a small value and around @xmath205 of the dissipated magnetic energy has been transformed to thermal energy eventually in that paper .
the periodic boundary conditions applied in @xmath56 direction in the paper by @xcite is the main reason to cause this difference .
the outflow plasmas are confined inside the simulation domain by periodic boundary conditions , they collide with the primary island at the top and slow down eventually . the outflow plasmas in this present work can gradually escape from the top boundary because of the open boundary conditions . from figure .
10(a ) , one can also see that the magnetic energy is dissipated faster and more thermal energy is generated in case b with @xmath62 than that in case a with @xmath61 during a period after secondary islands appear .
the reason is that the plasma in the lower @xmath2 case can be compressed more strongly and the shock heating becomes more important in the energy conversion process .
the termination shocks at the head of plasmoids also contribute a small part of the generated thermal energy .
part of the kinetic energy of the reconnection out flows can be converted to thermal energy at the termination shocks . in figure . 5
, one can see that the temperature increases at the termination shock above the primary island .
however , the temperature of the heated plasma at the termination shock is much lower than the highest temperature at the slow - mode shock fronts . at the fragment current sheet regions where the slow - mode shocks are not included , the heating by dynamic viscosity has been measured to compare with joule heating .
the joule heating can be measured as @xmath206 , and the viscous heating is measured as @xmath207 . at the small current sheet fragments , the current density @xmath208 is around @xmath209 am@xmath210 , the maximum @xmath211 is calculated around 10 s@xmath11 , the magnetic diffusion @xmath38 is around @xmath212m@xmath10s@xmath11 and @xmath213kgm@xmath11s@xmath11 .
then , we can find that @xmath214 is around @xmath215 .
therefore , the heating by dynamic viscosity can be ignored at the fragment current sheet regions in this work .
the one - dimensional energy spectra along the reconnection current sheet have been studied numerically in some papers @xcite .
after secondary islands appear , the numerical results in these previous papers demonstrate that the energy spectrum does not behave as a simple power law anymore .
the spectral index for both kinetic and magnetic energy varies with wave number , it usually increases to a higher value as the wave number increases . before studying the two - dimension spectra , we present the one - dimensional spectra along the current sheet for case b in figure . 11 . the similar method as that in our previous papers @xcite has been used to get the spectral index for both kinetic and magnetic energy .
firstly , each component of magnetic field , velocity and density along the current sheet at @xmath145 has to be transformed to fourier space .
then , the magnetic energy spectrum is calculated as @xmath216 and the kinetic energy spectrum is @xmath217 .
finally , we fit the spectrum to a power law ( @xmath218 and @xmath219 ) to obtain the energy spectrum index . figure .
11 shown that the spectra also do not behave as a simple power law after secondary islands appear .
we only fit a line to get the spectrum index @xmath220 within the region @xmath221 and @xmath222 within the region @xmath223 .
therefore , the spectral index we have obtained is an average value .
the spectral index is larger before secondary islands appear for both kinetic and magnetic energy spectrum .
after secondary islands appear , the spectra become harder .
the index for magnetic energy spectra decreases to around @xmath3 and the one for kinetic energy spectra is around @xmath224 .
these characteristics are very similar as those one - dimensional energy spectra presented in the previous papers@xcite .
however , we should note that the values of these spectral index vary in an range , they are not precisely fixed . in the paper by @xcite , the one dimensional spectral index for the magnetic energy is @xmath225 at @xmath226 .
but such a index varies with time as shown in the papers by @xcite , it is normally in the range @xmath227 for the magnetic energy spectrum after secondary islands appear .
we have chosen the level 5 data to plot figure .
11(b ) and ( c ) . at @xmath131
s , the current sheet is smooth and no secondary islands appear , the highest refinement level is 4 .
therefore , the level 4 data is used for plotting figure .
11(a ) . for the first time
, we have studied the two - dimensional energy spectra for both kinetic and magnetic energy .
firstly , the magnetic field , velocity and mass density in the region @xmath228 and @xmath229 are transformed to the two - dimensional fourier space .
then , we calculated the two - dimensional magnetic and kinetic energy as @xmath230 and @xmath231 .
there are @xmath232 grids in @xmath4 direction and @xmath233 grids in @xmath5 direction .
@xmath4 is defined as @xmath234 , @xmath235 , @xmath236 , @xmath237 , ... , @xmath238 , and @xmath5 is defined as @xmath234 , @xmath239 , @xmath240 , @xmath241 , ... ,
@xmath243 are the length scales we have selected in @xmath53 and @xmath56 direction , respectively .
the two - dimensional distributions of @xmath244 for @xmath131 s and @xmath132 s are presented in figure .
12 . since the high energy parts of the two dimensional spectra are mostly located at the positions with small @xmath4 and @xmath5 . in order to see the distributions of the two dimensional spectrum more clearly .
the imaginary parts of the spectra with negative coordinates are also presented in figure .
we have also zoomed into a smaller scale within @xmath245 and @xmath246 .
then , one can clearly see that the high energy parts are located in the center of the plots . in figure .
12 , we only need to focus in the first quadrant ( the top right quadrant ) with positive coordinates in the two panels .
the dark red color represents the highest magnetic energy and the dark blue color represents the lowest one . at @xmath131
s before secondary islands appear , the distribution of magnetic energy in fourier space is relatively smooth and the high magnetic energy is confined in the region near @xmath247 . at @xmath135
s after secondary islands appear , part of the high energy is obviously spread in to the space with larger @xmath4 and @xmath5 .
as we know , the larger the wave number is , the smaller the wave length is .
these evidences prove further that plasmoid instabilities can cause multiple levels of fine structures in the reconnection regions .
the magnetic energy can be cascaded to smaller and smaller scales , and it is dissipated in these small scales eventually . comparing with a one dimensional spectrum ,
the two dimensional ones show how asymmetric the spectra look like and give more detailed information . in the right panel of figure .
12 , one can clearly see that the magnetic energy is not uniformly cascaded in to the the fourier space after secondary islands appear , and most of the magnetic energy is still confined in the scales with small @xmath5 .
the level 4 data have been used to derive figure . 12 .
the distributions of the two - dimensional kinetic energy spectra which we do not show here is very similar as the magnetic energy spectrum shown in figure 12 .
the plasmoid distribution function is also important for revealing the statistical properties and understanding the dynamics of these plasmoids .
following the same method as @xcite and @xcite did , we have also calculated the plasmoid distribution function @xmath248 numerically .
@xmath249 is the number of plasmoids with magnetic flux larger than @xmath7 .
the magnetic flux of a magnetic island is calculated by @xmath250 , where @xmath251 is the magnetic flux at the @xmath149-point of the magnetic island and @xmath252 is the magnetic flux at the nearby @xmath84-point . for obtaining the plasmoid distribution function presented in figure . 13 ,
plasmoids appearing in different snapshots with an interval of @xmath253 are accumulated during the evolution of the current sheet .
13 shows that the plasmoid distribution function behaves as a power law closer to @xmath6 in the intermediate @xmath7 regime . in the large @xmath7 regime ,
the distribution function @xmath254 gradually deviates from the power law @xmath255 to a more rapid falloff .
studying the energy conversion and spectra of a corona current sheet are very important to understand the physical mechanisms of magnetic reconnection and to explain many observational features in flares . in this work
we have simulated a 2.5-dimensional corona current sheet with more realistic physical parameters , the temperature dependent high lundquist number ( @xmath212 ) has been used , both the radiation cooling and heating terms are included . here is a summary of the main results . \1 .
after the sweet - parker long current sheet is broken to multiple fragments , many petschek - like fine structures with slow - mode shocks attached at the edges of the plasmoids are formed . unlike the classical petschek slow - mode shocks , these shock structures are not steady and they can be distorted by the colliding plasmoids and turbulent outflows .
lots of turbulent structures appear inside the multiple plasmoids and in the down flow region .
the termination shocks can also be formed above the primary magnetic island and at the head of secondary islands .
these shocks play important roles in generating thermal energy in a corona current sheet .
\2 . for a numerical simulation with initial conditions @xmath62
, @xmath59 t and @xmath256 kg m@xmath66 , about @xmath257 of the dissipated magnetic energy is converted to kinetic energy before secondary islands appear .
after multiple slow - mode shocks appear at the edges of the magnetic islands , the generated thermal energy increases sharply , about @xmath258 of the dissipated magnetic energy can be transformed to thermal energy eventually .
the one dimensional energy spectra along the current sheet at @xmath145 have been studied .
after secondary islands appear , the average spectrum index for kinetic energy is around @xmath224 and it is around @xmath3 for the magnetic energy spectrum .
these spectra do not behave as a simple power law and the spectrum index increases with the wave number , which are similar as the previous studies @xcite .
for the first time , we have studied the two - dimensional energy spectra of the corona current sheet . comparing with the one dimensional spectra , two dimensional spectra
intuitively show that part of the high energy is cascaded to larger @xmath4 and @xmath5 space after secondary islands appear .
the spectra are asymmetric in the fourier space , most of the energy is always confined in the region with small @xmath5 .
the plasmoid distribution function has been calculated numerically by @xmath248 .
it behaves as a power law closer to @xmath6 in the intermediate @xmath7 regime , which is the same as the result from @xcite . in the large @xmath7 regime ,
the distribution function @xmath254 gradually deviates from the power law @xmath255 to a more rapid falloff .
however , the exponential tail in the large @xmath7 regime as presented in the paper by @xcite is not clearly identified from our data presented in figure .
it is probably because that we only use the snapshots before @xmath259 s of case b data and the plasmoids we have identified are still relative small .
as we know , the effect of heat conduction can smooth out the temperature in solar corona magnetic reconnection events .
the numerical simulations in the papers by @xcite have proved this point .
our previous paper @xcite also studied the effect of anisotropic heat conduction on the magnetic reconnection process . since the time step @xmath260 for explicit scheme used in the nirvana code
is limited by cfl condition , @xmath261 , where @xmath262 is the space step and @xmath263 is the thermal conductivity coefficient .
after secondary islands and smaller scale current sheet fragments appear , the extremely small @xmath262 and large @xmath263 make the time step @xmath260 becoming too small to continue the simulations for including the heat conduction in both case a and case b. therefore , the heat conduction terms are not included in this work .
the characteristic time scales of heat conduction in both the directions parallel and perpendicular to magnetic fields have been analytically calculated to compare with the alfvn crossing time .
we find that the heat conduction effect could be very efficient at the high temperature regions in the direction parallel to magnetic fields , but it can be ignored in the direction perpendicular to the magnetic fields . as shown in figure . 5 and 7 ,
much higher temperatures and steep temperature gradients are built up in the plasmoids and at the slow - mode shocks and x - points .
conductive energy transport from these regions into the inflow regions and into the big plasmoid at the bottom end of the current sheet is largely directed across the magnetic field .
therefore , even though the anisotropic heat conduction is included , the high temperature plasmas can still be confined inside the plasmoids in 2.5-dimensional simulations .
but , the heat conduction can be efficient in the third direction and the maximum temperature in plasmoids ( flux ropes ) will become smaller in full three dimensional space than that in 2.5-dimensional simulations . as mentioned in the paper by brta et al.(2011a , 2011b ) , the actual physical mechanism that provides
the energy transfer from the global scales , at which the energy is accumulated , to the much smaller scales , at which the plasma - kinetic dissipation takes place , is an open issue .
the spectrum studies presented in figure .
11 , 12 and 13 prove that the multiple cascading process is happening or already happened in the current sheet region , both kinetic and magnetic energy are cascaded from large scales to small scales during the plasmoid cascading process .
therefore , such kind of turbulent reconnection with multiple dynamic structures can explain well the energy transfer process in the solar flare eruption .
the hxr and radio observations ( e.g. , karlick et al . 1996 , 2000 ) also indicate that the particle acceleration takes place via multiple concurrent small - scale events distributed turbulently in the flare volume , rather than by a single compact acceleration process hosted by a single diffusion region .
such observations are usually referred to as signatures of fragmented / chaotic energy release in flares .
the spectra presented in figure .
11 , 12 and 13 can not be directly compared with the observation spectra right now . in the future
, we hope that the dynamic structures in the plasmoid instability process in our simulations can be used to study the particle acceleration by the testing particle method ( e.g. , li & lin 2012 ) .
a huge number of particles can be placed separately in the dynamic structures with turbulent current sheet fragments , slow - mode shocks or fast - mode shocks .
then we can analyze how these particles will be accelerated and calculate the particle energy spectrum distributions . according to these spectra
, one can expect to find out the contributions of turbulent current sheet fragments , slow - mode shocks and fast - mode shocks to particle accelerations separately .
finally , these spectra about the particle energy and number density distributions can be compared with observations .
we would like to thank the referee , dr .
yokoyama takaaki , dr .
roney keppens and yi - min huang for their helpful comments and discussions .
this research is supported by nsfc ( grant no .
11203069 ) , the key laboratory of solar activity grant klsa201404 , the western light of chinese academy of sciences 2014 , the program 973 grants 2013cba01503 , nsfc grant 11273055 , nsfc grant 11333007 , cas grant kjcx2-ew - t07 and cas grant xdb09040202 .
we have used the nirvana code v3.6 developed by udo ziegler at the leibniz - institut fr astrophysik potsdam .
the authors gratefully acknowledge the computing time granted by the yunnan astronomical observatories , and provided on the facilities at the yunnan astronomical observatories supercomputing platform , as well as the help from all faculties of the platform .
asai , a. , yokoyama , t. , shimojo , m. , & shibata , k. 2004 , , 605 , l77 brta , m. , bchner , j. , karlick , m. , & skla , j. 2011 , , 737 , 24 brta , m. , bchner , j. , karlick , m. , & kotr , p. 2011
, , 730 , 47 bhattacharjee , a. , huang , y .- m . , yang , h. , & rogers , b. 2009 , physics of plasmas , 16 , 112102 bemporad , a. 2008 , , 689 , 572 birn , j. , fletcher , l. , hesse , m. , & neukirch , t. 2009 , , 695 , 1151 cheng , x. , zhang , j. , liu , y. , & ding , m. d. 2011 , , 732 , ll25 ciaravella , a. , & raymond , j. c. 2008 , , 686 , 1372 dong , q .- l . , wang , s .- j . , lu , q .- m . , et al . 2012 , physical review letters , 108 , 215001 fermo , r. l. , drake , j. f. , & swisdak , m. 2010 , physics of plasmas , 17 , 010702 frank , a. , jones , t. w. , ryu , d. , & gaalaas , j. b. 1996 , , 460 , 777 guo , l .- j
. , bhattacharjee , a. , & huang , y .- m .
2013 , , 771 , l14 karlick , m. , sobotka , m. , & jiika , k. 1996 , , 168 , 375 karlick , m. , jiika , k. , & sobotka , m. 2000 , , 195 , 165 kliem , b. 1990 , astronomische nachrichten , 311 , 399 ko , y .- k . ,
raymond , j. c. , vrnak , b. , & vuji , e. 2010 , , 722 , 625 krucker , s. , hudson , h. s. , glesener , l. , et al .
2010 , , 714 , 1108 kulsrud , r. m. 2001 , earth , planets , and space , 53 , 417 kumar , p. , & cho , k .- s .
2013 , , 557 , a115 baty , h. , & keppens , r. 2002 , , 580 , 800 habbal , s. r. , holzer , t. e. , & leer , e. 1979 , , 64 , 287 hsieh , m. h. , tsai , c. l. , ma , z. w. , & lee , l. c. 2009 , physics of plasmas , 16 , 092901 huang , y .- m . , & bhattacharjee , a. 2010 , physics of plasmas , 17 , 062104 huang , y .- m . , & bhattacharjee , a. 2012 , physical review letters , 109 , 265002 innes , d. e. , mckenzie , d. e. , & wang , t. 2003 , , 217 , 267 lazarian , a. , & vishniac , e. t. 1999 , , 517 , 700 landi , e. , raymond , j. c. , miralles , m. p. , & hara , h. 2012 , , 751 , 21 li , y. , & lin , j. 2012 , , 279 , 91 lin , j. , ko , y .- k . , sui , l. , et al .
2005 , , 622 , 1251 lin , j. , li , j. , forbes , t. g. , et al .
2007 , , 658 , l123 lin , j. , cranmer , s. r. , & farrugia , c. j. 2008 , journal of geophysical research ( space physics ) , 113 , 11107 liu , r. , lee , j. , wang , t. , et al .
2010 , , 723 , l28 liu , r. 2013 , mnras , 434 , 1309 liu , w. , chen , q. , & petrosian , v. 2013 , , 767 , 168 loureiro , n. f. , samtaney , r. , schekochihin , a. a. , & uzdensky , d. a. 2012 , physics of plasmas , 19 , 042303 masuda , s. , kosugi , t. , hara , h. , tsuneta , s. , & ogawara , y. 1994 , , 371 , 495 mei , z. , shen , c. , wu , n. , et al .
2012 , , 425 , 2824 milligan , r. o. , mcateer , r. t. j. , dennis , b. r. , & young , c. a. 2010 , , 713 , 1292 nagai , f. 1980 , , 68 , 351 ni , l. , ziegler , u. , huang , y .- m . ,
lin , j. , & mei , z. 2012 , physics of plasmas , 19 , 072902 ni , l. , roussev , i. i. , lin , j. , & ziegler , u. 2012 , , 758 , 20 ni , l. , lin , j. , & murphy , n. a. 2013 , physics of plasmas , 20 , 061206 ni , l. , kliem , b. , lin , j. , & wu , n. 2015 , , 799 , 79 reeves , k. k. , linker , j. a. , miki , z. , & forbes , t. g. 2010 , , 721 , 1547 samtaney , r. , loureiro , n. f. , uzdensky , d. a. , schekochihin , a. a. , & cowley , s. c. 2009 , physical review letters , 103 , 105004 savage , s. l. , & mckenzie , d. e. 2011 , , 730 , 98 savage , s. l. , holman , g. , reeves , k. k. , et al .
2012 , , 754 , 13 shen , c. , lin , j. , & murphy , n. a. 2011 , , 737 , 14 shen , c. , lin , j. , murphy , n. a. , & raymond , j. c. 2013 , physics of plasmas , 20 , 072114 shibata , k. , & tanuma , s. 2001 , earth , planets , and space , 53 , 473 su , y. , veronig , a. m. , holman , g. d. , et al .
2013 , nature physics , 9 , 489 sun , j. q. , cheng , x. , & ding , m. d. 2014 , , 786 , 73 sui , l. , & holman , g. d. 2003 , , 596 , l251 susino , r. , bemporad , a. , & krucker , s. 2013 , , 777 , 93 takasao , s. , asai , a. , isobe , h. , & shibata , k. 2012 , , 745 , l6 tanuma , s. , yokoyama , t. , kudoh , t. , & shibata , k. 2001 , , 551 , 312 tsuneta , s. , & naito , t. 1998 , , 495 , l67 ugai , m. 1992 , physics of fluids b , 4 , 2953 uzdensky , d. a. , loureiro , n. f. , & schekochihin , a. a. 2010 , physical review letters , 105 , 235002 wang , t. , sui , l. , & qiu , j. 2007 , , 661 , l207 yokoyama , t. , & shibata , k. 1997 , , 474 , l61 yokoyama , t. , & shibata , k. 2001 , , 549 , 1160 zhang , q. m. , & ji , h. s. 2014 , , 567 , aa11 ziegler , u. 2011 , journal of computational physics , 230 , 1035 | in this paper , we study the energy conversion and spectra in a corona current sheet by 2.5-dimensional mhd numerical simulations .
numerical results show that many petschek - like fine structures with slow - mode shocks mediated by plasmoid instabilities develop during the magnetic reconnection process .
the termination shocks can also be formed above the primary magnetic island and at the head of secondary islands .
these shocks play important roles in generating thermal energy in a corona current sheet . for a numerical simulation with initial conditions close to the solar corona environment ,
the ratio of the generated thermal energy to the total dissipated magnetic energy is around @xmath0 before secondary islands appear .
after secondary islands appear , the generated thermal energy starts to increase sharply and this ratio can reach a value about @xmath1 . in an environment with a relatively lower plasma density and plasma @xmath2 , the plasma can be heated to a much higher temperature .
after secondary islands appear , the one dimensional energy spectra along the current sheet do not behave as a simple power law and the spectrum index increases with the wave number .
the average spectrum index for the magnetic energy spectrum along the current sheet is about @xmath3 .
the two dimensional spectra intuitively show that part of the high energy is cascaded to large @xmath4 and @xmath5 space after secondary islands appear .
the plasmoid distribution function calculated from numerical simulations behaves as a power law closer to @xmath6 in the intermediate @xmath7 regime . by using @xmath8 , the effective magnetic diffusivity
is estimated about @xmath9 m@xmath10s@xmath11 . |
disk galaxies are the main population of galaxies in the local universe .
the study of their global scaling relations is of paramount relevance for understanding the formation and evolution of galaxies in general .
the main disk
galaxy observed global scaling relations are those between maximum circular velocity , , and luminosity , @xmath0 ; between disk scale length , @xmath2 , and @xmath0 ; and between and @xmath2 .
each one of these relations can be established for different optical / near ir ( nir ) pass bands . under some assumptions , analogous relations can be calculated for the corresponding stellar and baryonic quantities .
the changes in the zero points , slopes and scatters of these relations , when changing from one color band to another , and to stellar and baryonic quantities , are expected to be intimately related to the stellar population properties and radial distributions of disk galaxies . in turn , these properties and distributions depend mainly on the disk assembly and star formation ( hereafter sf ) histories .
therefore , the study of the scale relations for different passbands and for the stellar and baryonic cases offers valuable information on the structural , dynamical and sf properties of disk galaxies , as well as important constraints on models of their formation and evolution .
the present paper is focused on this study .
the most robust and studied of the disk galaxy scaling relations is the one between @xmath0 and , the so called tully fisher relation ( tfr ) . originally discovered in the @xmath3 passband ( tully & fisher 1977 ) , it was afterward shown that it also applies to infrared bands . more recently , after applying a velocity ( luminosity)dependent extinction correction , tully & pierce ( 2000 ) have confirmed that the slope of the @xmath0-@xmath11 correlation gets steeper systematically from @xmath3 ( @xmath12 ) to @xmath4 ( @xmath13 ) bands .
this result suggests some dependence of galaxy colors on for a given galaxy luminosity or mass .
the amplitude of the scatter of the tfrs changes also with wavelength .
these dependencies of the slope and scatter of the tfr on wavelength are related mainly to stellar population effects , which on their own , are related to the assembly and sf histories of galaxies . regarding the intrinsic tfr scatters ,
they are found to be small , posing a challenge for scenarios aimed to explain the origin of disk galaxies ( eisenstein & loeb 1996 ; avila - reese et al .
1998 ; mo et al . 1998 ; firmani & avila - reese 2000 ) . the tfr has widely been studied for estimating extragalactic distances . in this case
the independent variable ( predictor ) is the circular velocity , and the galaxy samples are carefully pruned to minimize the scatter .
the latter implies a selection of samples for a narrow range of morphological types and surface brightnesses ( hereafter sbs ) .
however , for studies that aim to understand the origin of the tfr and use it to constrain models of galaxy formation rather than circular velocity , the luminosity should be the predictor , and the sample should be as wide as possible in morphological types , sbs , and colors ( courteau et al . 2007 ) .
hereafter , we will refer to the tfr as the -@xmath0(-@xmath6 ) correlation . by using nir magnitudes which are good tracers of stellar masses and gas contents
when available , some authors have calculated the stellar and baryonic tfrs ( e.g. , mcgaugh et al .
2000 ; bell & de jong 2001 ; mayer & moore 2004 ; gurovich et al .
2004 ; pizagno et al . 2005 ; de rijcke et al .
2007 ) . as discussed in firmani & avila - reese ( 2000
; hereafter fa - r ) , on one hand the baryonic tfr appears to be closely related to the cosmological halo maximum circular velocity mass relation , and on the other hand , when compared to the stellar tfr , it reflects some aspects of the sf efficiency for different types of galaxies .
the behavior of the scatter , when one changes from @xmath8 to @xmath14band tfrs and to the stellar and baryonic tfrs , certainly contains relevant information on the processes of disk galaxy evolution such as gas infall , sf , and stellar population histories .
the radius for a given luminosity ( or the disk sb ) has been considered as a potential third parameter that correlates with the scatter of the tfr ( e.g. , kodaira 1989 ; willick 1999 ) .
the scale length @xmath2 is indeed another structural parameter of disk galaxies that span a wide range of values .
some authors have reported that the scatter of the tfr is reduced up to 50% when @xmath2 is introduced as a third parameter ( e.g. , koda , sofue & wada 2000 ; han et al .
however , if the scatter of a given relation decreases upon introducing a third variable , it does not necessarily mean that such a variable is a statistically significant parameter .
indeed , some authors have shown that @xmath2 ( or central sb ) is not a physical third parameter in the tfr ( e.g. , han 1991 ; tully & verheijen 1997 ; willick et al .
1997 ; courteau & rix 1999 ; fa - r ) , but the radius could be a third parameter in the baryonic tfr ( fa - r ; avila - reese , firmani & zavala 2002 ; zavala 2003 ; dutton et al . 2007 ) .
it is well known that the radius correlate with @xmath0 ( freeman 1970 ) and ( tully & fisher 1977 ) , although these correlations are much more scattered than the tfr .
disk galaxies in the log@xmath0log@xmath11log@xmath2 parameter space indeed lie in a thin oblique plane whose edge on projection is close to the log@xmath15log ( tfr ) projection ( koda , sofue & wada 2000 ; han et al .
2001 ; shen , mo & shu 2002 ; gnedin et al .
this is the so called fundamental plane of disk galaxies .
here we construct from observations and analyze separately each one of the three projections of these planes . in most of the previous works , the disk
galaxy scaling correlations were constructed for samples biased in morphological types and sbs ( pruned for reducing the scatter ) , and without selecting the environment .
however , _ for theoretical and interpretative studies , the observational galaxy sample should be as wide as possible in the range of morphologies , sbs , and colors _ in order to include the whole population of disk galaxies , later to be compared with model predictions .
besides , since galaxy models are commonly used for isolated galaxies , the observational sample should also be selected to contain galaxies _ not affected by the environment_. on the other hand , the limited information of the samples employed in previous works , did not allow one to explore how the scaling correlations and their scatters change from luminous to stellar and baryonic quantities . in this paper we shall use a sample of @xmath16 disk galaxies compiled from the literature and homogenized in zavala et al .
( 2003a ; hereafter zahf ) and zavala ( 2003 ) .
this sample is not in any sense complete , an aspect that is crucial for volume averaged population studies ( e.g. , the luminosity function ) , but is less relevant for studies on the correlations of physical properties .
the sample includes high and low sb ( hereafter hsb and lsb , respectively ) disk _ normal _ galaxies of all morphological types , and it provides photometric information , at least in @xmath3 and @xmath4 bands , as well as information on the rotation curve and the hi integrated flux .
none of the previous related works have presented a sample with these characteristics , which are crucial for the analysis proposed here . by using this sample
, we will explore the differences in the slope and scatter of each scaling relation defined in the @xmath3 and @xmath4 bands , and for stellar and baryonic quantities .
the results of this exploration will be interpreted in the light of disk galaxy formation and evolution models .
the current theory of galaxy formation and evolution is based on the hierarchical @xmath10 cold dark matter ( ) cosmological model ( for recent reviews , see baugh 2006 ; avila - reese 2007 ) .
several authors have attempted to predict the scaling relations of disk galaxies and their scatters within this model ( e.g. , avila - reese et al .
1998,2002 ; mo et al .
1998 ; steinmetz & navarro 1999 ; navarro & steinmetz 2000 ; fa - r ; koda et al .
2000 ; shen et al .
2002 ; zavala 2003 ; courteau et al .
2003,2007 ; dutton et al .
2005,2007 ; gnedin et al .
2007 ; courteau et al . 2007 ) .
the results are encouraging .
we shall use some of these results , mainly those of fa - r , to interpret and discuss the new inferences from observations presented here .
note that it was a common practice in the literature to compare model results with the observed _ luminous _ properties and correlations , while most of the galaxy models did not actually include or self consistently treat the process of gas transformation into stars , as well as did not follow the evolutionary path of halos and galaxies .
as fa - r showed ( see also avila - reese et al .
2002 ; dutton et al .
2005 ) , on one hand , there are systematical differences in the predicted scaling relations and their scatters among the luminous , stellar , and baryonic quantities ; differences that we will explore here .
on the other hand , properties such as the stellar and baryonic @xmath6 , @xmath2 , , gas fraction , and integral colors depend not only on the initial parameters , but also on the evolutionary path of the halo disk system ( stringer & benson 2008 ) .
contrary to the `` static population '' models , such as those of mo et al .
( 1998 ) and extensions of them , the fa - r models are evolutionary . with some simplifications , they follow self consistently ( see appendix b ) : the halo and disk assembly , taking into account the halo adiabatic contraction ; the self regulated disk sf and feedback ; the luminosity evolution ; and the secular bulge formation . during the completion of the present work , several of the papers mentioned above , with results and a philosophy similar to ours , appeared in the literature .
although some overlapping is present , our contribution is original in the sense that we focus the research _ on the changes in the scaling relations and their scatters when moving from optical to nir bands and to stellar and baryonic quantities_. our sample is similar in number to others used in recent works ( e.g. , gnedin et al .
2007 ) and smaller with respect to dutton et al .
( 2007 ) and courteau et al .
( 2007 ) , who use @xmath17 objects with @xmath18 band photometry and @xmath19 kinematics , but only @xmath20 out of them , mainly hsb galaxies , have available @xmath14 band photometry .
these large samples are undoubtedly useful and we will compare , when possible , with their results . however , we can not use them because they lack several observables ( e.g. , gas content , optical photometry ) and requirements ( e.g. , inclination limits , lsb galaxies in both optical and @xmath4 bands ) that we have imposed for our study .
we hope that our study will serve as a benchmark for similar analyzes based on large multiwavelength galaxy surveys , which are currently being completed .
this paper is organized as follows . in
2 we describe the disk galaxy sample to be used .
the three main scaling correlations and their scatters are constructed and discussed in 3 for bands @xmath3 and @xmath4 and for the inferred stellar and baryonic quantities .
some other correlations among global galaxy quantities are also reported . in 4 , in the light of previous based disk galaxy formation and evolution models , we interpret the changes in the slopes and scatters of the scaling relations when changing from @xmath3 to @xmath4 bands and to stellar and baryonic quantities .
a summary of the results and the conclusions of this work are presented in 6 .
in the literature there are only a few small galaxy samples with the requirements and data we need ( broad in morphological types and sbs , with information on detailed surface
photometry in two separated pass bands , e.g. @xmath3 and @xmath4 , on global dynamics , and on total hi gas flux ) .
here we use a compilation and completion of these samples presented in zahf and zavala ( 2003 ; see also graham 2002 ) . in the following , we present the sub
samples , describe the homogenization and data correction processes that we have implemented , and the way in which we calculated the stellar and baryonic quantities ( see zahf for additional details ) . the sub
samples are : ( i ) de jong & van der kruit ( 1994 ; see also de jong 1996a ) , which comprises undisturbed field , mostly hsb , disk galaxies , ( ii ) verheijen ( 1997 ; see also verheijen & sancisi 2001 ) , composed of disk hsb and lsb galaxies from the least massive and most spiral rich nearby cluster , ursa minor , and ( iii ) bell et al .
( 2000 ) , a sample of lsb galaxies observed partially by them ; the rest were compiled from the literature , mainly from de blok et al .
( 1995,1996 ) . in our sample
we also included the milky way and andromeda galaxies .
the raw photometric magnitudes , disk scale lengths , @xmath2 , and central sbs were taken directly from the source papers . in zahf and zavala ( 2003 )
we have checked that there were no significant systematic differences in these parameters with inclination ; we also found no systematics due to the different completeness criteria and profile fitting procedures used by the original authors of each sub sample .
in fact , the dominant source of error in the photometric parameters is caused by the uncertainty in the sky level ( bell et al 2000 ; de jong 1996a ) . regarding the dynamical information , and to make the data as uniform as possible , we used the velocity line width ( defined at the 20% level ) for almost all the galaxies in the sample .
only a few ( lsb ) galaxies in the sample lack a measurement but have measured rotation curves ; in these cases we have used the inferred directly from the rotation curve . if properly corrected , @xmath21 ( verheijen 1997 ; verheijen 2001 ) . for the verheijen sub
sample , we adopt the values given in verheijen & sancisi ( 2001 ) , while for the de jong sub sample and most of the lsb galaxies from the bell et al . sub sample , was extracted from the hyperleda information system .
the integral fluxes in the 21cm line ( used to calculate the neutral gas mass component in galaxies ) were taken from hyperleda , verheijen & sancisi ( 2001 ) , and de blok et al .
( 1996 ) .
a condition imposed on our sample is that galaxies should be in a restricted range of inclinations ( @xmath22 ) .
we excluded from the sample any galaxies with clear signs of interaction ( mainly from the verjeihen sub
sample ) and with rotation curves that are still increasing at the last measured outer radius .
only some lsb galaxies showed this feature .
the final sample consists of 76 galaxies : 42 out of 86 , 29 out of 52 , and 5 ( 8) out of 23 from the de jong ( 1996a ) , verheijen ( 1997 ) , and bell et al .
( 2000 ) sub samples , respectively , plus the milky way and andromeda galaxies . from the bell et al .
( 2000 ) sample , 8 lsb galaxies are useful for us , but three of them are in common with the de jong sample .
the local distances to the galaxies were calculated using the kinematic distance modulus given in hyperleda ( see more details in zahf ) .
the value of the hubble constant used in this database is the same as that we assume here , @xmath23mpc@xmath24 . since the lsb galaxies from the bell et al .
( 2000 ) sub sample are not included in hyperleda , their distances were taken directly from the source paper , properly corrected for the h@xmath25 value used here .
the compiled sample is not complete in any sense , but it comprises representative ranges of the basic features of _ normal _ disk galaxies ( see fig . 1 in zahf ) .
these features are mainly the morphological type , magnitude , disk central sb , integral color , and gas fraction .
the total magnitudes were corrected for galactic extinction ( following schlegel et al .
1998 ) , @xmath4 term , and internal extinction ( see zahf for details ) . for the latter we used the empirical velocity ( luminosity ) dependent extinction coefficients determined by tully et al .
the central sbs were corrected for galactic extinction , @xmath4 term , cosmological sb dimming , and inclination geometrical and extinction effects ( see zahf ) . for the inclination correction
, we followed verheijen ( 1997 ) , considering lsb galaxies optically thin in all bands .
the hsb galaxies were also considered optically thin in the @xmath4 band .
we have defined lsb galaxies as those with central sbs in the @xmath4 band , after sb correction , larger than 18.5 mag / arcsec@xmath26 ( verheijen 1997 ) . the 21 cm line widths at the 20% level , , for the verheijen susbsample were taken directly from verheijen & sancisi ( 2001 ) .
galaxies with taken from leda were de corrected ( for instrumental corrections ) to get the raw data ( paturel et al . 1997 ) , and then corrected again for broadening due to turbulent motions and for inclination , this time following the procedure by verheijen & sancisi ( 2001 ) .
in zahf , we did not take into account the observational errors .
here we attempt to estimate these errors for both the photometric and dynamic parameters , and use them for our results when possible . in appendix
a , we present our approach to estimate the errors . the stellar mass and disk central surface density , and , were derived from the @xmath14band luminosity ( which includes the bulge ) and the extrapolated disk central sb , , respectively , by using the appropriate stellar mass to light ratio , @xmath27 .
the application of population synthesis techniques to disk galaxy evolution models shows that @xmath27 depends mainly on the integral color ( tinsley 1981 ; bruzual 1983 ; bell & de jong 2001 ) . in zahf
we have used a @xmath27 inferred from the latter paper . in a more recent work , bell et al .
( 2003a ) used galaxy evolution model fits for a large sample of galaxies from the two micron all sky survey ( 2mass ) and the sloan digital sky survey ( sdss ) .
they obtained a correlation of @xmath27 with @xmath28 color which is shallower and more scattered than that in bell & de jong ( 2001 ) .
unfortunately , bell et al . (
2003a ) did not report the correlation of @xmath27 with the @xmath5 color .
we have obtained this correlation by a re scaling procedure that makes use of the @xmath27-@xmath28 correlation given in bell at al .
( 2003a ) , and the @xmath27-@xmath28 and @xmath27-@xmath5 correlations given in bell & de jong ( 2001 ; their model `` formation epoch with bursts '' was used ) .
we obtain the following result : @xmath29 for low sb values , at least for blue galaxies , @xmath27 seems to anti - correlate with color ( verheijen 1997 ) . in fig . 1 of bell & de jong ( 2001 ) ,
it is also shown how the inferred values of @xmath27 , for low sbs , no longer decrease with decreasing sb , and even start to increase as sb decreases , although the scatter is large . in the absence of detailed models , we approximate the @xmath27-@xmath5 correlation for lsb galaxies in such a way that for @xmath30 , @xmath27 is given by eq .
( [ gammahsb ] ) , while for @xmath31 , @xmath32 the latter dependence is from a linear eye
fit to the @xmath27-@xmath5 correlation of the _ blue _ lsb galaxies in verheijen ( 1997 ) , for his case of a hernquist halo model and the constrained decomposition method . after using the new @xmath27-@xmath5 correlations ,
the stellar and baryonic masses of the galaxy calculated here are slightly different from those used in zahf .
the scale length of the stellar disk , , is assumed to be equal to the scale length in the band @xmath4 , .
most of the stars in the disk are properly traced using this band .
the disk gas mass , , is estimated as : @xmath33\ ] ] where the factor 1.4 takes into account helium and metals , and m@xmath34 is the mass in molecular hydrogen .
the m@xmath35m@xmath36 ratio has been found to depend on the morphological type @xmath37 ( young & knesek 1989 ) . using the latter paper ,
mcgaugh & de blok ( 1997 ) estimated that m@xmath38m@xmath39 .
for t@xmath40 , this empirical fitting formula could overestimate the gas mass in galaxies , thus we assume : m@xmath38m@xmath41 ( the value of the fit at @xmath42 ) for @xmath43 .
the galaxy baryonic mass is defined as = + , and the gas mass fraction as = /.
unfortunately , we do not have information on the gas surface density profile parameters ( @xmath44 and h@xmath45 ) for the galaxies in our sample .
we need them in order to calculate the baryonic disk central surface densities , , and scale lengths , . as discussed in zahf , we assume that the total gas surface density follows an exponential distribution with a scale length 3 times that of .
thus , @xmath46 . the baryonic quantity is then calculated as = + @xmath47 . the corresponding baryonic radius will be then @xmath48^{0.5}$ ] .
in fact , the addition of the ( uncertain ) gas disk parameters will have little impact on our final results .
this inclusion is important only for gas rich galaxies ( late type , blue lsb galaxies ) .
we believe that in spite of the uncertainties , some real systematical variations of the baryonic disk scale length and surface density with the gas content of galaxies can still be taken into account through our procedure . we have also estimated the errors in the composite quantities by adequately propagating the errors of the primary quantities .
see appendix a for details .
for the sample of 76 galaxies presented in 2 , we proceed to construct the physical scaling relations @xmath11-@xmath0 ( -@xmath6 ) , @xmath2-@xmath0 ( -@xmath6 ) , and @xmath11-@xmath2 in the bands @xmath3 and @xmath4 , and for stars and baryons .
some authors recommend to use * @xmath49 * , the outer asymptotic flat part of the rotation curve , as an estimate of the circular velocity , instead of . we stress that for studies aiming to understand the origin of the tfr and its scatter ( the same applies for the @xmath11-@xmath2 relation ) as well as to compare them with galaxy model predictions , the adequate quantity to be used is _ because it maximizes the contribution of the disk component for a given halo_. by using @xmath49 , which traces the dynamics of the galaxy in outer regions , where the halo component tends to dominate , we lose information about the disk contribution .
in fact , here we use neither nor @xmath49 , but .
however , as mentioned in 2 , there is some evidence that @xmath50 . for the luminosities and masses ,
here we use the total ones , i.e. the sum of disk and bulge .
for the cases when bulges form by disk secular evolution processes ( e.g. avila - reese & firmani 2000 ; avila - reese et al .
2005 ) , this is a good approximation for total disk luminosities and masses . for the scale length , the corresponding @xmath3 and @xmath4 band radii are used in the luminous scaling relations . for the stellar and baryonic scaling relations we assume respectively
= and a scale length , , which takes into account the sum of the stellar and gas disks ( see 2.2 ) . in figs .
[ tf][vr ] , the observational points with the estimated error bars are plotted in the diagrams corresponding to the disk galaxy scaling relations studied here . the variables are given logarithmically .
circles and triangles are used for galaxies with @xmath51 ( red ) and @xmath52 ( blue ) , respectively .
the open and solid symbols are for galaxies with @xmath53 ( hsb ) and @xmath54 ( lsb ) , respectively .
to quantify the correlations and compare them among the different pass bands and mass quantities ( and eventually to model predictions ) , one needs to fit the correlations . to this end
, a linear regression analysis should be used .
we recall that our aim here is to explore variations for the scaling relations in the different cases ( @xmath3 and @xmath4 bands , stars , and baryons ) rather than study in detail the relations for each one of the cases .
therefore , it is important for us _ to use the same regression method for all the cases rather than to choose an optimal method_. for studies aimed at theoretical interpretations it is better to use a linear fit that is symmetric to interchanges of the two variables , e.g. the bisector or orthogonal regression linear methods .
here we fit the correlations using the forward , backward , bisector , and orthogonal linear regression models .
however , for further analysis and discussion , _ we will use only the orthogonal regression results_. the origin of coordinates for each correlation is shifted to the barycenter of the corresponding variables , where the slope and zero point are uncorrelated .
the corresponding linear regressions , _ not taking into account _ the errors in the variables , were carried out for our galaxy sample in zavala ( 2003 ) by using the slopes routine ( isobe et al .
1990 ; feigelson & babu 1992 ) .
this seems to be a feasible approximation when the intrinsic scatter in the regression line dominates any errors arising from the measurement and error propagation procedures ( isobe et al .
the tfr is the tightest one among the scaling relations .
studies of the tfr have shown indeed that the intrinsic scatter of the linear fit is larger than the measurement and correction errors ( for both magnitude and velocity ) .
for example , giovanelli et al . (
1997 ) showed that the latter errors account for @xmath55 mag ( from smaller to larger velocities ) , while the total scatter to the fit is @xmath56 mag for the same range of velocities , implying that the _ intrinsic _ scatter dominates over the measurement and correction uncertainties . for the other scaling relations ,
this difference is expected to be even larger .
since we have obtained estimates of the errors in the variables of the scaling relations ( appendix a ) , here we carry out the corresponding `` error in variable '' regression models .
thus , the analysis in the present paper is more exact than in zavala ( 2003 ) and it allows us to roughly estimate the values of the intrinsic scatter in the different scaling relationships . because the errors differ from point to point , the model should be heteroscedastic .
we assume that the scaling relations have an ( unknown ) intrinsic scatter , which we want to estimate , explore , and confront with theory .
the fitting regressions taking into account heteroscedastic measurement errors are carried out using the method and routine presented in akritas & bershady ( 1996 ) .
the total average intrinsic scatter , , is estimated as the square root of the variance of the fit , @xmath57 , subtracted from the average ( in both variables ) observable variance , @xmath58 ( @xmath59 $ ] , where @xmath60 is the number of data points and @xmath61 is the slope of the correlation ) , i.e. , @xmath62 .
more elaborate fitting procedures could be used to estimate the intrinsic scatter ( e.g. , gnedin et al .
however , given the level of approximation we have used to determine the errors for the data , we consider that our first
order approximation is enough .
tables @xmath63 show the results from the different regression methods applied to the three scaling correlations in the four cases ( baryonic and stellar quantities , and @xmath4 and @xmath3 bands ) .
we report the zero point , @xmath64 , the slope and @xmath61 , with their respective standard deviations ( calculated in the barycenter of the variables ) , as well as the corresponding square roots of the variance of the fit , , the bi weighted average of the individual ( in both variables ) variances , , and the total average intrinsic variance , .
we recall that our analysis , including the scatters , refers to the logarithm of the quantities involved . in the panels of figs .
[ tf ] to [ vr ] , we plot the forward ( dashed line ) , backward ( dotted line ) , and orthogonal ( solid line ) doubly weighted linear fits .
llllllll [ tab : tf ] + forward & 0.303 & 0.012 & -0.979 & 0.127 & 0.029 & 0.058 & 0.050 + inverse & 0.333 & 0.014 & -1.300 & 0.151 & 0.031 & 0.063 & 0.055 + bisector & 0.317 & 0.012 & -1.140 & 0.132 & 0.030 & 0.060 & 0.052 + orthog . &
0.306 & 0.012 & -1.024 & 0.129 & 0.028 & 0.058 & 0.051 + - + forward & 0.274 & 0.011 & -0.639 & 0.121 & 0.036 & 0.058 & 0.045 + inverse & 0.294 & 0.013 & -0.843 & 0.137 & 0.038 & 0.061 & 0.048 + bisector & 0.284 & 0.012 & -0.741 & 0.127 & 0.037 & 0.059 & 0.046 + orthog .
& 0.274 & 0.012 & -0.650 & 0.123 & 0.035 & 0.058 & 0.045 + -@xmath65 + forward & 0.258 & 0.012 & -0.503 & 0.126 & 0.014 & 0.052 & 0.050 + inverse & 0.272 & 0.012 & -0.652 & 0.132 & 0.015 & 0.057 & 0.054 + bisector & 0.272 & 0.012 & -0.652 & 0.132 & 0.014 & 0.054 & 0.052 + orthog .
& 0.261 & 0.012 & -0.520 & 0.128 & 0.014 & 0.051 & 0.049 + -@xmath66 + forward & 0.310 & 0.015 & -0.917 & 0.152 & 0.016 & 0.064 & 0.062 + inverse & 0.361 & 0.016 & -1.430 & 0.169 & 0.018 & 0.073 & 0.072 + bisector & 0.335 & 0.014 & -1.170 & 0.149 & 0.017 & 0.068 & 0.066 + orthog .
& 0.314 & 0.015 & -0.963 & 0.154 & 0.016 & 0.065 & 0.063 + llllllll [ tab : rl ] + forward & 0.287 & 0.035 & -2.510 & 0.379 & 0.065 & 0.207 & 0.196 + inverse & 0.692 & 0.098 & -6.770 & 1.030 & 0.085 & 0.032 & 0.313 + bisector & 0.474 & 0.042 & -4.470 & 0.444 & 0.073 & 0.236 & 0.225 + orthog .
& 0.322 & 0.039 & -2.870 & 0.422 & 0.067 & 0.208 & 0.197 + - + forward & 0.285 & 0.031 & -2.510 & 0.327 & 0.061 & 0.196 & 0.187 + inverse & 0.592 & 0.073 & -5.700 & 0.765 & 0.087 & 0.282 & 0.269 + bisector & 0.430 & 0.034 & -4.020 & 0.366 & 0.072 & 0.218 & 0.206 + orthog . &
0.310 & 0.033 & -2.770 & 0.352 & 0.063 & 0.196 & 0.187 + -@xmath65 + forward & 0.262 & 0.030 & -2.300 & 0.324 & 0.052 & 0.201 & 0.194 + inverse & 0.593 & 0.075 & -5.770 & 0.794 & 0.056 & 0.294 & 0.289 + bisector & 0.418 & 0.035 & -3.940 & 0.378 & 0.054 & 0.223 & 0.217 + orthog .
& 0.285 & 0.033 & -2.540 & 0.355 & 0.052 & 0.201 & 0.194 + -@xmath66 + forward & 0.393 & 0.039 & -3.460 & 0.403 & 0.053 & 0.206 & 0.198 + inverse & 0.718 & 0.065 & -6.730 & 0.658 & 0.058 & 0.278 & 0.271 + bisector & 0.545 & 0.037 & -4.980 & 0.387 & 0.055 & 0.223 & 0.216 + orthog . & 0.441 & 0.044 & -3.940 & 0.452 & 0.054 & 0.207 & 0.200 + llllllll [ tab : vr ] - + forward & 0.332 & 0.078 & 2.030 & 0.044 & 0.024 & 0.160 & 0.158 + inverse & 1.520 & 0.318 & 1.420 & 0.174 & 0.093 & 0.359 & 0.348 + bisector & 0.768 & 0.056 & 1.810 & 0.044 & 0.048 & 0.201 & 0.195 + orthog .
& 0.506 & 0.109 & 1.937 & 0.063 & 0.033 & 0.167 & 0.164
+ - + forward & 0.390 & 0.068 & 2.030 & 0.035 & 0.024 & 0.149 & 0.147
+ inverse & 1.220 & 0.196 & 1.670 & 0.096 & 0.063 & 0.274 & 0.267 + bisector & 0.727 & 0.053 & 1.880 & 0.037 & 0.039 & 0.177 & 0.172 + orthog .
& 0.534 & 0.085 & 1.970 & 0.046 & 0.030 & 0.155 & 0.152
+ - + forward & 0.378 & 0.057 & 2.010 & 0.035 & 0.023 & 0.142 & 0.140 + inverse & 1.050 & 0.157 & 1.680 & 0.088 & 0.055 & 0.257 & 0.241 + bisector & 0.662 & 0.049 & 1.870 & 0.038 & 0.036 & 0.167 & 0.163 + orthog .
& 0.481 & 0.067 & 1.960 & 0.043 & 0.028 & 0.146 & 0.143 + as fig .
[ tf ] and table 1 show , the -@xmath66 , -@xmath65 , - and - correlations are tight , although the differences between the lines obtained from the different regression methods are slightly larger than the dispersion associated with any given line .
therefore , the choice of the fitting method will slightly affect the results .
for all the cases , the slopes of the forward correlations are shallower than those of the inverse correlations . for the bisector and orthogonal correlations , their slopes are between those of the forward and inverse correlations .
the orthogonal correlations have shallower slopes than the bisector correlations , which are very close to those obtained from the forward fit . in the following ,
we analyze only the correlations fitted with the orthogonal regression method .
a deep statistical analysis of each one of the correlations presented here is out of the scope of the paper . from the baryonic - to stellar -@xmath67 tfr ,
the slope is shallower by @xmath68 , and the mass at = 160 km / s is 1.36 times ( 0.134 dex ) smaller ; for larger velocities , the mass difference becomes smaller . from - to -@xmath65 ,
the slope remains almost constant . from -@xmath65
to -@xmath66 the slope is steeper by @xmath69 , and some segregation by color appears .
keeping in mind that most of the previous works used to report the forward slopes of the @xmath0-(@xmath6- ) correlations , we remark that the equivalent slopes here would be the inverse of our inverse linear regressions ( see table 1 ) : 3.00 , 3.40 , 3.67 , and 2.77 for the baryonic , stellar , @xmath14 and @xmath8band cases , respectively .
it should be noted that the sample used here is broad in galaxy properties and _ it was not pruned to reduce the tfr scatter_. the few works that carry out analysis of samples similar to ours , although with different fitting methods , give slope values for the luminous tfrs ( kannappan , fabricant & franx 2002 ; verheijen & sancisi 2001 ; dutton et al .
2007 ; courteau et al . 2007 ) and for the stellar tfr ( pizagno et al . 2005 ; gnedin et al .
2007 ) that are close to our estimates .
the recent work by pizagno et al .
( 2007 ) , that analyzes a sample of 162 sdss disk galaxies with h@xmath70 rotation curves , reports slopes of the luminous tfrs that are much steeper than those obtained here .
the baryonic tfr obtained in mcgaugh ( 2005 ) is shallower than ours , however , he uses @xmath49 instead of .
the estimated average intrinsic scatters , , in the - , -@xmath67 , -@xmath65 , and -@xmath66 correlations are 0.051 , 0.045 , 0.049 and 0.063 dex , respectively .
the intrinsic scatter is larger ( by @xmath71 ) for the baryonic correlation than for the stellar one ; the largest scatter is in the @xmath8band .
the intrinsic scatter along the @xmath6 or @xmath0 axis expressed in magnitudes , @xmath72 , oscillates between 0.41 mag ( stellar correlation ) and 0.5 mag ( @xmath3 band correlation ) for all the regression methods used here ( see values for the orthogonal fit in fig .
[ tf ] ) .
recall that our observational error estimates and their propagation are approximations .
nonetheless , our estimates for the intrinsic scatter agree with other similar works .
for example , dutton et al . ( 2007 ) estimate @xmath73 in the @xmath18band , which falls in between our values in the @xmath3 and @xmath4 bands .
pizagno et al .
( 2007 ) , who carried out a deep analysis of the errors for their sample , find typical values of the intrinsic scatter for the optical / nir tfrs of 0.400.45 mag , which is close to our findings .
the -@xmath66 , -@xmath65 , -@xmath67 , and - correlations are scattered and strongly segregated by disk central sb ( fig .
[ rl ] ; table 2 ) . for
a given luminosity or mass , the disk scale length , or central sb , span a large range of values .
we see from fig .
[ rl ] that a segregation by color is also present .
there is no significant difference between the slopes and zero points of the baryonic and stellar @xmath2-@xmath6 correlations .
both the mass and the radius increase when changing from stellar to baryon quantities , but the change is small and shifts galaxies along the same correlation , whose slope is @xmath74 . from the stellar to the @xmath14band correlation
, the slope slightly decreases , while from the @xmath14 to the @xmath8band correlation the slope strongly increases , from @xmath75 to @xmath76 .
the intrinsic scatter measured in the log of the radii is approximately the same in all the cases , @xmath77 .
when translated to the ( @xmath78axis ) scatter in log@xmath0 or log@xmath6 and expressed in magnitudes , the scatter is around 1.1 mag for the @xmath8band correlation and 1.51.7 mag for the other correlations .
hence , in the @xmath3 band , the dispersion along the luminosity is decreased by some compensation effect .
the observational errors in the @xmath2-@xmath0(-@xmath6 ) diagrams are significantly smaller than the corresponding average intrinsic scatters . note that , contrary to the tfrs , for the @xmath2-@xmath0 ( -@xmath6 ) correlations , the use of a broad range in sbs ( from hsb to lsb galaxies ) significantly spreads the correlation and influences the slope and zero point values .
this can be clearly appreciated in fig .
[ rl ] : if we exclude the lsb galaxies , then the fit changes notably . in
most of the previous works aimed to estimate the disk scaling relations , _ lsb galaxies were not considered .
this is unfortunate if the observations are to be used to compare and constrain theoretical models .
_ since the galaxy morphological type correlates with the sb , one expects that the @xmath2-@xmath0 correlation also will depend on the type ; graham & worley ( 2008 ) have indeed shown this explicitely . the - , - , and - correlations are shown in fig .
[ vr ] and their fitting parameters are given in table 3 .
the - correlation is the same as the - one because we assumed that @xmath79 .
the @xmath80 correlations are highly scattered , and segregated by disk surface brightness and color . for these correlations ,
the differences between forward and inverse fits are large .
the relative errors in the fitted slope and zero point parameters are the largest ones among the three scaling relations studied here . because of the large dispersions in the fitted slopes and zero points of the @xmath80 correlations , there is no significant difference among these parameters when we change from the @xmath3 to the @xmath4 band and to the baryonic case . for the orthogonal regression ,
the slopes and zero points are @xmath81 and @xmath82 , respectively .
the intrinsic scatter slightly decreases from the baryonic to the @xmath14band , @xmath83 and 0.15 , respectively .
the estimated inverse intrinsic scatter ( along the @xmath0 axis ) increases from the @xmath4 to the @xmath3 band . in order to complement the presented scaling correlations and their changes
when moving from luminous to stellar and baryonic quantities , we explored the possibility of other correlations for our galaxy sample involving luminosities ( masses ) , global color , central surface densities ( brightnesses ) , and stellar mass content .
[ colorl ] shows the @xmath5 color vs @xmath66 , @xmath65 , , and , with the corresponding pearson correlation coefficients and slopes . in spite of the @xmath0dependent correction by internal extinction that we have applied ,
a clear correlation of color with @xmath0 and @xmath6 remains : more massive ( luminous ) galaxies are on average redder than the less massive ( luminous ) ones .
the @xmath8band correlation is the most scattered , as expected .
the slope becomes slightly steeper when changing from to mainly because the less massive , bluer galaxies have typically larger gas mass fractions , shifting more to the right side in the log@xmath6 axis than the massive galaxies .
the slope also becomes steeper when changing from @xmath65 to @xmath66 , in this case because less luminous galaxies being bluer on average , shift more to the right side in the log@xmath0 axis than the redder luminous galaxies .
noisy log
log correlations are also observed between @xmath5 and the different sbs and surface densities , and between the latter quantities and the luminosities and masses ( fig .
[ colorsb ] ) .
the slopes of the orthogonal fits are given inside each corresponding panel .
the general trend is that the higher the surface density ( or sb ) , the redder and more massive ( luminous ) is the galaxy .
what lies on the basis of these trends for disk galaxies ?
we have also explored correlations related to the galaxy stellar s mass fraction , ( or gas mass fraction @xmath84 ) .
this quantity is an important property of galaxies related to their gas infall and sf rate histories ; it will be discussed in 4.2 ( see fig .
[ stfrac ] therein ) .
the fraction correlates strongly with the @xmath5 color and weakly with the surface densities ( or sbs ) and luminosities ( masses ) , in that order .
for all cases , the lowest sb galaxies break the correlations , suggesting a kind of a threshold for the gas surface density , below which the sf efficiency is almost independent of it ( and of @xmath6 or @xmath0 ) . to gain a more quantitative description of the disk galaxy scaling relations and their implications
, we may explore the behavior of their residuals .
the possible correlations of the residuals among them , and other galaxy properties , bring valuable information on galaxy formation and evolution processes .
in particular , the changes of the residual correlations when moving from optical to nir bands and to stellar and baryonic quantities provide clues on the sf and stellar population evolution processes as well as on the dynamics of disk galaxies . in the different panels of fig .
[ residuos ] , the residuals of the -@xmath6(-@xmath0 ) correlations are plotted against the corresponding residuals of the @xmath2-@xmath6(-@xmath0 ) , @xmath5-@xmath6(-@xmath0 ) , and @xmath85-@xmath6(-@xmath0 ) correlations for the baryonic , stellar , @xmath14 and @xmath8band cases , respectively . the orthogonal regression fit was used to obtain the residuals . as can be seen in fig .
[ residuos ] , for most of the cases , the residuals are weakly correlated or completely uncorrelated with each other . for the baryonic case
, we find an anti correlation of @xmath86 ( ) with @xmath86 ( ) with a slope of @xmath87 ( orthogonal linear regression ) : on average , the more a galaxy deviates in the direction of the high velocity side in the - relation , then the more it deviates in the direction of the small radius side in the - relation .
it is illustrative to know that for the extreme case of exponential disks without dark matter halos , the slope would be @xmath88 , while for completely halo dominated galaxies , no correlation is expected . for the stellar and @xmath14 and @xmath8band cases , the trend described above almost disappears .
these results show that _ the residuals of the -@xmath0 ( -@xmath6 ) and @xmath2-@xmath0 ( -@xmath6 ) correlations can or can not correlate , depending not only on the inner dynamics of the galaxies themselves , but also on their sf histories , a question highlighted in fa - r_. from the left column of fig .
[ residuos ] , we also learn that for all cases , the lsb galaxies are more scattered and uncorrelated than the hsb ones in the residuals plane .
as expected , the residuals of the @xmath2-@xmath0 ( -@xmath6 ) correlations anti correlate strongly with the corresponding disk central surface densities ( or sbs ) as : @xmath89 , @xmath90 , @xmath91 , and @xmath92 .
the surface density ( or sb ) is clearly a significant third parameter in all the @xmath2-@xmath0 ( -@xmath6 ) correlations .
the tightest anti correlation is for the baryonic case ; then its strength decreases successively for the stellar , @xmath14band , and @xmath8band cases .
the residuals of the -@xmath2 correlations correlate significantly with the corresponding surface densities ( or sbs )
. these residuals also correlate with those of the -@xmath0 ( -@xmath6 ) correlations , but with a considerable segregation by sb ( or surface density ) , resembling the @xmath2-@xmath0 ( -@xmath6 ) correlations .
the findings reported above imply that the radius could be a statistically significant third parameter in the baryonic tfr . applying a backward step
wise multiple linear regression procedure , where @xmath93 tests are used to compute the significance of each independent variable ( or @xmath0 ) , @xmath94 , @xmath2 , @xmath5 , @xmath95 , and the morphological type @xmath37 are the independent variables . for each case
( baryonic , stellar , and @xmath3 and @xmath4 bands ) , one starts from a multiple variable linear regression using _ all _ the variables and then , through the backward step
wise procedure , the variables that are not statistically significant are eliminated step by step .
] , zavala ( 2003 ) has indeed found that the radius is a third ( statistically significant ) parameter in the baryonic tfr .
however , the radius was no longer found to be a third parameter in the stellar and luminous tfrs .
we confirm here the results by zavala ( 2003 ) , with the exception that for our analysis in the @xmath3 band , the radius is statistically more significant than in the analysis of zavala ( 2003 ) , however , we still find that the radius is not a third parameter in the @xmath8band tfr ; instead , the @xmath5 color is a third parameter in this case . since we are interested in the changes in the scaling correlations when changing from one band to another and to stellar and baryonic quantities , it is important then to analyze also the possible dependences of their residuals on galaxy color and stellar fraction . in the mid and right panels of fig .
[ residuos ] , the residuals of the -@xmath0 ( -@xmath6 ) correlations are plotted vs the residuals of the @xmath5-@xmath0 ( -@xmath6 ) and -@xmath0 ( -@xmath6 ) correlations , rather than just vs @xmath5 and . in this way , the comparisons are at a given @xmath0 or @xmath6 . in the baryonic case ,
there are noisy trends : the residuals of the - correlation increase on average for increasing residuals of the @xmath5- and - correlations ( the galaxies become redder and with higher ) ; however , this is only true for the hsb galaxies . in the stellar and @xmath14band cases
, one does not see any clear trend . in the @xmath3 band ,
the residuals @xmath96 correlate with the residuals @xmath97 , although mainly due to the hsb galaxies . the backward step
wise analysis shows that the @xmath5 color is indeed the third parameter in the @xmath8band tfr ( see also kannappan et al .
2002 ; pizagno et al .
2007 ; courteau et al . 2007 ) .
there is also a weak correlation of @xmath96 with @xmath98 .
the results reported in 3 reveal non
negligible changes in the slopes and scatters of the scaling correlations of disk galaxies when changing from @xmath3 to @xmath4 bands and to stellar and baryonic quantities . while the most robust aspects of the scaling relations seem to be a consequence of the cosmological initial conditions of disk galaxy formation ( e.g. , fa - r ; shen et al .
2002 ; dutton et al .
2007 ) , the changes found here are expected to be related mainly with the sf and gas infall histories of galaxies . in the following ,
we focus our discussion on the findings presented in 3 and on whether or not these changes are expected in based models of disk galaxy evolution .
the scenario of disk galaxy formation sketched in appendix b makes certain predictions regarding the baryonic scaling relations of disk galaxies ( appendix c ) .
the main simple physical ingredients of disk galaxy formation and evolution are the halo mass , the mass assembly history ( mah ) , the spin parameter @xmath99 , and the disk galaxy mass fraction @xmath100 .
below , we highlight some of the main aspects of these predictions with the goal of interpreting the observational inferences presented in 3 . at some points
, we will discuss in more detail the model results obtained in fa - r . in that work ,
the cosmology used corresponded to a flat universe with @xmath101 , @xmath102 , @xmath103 , @xmath104 ; was assumed to be constant . *
the - relation . * as described in appendix c1 , if the initial condition parameters @xmath99 and do not depend significantly on mass , then one expects that the baryonic tfr should have a slope similar to the cosmological one .
_ this seems to be the case according to our results _ ( @xmath105 from observation vs @xmath106 from simulations ) .
the models of fa - r and dutton et al .
( 2007 ) show that for a reasonable range of values ( between @xmath107 ) , changes of with mass would not have a significant effect on the baryonic tfr slope ( taking into account the disk gravitational drag on the halo ) ; for a given , as decreases , decreases , but also decreases ( actually is the g function defined in the appendix c1 that decreases ) , and the galaxy model shifts along the relation . in the same way , a moderate correlation ( anti correlation ) of @xmath99 with mass would only slightly decrease ( increase ) the slope .
conversely , steep dependences of and/or @xmath99 upon mass would imply a significant change in the slope of the tfr . regarding the intrinsic scatter , we have measured from the observational sample a value of = 0.051 for the baryonic tfr .
this value , in spite of the broadness of the sample , _ is in marginal agreement with the theoretical expectations discussed in appendix c1 .
_ for example , for the fa - r models with = 0.05 , @xmath108 ; by relaxing some model assumptions , this value is expected to be larger .
thus , from the comparison of observations with theory , it seems that there is no room for other significant sources of scatter , either physical or systematical , as for example disk ellipticity and non circular motions ( see 5 for further discussion ) .
further pieces of information related to the scatter in the baryonic tfr are given by the correlations between the residuals of this relation , @xmath109 , and those of the - , @xmath5- and - correlations ( fig .
[ residuos ] ) .
the fact that @xmath109 anti correlates with @xmath110 shows that _ the radius ( or surface density ) plays a role in the scatter of the baryonic tfr_. as mentioned in 3.5 , the radius is found to be indeed a statistically significant parameter in the -- multi variable correlation . because @xmath110 correlates strongly with ( 3.5 ) , the anti
correlation seen in fig .
[ residuos ] implies that , for a given , higher surface density disks produce larger values of . for a given @xmath99 distribution , the larger ,
the higher the disk surface density , and hence the more pronounced is the effect upon because the disk contribution to the total rotation curve is larger .
this implies a steeper and stronger anti correlation between @xmath109 and @xmath110 . in the extreme case of complete dominion of an exponential disks ( no dark matter halo ) ,
this anti
correlation should have a slope of @xmath88 ( courteau & rix 1999 ) .
our observational results point to a slope shallower than @xmath88 , @xmath7 , _ showing that the dark halo contribution at a radius where the rotational velocity is maximal , is significant _ , as predicted by models based on halos . according to fig .
[ residuos ] , the residuals are more tightly anti correlated for the hsb sub - sample than for the lsb sample .
this implies that the dark halo component becomes more dominant for lower values of the disk sb , to the point that the contribution of the disk to the total rotation curve at radii close to its maximum , is negligible for the lowest sb galaxies .
this dependence of the baryon and dark matter contents with sb has been deeply explored in zahf , who showed that for the same galaxy sample used here , only the highest sb disk galaxies are maximum disk , while the lower the sb , the more dark halo dominated ( sub maximum disk ) the galaxies become ( see also e.g. , casertano & van gorkom 1991 ; de blok & bosma 2002 ; catinella et al . 2007 ) .
it is important to mention that the slope of the residual correlation is sensitive to . for the modeling of disks formed inside halos discussed above
, the slope will be significantly steeper than @xmath111 if the average value of is larger than @xmath112 ( e.g. , gnedin et al .
2007 ; dutton et al . 2007 ; pizagno et al . 2007 ) . for the cosmology used by fa - r and = 0.05
, they found that the slope is @xmath113 ( see their fig .
an interesting question is _ what slope would be predicted by alternative theories , like the modified newtonian dynamics , and would agree with the value of @xmath7 found here using a broad sample of disk galaxies .
_ we stress that observational studies aimed to constrain the baryon and dark matter contents in disk galaxies should use the baryonic relations instead of the luminous or stellar ones . for example courteau & rix ( 1999 )
obtained the residuals of the observed ( infrared ) tfr and @xmath2-@xmath0 relation for a sample of mostly hsb galaxies , and interpreted the lack of correlation between these residuals as evidence of dark matter dominance in all galaxies .
as fa - r showed , the dependency among the residuals changes from the baryonic to the stellar relations ( see also dutton et al . 2007 ; courteau et al .
our observational results confirm this fine effect . *
the - relation . * for the galaxy sample analyzed here , the - correlation is highly scattered (= 0.2 for the orthogonal fit ) , with the scatter indeed anti correlating strongly with ( @xmath114 ) , and less with the @xmath5 color , as the based models predict ( appendix c2 ) .
the slope of the correlation , for the orthogonal fit , is @xmath115 ( table 2 ) and is steeper for the bisector and inverse fits , i.e. is close to the slope of the cdm halo - relation ( appendix b ) . _ thus , according to the arguments given in appendix c2 , there is almost no room for a significant dependence of @xmath99 or on , in the range of masses of the galaxies studied here , although if both parameters anti correlate ( or correlate ) at the same time with then the slope changes may be compensated . _ * the - relation . *
the slope of the ( noisy ) - correlation found here is @xmath116 ( orthogonal fit ) , i.e. shallower than the cosmological one ( see appendix b ) .
this , according to appendix c3 _ favors , if any , an anti correlation of with mass_. the intrinsic scatter , = 0.16 , is much larger than the scatter around the baryonic tfr . as predicted by the models , the residuals of the - correlation correlate significantly with both log and the @xmath5 color ( @xmath117 and 0.67 , respectively ) ; the slopes of these correlations ( orthogonal fit ) are 0.34 and 0.28 , respectively .
the main changes when moving from the baryonic to the stellar scaling correlations are the decrease of the slope and intrinsic scatter in the -@xmath6 correlations .
the scatters of the @xmath2-@xmath6 and @xmath2- correlations also decrease but the significance of these decrements are very marginal .
more interestingly , the ( weak ) anti correlation between the residuals of the baryonic - and - correlations tends to disappear in the stellar case .
this is a fact related to other of our results , namely that _ the radius ( or disk surface density ) is a third parameter in the baryonic tfr but no longer in the stellar one_. all these differences are related to the gas ( or stellar ) mass fraction of galaxies , and they show that the mass infall and sf histories vary systematically among galaxies . are these variations consistent with galaxy models based on the framework ?
let us first discuss the change in the slope of the -@xmath6 relation . if the stellar mass fraction , @xmath118 , depends on as @xmath119 , then @xmath120 .
therefore , from @xmath121 one passes to @xmath122 . in the lesft panel of fig .
[ stfrac ] , we show vs for our galaxy sample . although with a high scatter , correlates with roughly as @xmath123 implying then that the slope from the baryonic to stellar tfrs should decrease by a factor @xmath124 .
the corresponding slopes calculated with the orthogonal fits are 0.31 and 0.27 , respectively .
what causes the dependence of ( or ) on ?
it could be that rather than a fundamental dependence , it is a consequence of other correlations .
for example , fig .
[ stfrac ] shows that correlates stronger with ( and also with galaxy color ) than with . from the point of view of the models
, the stellar fraction of normal disk galaxies depends mainly on two factors : the efficiency of gas transformation into stars , given basically by the disk surface density ( determined by @xmath99 , , and ) , and the gas infall history , connected to the halo mah .
based on these processes and by means of semi numerical models , fa - r and avila - reese & firmani ( 2000 ) have shown that ( i ) lower surface density disks transform gas into stars with less efficiency than disks of high surface density , ( ii ) more massive disks tend to be of higher surface density ( see also dalcanton et al .
1997 ) , and ( iii ) the present day correlates significantly with ( or ) , and because of item ( ii ) , with .
note that the dependence of on is a non trivial prediction because several physical / evolutionary processes combine to give rise to the stellar ( or gas ) mass fraction and surface density ( brightness ) of evolved disk galaxies . in fig .
[ stfrac ] , together with the observational inferences , the model predictions presented in fa - r are plotted in the - diagram ( magenta stars ) .
models and observations occupy the same region , though the models have on average slightly lower and values ; the introduction of interaction driven sf would work in the direction of increasing both quantities . as a consequence of the mentioned predictions , fa - r showed that the scatter around the stellar tfr due to @xmath99 ( ) should diminish compared to the one around the baryonic tfr ; from the baryonic to the stellar tfr , the average scatter decreases from @xmath108 to @xmath125 .
galaxies that in the baryonic tfr were shifted to the high side , because of their higher surface densities , transform gas into stars more efficiently and shift also to larger values in the stellar tfr diagram compared to galaxies of lower surface densities . as a result , the dispersion in the stellar tfr due to disk surface density ( mainly associated to @xmath99 and ) decreases , and _ the tfr becomes almost the same for hsb and lsb galaxies : the anti correlation between the residuals of the tfr and those of the - relation disappears . _ the dispersion in the halo mah remains as the main source of scatter in the stellar tfr .
however , regarding the original mah dispersion , it is also predicted that there is a kind of compensation in the shift of galaxies in the -@xmath6 diagrams : galaxies that were shifted to the high side because they were formed in halos assembled earlier ( more concentrated ) , are also shifted to the larger side because these galaxies had more time to consume their gas ( they end up with higher ) .
the predictions of the based disk galaxy models provide an adequate context for interpreting the observational results plotted in the two upper rows of fig .
[ residuos ] .
the loss of dependence of the -@xmath6 correlation residuals upon the @xmath2-@xmath6 correlation residuals when changing from baryonic to stellar quantities can be understood on the ground of the ( self regulated ) sf efficiency dependence on disk surface density described above ( see also dutton et al .
this also explains the reduction of the scatter around the tfr from the baryonic case to the stellar case ( see table 1 ) , though the model results are more pronounced than observations both in the changes of the scatter values and in the slopes of the residual correlations . in the same way , the distribution of galaxies in the diagrams of the residuals for the -@xmath6 , @xmath5-@xmath6 , and -@xmath6 correlations ( fig .
[ residuos ] ) can be interpreted as follows . upon the understanding that disk color and are related mainly to the halo mah ( concentration ) ,
the correlations between the baryonic residuals @xmath109 and @xmath126 , and between @xmath109 and @xmath127 , are expected ( see above ) .
this is because for a given mass , halos assembled earlier have larger , and the galaxies formed inside them transformed gas into stars earlier ( therefore are redder ) and more efficiently ( therefore have higher values of ) . in the corresponding panels of fig . [ residuos ] , one indeed sees these predicted tendencies , mainly for the hsb galaxies . for the lsb galaxies , the tendencies again ( see 3.4 ) almost disappear .
it might be because the ( low ) sf rate in lsb galaxies no longer depend on the gas infall rate ; this dependence is what allowed us to associate the observables , color and , to the halo mah . in the residual plots corresponding to stellar quantities , the tendencies become negligible because from to
, the shifts in in the tfr , due to the halo mah ( concentration ) , tend to be compensated by shifts in ( see the previous paragraph ) .
therefore , _ in the - ( stellar ) diagram the mentioned sf effects tend to reduce the scatter and its dependence on surface density , color , and compared to the - ( baryonic ) diagram . _ the slopes and scatters of the @xmath14band scaling correlations are similar , within the uncertainties , to those of the stellar scaling correlations .
the only quantity that actually changes in the diagrams is @xmath65 for , through the @xmath14band mass to luminosity ratio , @xmath27 .
this ratio increases slightly with the disk color for hsb galaxies and slightly decreases with it for blue lsb galaxies ( 2.2 ) .
hence , a significant trend of @xmath27 with mass or is not expected . for the scale length
, we assumed that = .
interesting differences are seen between the @xmath14 and @xmath8band scaling correlations . in the -@xmath0 diagrams , the slope and intrinsic scatter increase significantly from the @xmath4 to the @xmath3 band . calculating the ratio of the @xmath14 and @xmath8band tfrs
, we obtain : @xmath128 , where @xmath129 is @xmath130 times the difference of the @xmath8 and @xmath14band slopes .
this correlation is a consequence of the color@xmath0 correlation shown already in fig .
[ colorsb ] . that the less luminous galaxies systematically tend to be slightly bluer , implies already a steepening of the slope when changing from the @xmath14 to the @xmath8band tfr .
furthermore , the residuals of the @xmath8band tfr are correlated with those of the @xmath5-@xmath66 correlation , and less with the residuals of the -@xmath66 and -@xmath66 correlations : redder galaxies , with higher sb and tend to be shifted to the tfr large side ; the opposite happens for bluer galaxies , with lower sb and ( lowest row panels in fig .
[ residuos ] ) .
in fact , the step wise backward analysis mentioned in 3.5 shows that the @xmath5 color is a statistically significant parameter in the -@xmath66-@xmath5 multi variable correlation , i.e. _ color is the third parameter in the @xmath8 band tfr , _ a result that other authors have also found in the optical bands ( e.g. , verheijen & sancisi 2001 ; courteau et al .
2007 ; kannappan et al .
2002 , and more references therein ) .
it is obvious that a galaxy in the -@xmath65 diagram shifts to the large ( low ) @xmath8band luminosity side if it is bluer ( redder ) than the corresponding average . why do galaxies shifted to the high ( low ) side in the @xmath3 band tfr
tend to be redder , with higher sbs and larger values of ( bluer , with lower sbs and smaller values of ) ? according to
the galaxy models discussed above , disks formed in halos with low @xmath99 and/or early mass assembling ( high concentration ) , tend to consume gas into stars early in such a way that , for the same or @xmath65 , they result with older stellar populations ( lower @xmath66 and redder ) and a higher than those formed in halos with high @xmath99 and/or late mass assembling ( low concentration ) .
thus , _ while the dispersions introduced by @xmath99 , , and the mah are reduced by compensating effects in the stellar and @xmath14band tfrs ( 4.2 ) , the opposite happens in the @xmath8band tfr_. we see indeed that the tfr in blue bands is systematically more scattered than in the nir bands ( table 1 ) . regarding the -@xmath66 correlation , its slope is significantly steeper than that of the -@xmath65 correlation .
again this is because , as the galaxy is more luminous , it tends to be redder , i.e. , the @xmath66to@xmath65 ratio decreases with @xmath0 .
however , the situation is more complex than in the tfr case , because in the @xmath2-@xmath0 diagrams ( i ) the scatter is large and galaxies are strongly segregated by sb and color , and ( ii ) the @xmath8 and @xmath14band radii are not the same as is the case of in the tfr diagrams ; the to ratio tends to be larger for redder , more luminous galaxies .
thus , on one hand , the redder , higher sb galaxies shift to the low@xmath0 and high@xmath2 sides in the @xmath2-@xmath0 diagram when changing from band @xmath4 to band @xmath3 ; the shift tends to be larger for the more luminous galaxies ( see fig .
[ rl ] ) . on the other hand ,
the bluer , lower sb galaxies shift slightly , mainly to the low@xmath0 side , while the shift is larger for those of larger luminosities . as a consequence ,
the slope of the correlation increases in the @xmath3 band and the scatter along the luminosity axis decreases . in the -@xmath2 diagrams ,
the correlations do not significantly change when changing from the @xmath4 to the @xmath3 bands .
this confirms that the most relevant shifts in the @xmath2-@xmath0 diagrams are due to @xmath0 rather than @xmath2 .
the models predict that the scatter around the -@xmath2 relation should thicken slightly in the band @xmath3 with respect to the band @xmath4 : galaxies of lower sbs and/or bluer ( larger @xmath99 , late assembling halos , and lower ) , which are shifted to the low- side in the @xmath4 band , tend to shift also to larger @xmath8band radii with respect to their radii ; for galaxies that are of higher sb and redder , the radii tend to be the same in both bands .
according to table 3 , the direct intrinsic scatter around the - and - correlations is almost the same , while the inverse scatter ( along the @xmath2 axis ) , is larger in the @xmath3 band .
as shown , the predictions obtained with the based models by fa - r offer an excellent qualitative description of several observational details of isolated disk galaxies . a systematic comparison of observations with results from models calculated with the most recent cosmological parameters will be presented elsewhere . with the aim to show that the models are able to provide even quantitative predictions , we plot those results from fa - r that can be compared with the observational inferences presented here . in fa - r , the authors plotted averages and scatters for three mass bins in the stellar tfr diagram ( their fig .
7 ) , that we reproduce here in the corresponding panel of fig .
[ tf ] ( magenta stars with vertical error bars ) .
we diminished their points by the factor ( 0.70/0.65)@xmath131 in order to take into account the slightly different value of @xmath132 used in their models .
the agreement is good , even more if we consider that models calculated with a @xmath133 , would have smaller values of .
the agreement is also quite good in the baryonic tfr diagram ( magenta stars in the lower right panel of fig . 1 ;
the values given in fa - r were used to calculate from ) .
it should be stressed that models and observations are now being fairly compared ; in both cases we have nearly isolated normal disk galaxies in a wide range of sbs and morphological types . in fa - r ,
model results for the - relation were also presented , separated in hsb and lsb galaxies , but models with very high sb were not taken into account .
we estimate the `` weight '' of these models in the fit for the hsb galaxies , and reproduce both the lsb and hsb+very hsb fits in the corresponding panel of fig .
[ vr ] ( thick magenta dashed lines ) , correcting the radii by a factor of ( 0.70/0.65)@xmath24 . the agreement is rather good within the large scatter . finally , model results by fa - r in the - diagram ( fig .
[ stfrac ] , magenta stars ) , and ( ii ) in the diagram of residuals of the baryonic - and - correlations ( upper left panel of fig .
[ residuos ] , magenta dashed line ) were already presented above , evidencing a reasonable agreement with observations ; for the stellar case , the residuals do not correlate , as seen also for the observations .
the hierarchical based models of disk galaxy evolution discussed above have been useful for interpreting the results reported in 3 about the slopes and scatters of the observed scaling relations in the different cases , as well as the behaviors of the residuals from these and other correlations .
even the zero points of the relations predicted by fa - r seem to agree quantitatively with observations .
note that the models by fa - r , differentiate from others in the literature in that they _ include a detailed description of disk sf and feedback and follow self
consistently the whole process of halo
disk evolution_. although a detailed comparison with observations , using models calculated with more recent cosmological parameters , was left for a forthcoming paper , we may anticipate here some potential shortcomings of the model predictions as well as solutions .
\(1 ) _ the models are not able to explain the observed correlation of galaxy color with mass or luminosity_. in the hierarchical scenario , isolated more massive halos tend to assemble a given fraction of their present
day mass later than the less massive ones . then disks formed inside the more massive halos would tend to be younger ( bluer ) than those formed in the less massive halos .
however , the models also show that the more massive the halo is , the higher the surface density of the disk formed inside it ( dalcanton et al .
1997 ; fa - r ) ; the higher the disk surface density , the more efficient the sf process and therefore , the redder the galaxy . as a result of these two evolutionary compensating effects ,
the models predict almost no dependence of color on @xmath6 or @xmath0 ( e.g. , avila - reese & firmani 2000 ; avila - reese et al .
2005 , see table 5 therein ) .
a possible way to produce a steeper color@xmath6 correlation in the models is by obtaining a steeper disk surface density@xmath6 dependence ; in fact , the model prediction for this dependence is also too shallow as compared with observations ( see below ) .
it should be said that the models do not include satellite galaxies and interacting induced sf . for massive systems
it is probable that the late gas infall is reduced by the large cooling time and because a fraction of the gas could have been trapped before into substructures that tend to be incorporated into the central galaxy through ( minor ) mergers ; all these processes make the most luminous galaxies redder and with larger values than obtained in the fa - r models .
an indirect dependence of galaxy color on @xmath0 may also appear through environment .
however , the disk galaxy sample studied in this paper includes field galaxies and galaxies from the uma low
density , low mass cluster of galaxies .
therefore , it is less probable that environmental effects should produce the color@xmath0 correlation .
this correlation , which in some sense resembles the so called _ anti hierarchical or downsizing galaxy formation _ reported for early
type luminous galaxies ( e.g. , cimatti , daddi & renzini 2006 ) , could be produced by a mass ( luminosity)dependent dust extinction larger than that used here to correct the observations .
another alternative is related to the agn feedback mechanism invoked recently to solve the sharp cut off at the high end of the ( elliptical ) galaxy luminosity function , as well as the observed downsizing of red early type galaxies ( e.g. , bower et al .
such a mechanism should be , however , strongly dependent on the galaxy mass in order to produce the observed continuous and monotonic change of color with mass ( luminosity ) of disk galaxies .
\(2 ) _ the predicted dependence of surface density on mass ( or sb on luminosity ) is too shallow as compared to that inferred from observations _ , although the latter is very noisy .
such a dependence can be estimated by combining the two simple model relations : @xmath134 by rejecting the small dependence of @xmath135 on mass ( see appendix c2 ) , we then obtain that @xmath136 the semi
numerical evolutionary models show that the dependence of on is even shallower than 1/3 . on the other hand ,
the slope ( orthogonal regression ) that we have found for the ( noisy ) correlation between and in our galaxy sample is @xmath137 , and it does not change significantly for the stellar and luminosity cases . assuming that there is no a dependence of and @xmath99 on mass , model predictions and observations would not agree : the surface density ( brightness ) of observed galaxies increases on average more rapidly with @xmath6 ( @xmath0 ) than the models predict .
again , the inclusion of mergers and interaction induced sf would work in the direction of increasing in the more massive galaxies . on the other hand , recent inferences of the halo @xmath99 distribution for a large sdss sub
sample of galaxies show this parameter to be systematically smaller and less scattered for more luminous galaxies ( cervantes - sodi et al . 2007 ; berta et al . 2008 ; see also hernandez et al .
2007 ) . by taking into account an anti correlation of @xmath99 with mass in the models , both the color ( 1 ) and surface density ( 2 ) problems could be ameliorated without affecting significantly the scaling correlations .
the question is why more massive galaxies would end with smaller values of @xmath99 and consequently , higher surface densities than the less massive ones .
cervantes - sodi et al .
( 2007 ) suggested that the trend of @xmath99 with mass can be related to the halo angular momentum acquisition .
a complementary explanation may lie in the baryon processes .
for example , the outer gas in the most massive galaxy halos may not have enough time to cool and fall to the disk ( for numerical results see e.g. , van den bosch et al .
2002 ; keres et al . 2005 ) ; the material located in the outer regions is the richest in the specific angular momentum ( bullock et al . 2001b ) .
on the other hand , in less massive halos the outer gas may flow through filaments to the disk , even before being shock heated by the halo collapse ( keres et al . 2005 ) . as a result
, the specific angular moment content will be higher in low mass galaxies and lower in the massive ones , which implies lower and higher disk surface densities , respectively .
note that this scenario also implies that more massive galaxies would tend to have lower values of .
a mild anti
correlation of with mass in the models is allowed according to the observational results presented here ( 4.1 ) : the - relation would agree with the anti correlation , while the - relation would do so in case @xmath99 also anti correlates with mass , and regarding the - relation , there is room for a mild variation of with mass .
detailed model calculations are necessary for obtaining more quantitative conclusions .
\(3 ) _ are the scatters around the model and observed tfrs in agreement ?
_ for the baryonic case , the agreement is marginal , given the scatter in the models is slightly larger than that in the observations ( 4.1 )
. the difference could be even larger if ( i ) some of the simplifying assumptions introduced in the models are relaxed , which would increase the predicted scatter , and/or ( ii ) systematical errors such as disk ellipticity are present , which would give rise to a reduction in our estimate of the observational intrinsic ( physical ) scatter .
therefore , the current models of disk galaxy formation based on the cosmology have a potential problem in predicting the intrinsic scatter around the baryonic tfr .
this scatter is smaller in the models for lower values of or for a distribution of @xmath99 narrower than that of the halos .
it could also be possible that the potential systematic errors , such as disk ellipticity , confabulate with the disk surface density or baryonic to total mass ratio ( with @xmath99 and from the model s point of view ) in such a way that the observed tfr scatter is systematically diminished .
for the stellar case , the scatter in the models becomes slightly smaller than that in the observations ( 4.2 ) .
although in both cases ( models and observations ) , the scatter decreases from the baryonic tfr to the stellar tfr , for models , the decrease is more pronounced than it is for observations .
however , there are known effects not considered in the fa - r models , like the interaction induced sf , that will work certainly in the direction of increasing the scatter around the stellar tfr ( e.g. , barton et al .
2007 ) , but not around the baryonic one .
\(4 ) _ are current models able to match both the tfr zero point and the galaxy luminosity function ?
_ first , we note that the disk galaxy evolutionary approach , like the one of fa - r , is different by construction to the the semi analytic models ( sam ) or the halo occupation models ( hom ) : the former is focused on the modeling of the galaxy internal physics and evolution , while the latter are constructed for reproducing galaxy population global properties as the luminosity function .
the parameters that the sam and hom need for reproducing in detail the observed luminosity function apparently produce a luminous tfr whose zero point is shifted to the higher velocity side with respect to observations ( baugh 2006 , and references therein ) .
some astrophysical solutions were proposed to alleviate this apparent problem ( e.g. , dutton et al .
2007 ; gnedin et al . 2007 ) .
however , before introducing modifications to the common disk galaxy formation scenario , we may mention at least four aspects that should be taken into account in order to compare fairly the predictions of sam / hom with observations in regards to the tfrs .
( 1 ) since the tfrs are related to internal disk properties , a more detailed modeling of these properties than that carried out in the sam is necessary ; for example , in the sam , disks are assumed to have a purerely exponential mass distribution , the sf
feedback processes are artificially linked to the halo circular velocity , etc .
( 2 ) is typically used to define the tfrs , while in the sam / hom only approximations to this velocity are calculated . ( 3 ) the tfrs refer only to the population of normal disk galaxies , while the luminosity function that the sam / hom fit to observations refers to the overall galaxy population .
( 4 ) the halos where disk galaxies form could be a sub group of the overall halo population , for example those that did not suffer a major merger since @xmath138 ; therefore , the corresponding `` disk galaxy '' halo mass function , and not the overall one , should be used in the hom as the starting point in the procedure of shaping the luminosity function .
from the point of view of the sam and hom , the passage from the halo mass function to the galaxy luminosity function implies a specific to@xmath0 ratio as a function of mass .
recent studies using galaxy
galaxy weak lensing , allowed one to directly determine the central galaxy to@xmath0 ratios ( e.g. , hoekstra et al . 2005 ; mandelbaum et al .
for the fa - r models , which would agree very well with the baryonic and stellar tfrs if corrected for @xmath133 , we have the @xmath8band luminosities available . the results from mandelbaum et al .
( 2006 ) for _ late type _ galaxies ( note that they differ from those of the early type galaxies ) are given in the @xmath139 band . by using @xmath140 for late type galaxies ( de jong 1996b ) and @xmath141
, we estimate from the fa - r models @xmath142/(/@xmath143 ) = @xmath144 and @xmath145 , for the luminosities @xmath146 and @xmath147 , respectively .
to compare the fa - r @xmath148 ratios with those of mandelbaum et al .
, the values just reported should be increased by @xmath149 due to the different definitions of .
the good agreement between the models and observations within their scatters is encouraging .
the model predictions also agree roughly with the @xmath142 and @xmath150 ratios calculated from the observed luminosity function and the stellar mass to@xmath0 ratios inferred from kinematical mass modelling ( shankar et al .
we can also compare tracers of the inner luminous ( or baryonic)to dark matter ratios of models and observed galaxies .
for example , in zahf the disk to total velocity ratio at the maximum of the rotation curve was estimated for the same galaxies studied here . the models and observations
were found to agree roughly , both in the zero point as well as in the dependence of this ratio on the disk surface density or sb ( see fig . 3 in zahf ) .
summarizing , the fa - r models based on the scenario seem to be able to predict both the tfr , including its zero point , and the inferred , directly from observations , to luminosity ratios of central late type galaxies , as well as the inner luminous and baryonic to dark matter contents .
these ratios , specially the former , are closely related to the `` transfer '' function needed for passing from the halo mass function to the galaxy luminosity function .
we speculate that if the sam and hom take into account the shortcomings mentioned above , then the potential problem of fitting both the tfr zero point and the luminosity function will be overcome . on the other hand ,
some of the initial condition parameters used in models such as those of fa - r , if any , could be modified accordingly to reproduce the disk
galaxy luminosity function , while the tfrs ( and the other scaling relations ) would be yet in agreement with the observations .
for example , the anti correlation of with mass for luminous galaxies ( motivated by the large cooling time in massive halos ) that the sam / hom evoke , could be allowed by the models as discussed in items 1 and 2 above .
a compiled sample of 76 normal ( non interacting ) local disk galaxies of all morphological types and sbs has been used to construct and compare , among them , the luminous ( bands @xmath3 and @xmath4 ) , stellar and baryonic scaling relations . the required observational information for building up this sample implied detailed surface photometry in bands @xmath3 and @xmath4 , a dynamical determination of the rotation curve amplitude , and hi gas integrated flux data .
we have corrected and processed the observational data homogeneously , and obtained several composite disk quantities and their uncertainties : global @xmath5 colors , stellar masses and surface densities , gas ( or stellar ) fractions , and baryonic masses , surface densities and scale lengths .
the objectives of this work were ( i ) to explore the changes of the disk
galaxy scaling relations and their residuals as moving from optical to nir bands , and to stellar and baryonic quantities , ( ii ) to look for interpretations of the results in the context of based evolutionary disk galaxy models , and ( iii ) to show the necessity of larger samples , similar to the one compiled here , for constraining models of galaxy formation and evolution .
a summary of the results is as follows .
the slope of the baryonic tfr is shallower than the slopes of the stellar and @xmath14band tfrs ( @xmath151 vs @xmath152 and 0.26 ; orthogonal regression ) and , in agreement with previous works , the slope of the @xmath8band tfr is steeper than that of the @xmath14band tfr ( @xmath74 vs @xmath152 ) .
the estimated average intrinsic scatters in log around the baryonic , stellar , @xmath14 and @xmath8band tfrs are 0.051 , 0.045 , 0.049 , and 0.063 dex , respectively .
thus , the baryonic tfr is more scattered than the stellar tfr .
a statistical analysis shows that the radius ( or surface density ) is a significant third parameter in the former relation , while the latter , as well as the @xmath14band tfr , do not admit a third parameter . in the case of the @xmath8band tfr , which is the most scattered
, the @xmath5 color is a significant third parameter .
the @xmath2-@xmath153 correlations are scattered and segregated strongly by disk sb and weakly by color .
the slopes of the baryon , stellar , and @xmath14band correlations are around 0.30 ( orthogonal regression ) , while for the @xmath8band relation , this increases to 0.44 . the scatter around the latter , projected in the luminosity axis , is smaller than that in the other correlations .
the @xmath154 correlations are the most scattered and are also segregated strongly by disk sb and less by color .
the slopes have values of around 0.5 .
the residuals of the baryonic - and - correlations are anti - correlated , mainly for the hsb galaxies .
the slope is @xmath87 ( orthogonal regression ) , showing that the smaller the disk radius , for a given mass ( higher surface density ) , the larger the disk contribution to .
for the stellar and @xmath14band cases , the anti correlation almost disappears .
thus , our results show that the correlations among the mentioned residuals are different for the baryonic , stellar , @xmath14 and @xmath8band cases .
while the correlation ( or the lack of it ) among the residuals in the baryonic case can be used to explore the dynamical importance of the halo / disk in galaxies , in the stellar and luminous cases the effects related to sf processes distort any pure dynamical interpretation of the results . in spite of the velocity dependent internal extinction correction that we have applied , the @xmath5 color correlates significantly with the logarithms of @xmath66 , @xmath65 , , and ( orthogonal regression slopes of 0.61 , 0.53 , 0.56 , and 0.60 , respectively ) .
noisy but steep correlations are observed between the surface densities and brightnesses and the corresponding masses and luminosities . the stellar fraction , , correlates strongly with @xmath5 and less with surface densities ( brightnesses ) and luminosities ( masses ) . in all cases ,
the lowest sb galaxies break the latter correlations , suggesting a kind of threshold in the surface density , below which the sf rate is almost independent of it ( and of @xmath6 or @xmath0 ) and of the gas infall rate .
we have discussed previous models of disk galaxy formation and evolution ( including halo contraction , sf , and disk feedback ; e.g. , fa - r ) within hierarchically growing halos , under the assumption of gas detailed angular momentum conservation .
we showed the potentiality of these models to describe the observational inferences , as well as how such inferences can help us to constrain several of the model parameters .
a crucial aspect of this endeavor is the correct comparison of models with observations .
the basic models refer to isolated normal disk galaxies .
therefore , the observational sample should also be for nearly isolated normal disk galaxies .
we highlight the following conclusions : @xmath155 the slopes of the baryonic - , - , and - correlations can be interpreted as a direct imprint of the cosmological halo - , - , and - relations , especially the former ( baryonic tfr ) , which is the tightest one .
the models show that this relation is robust to systematical or statistical variations in the disk baryonic mass fraction , , for @xmath156 .
then , the main sources of scatter are the dispersions in the halo spin parameter @xmath99 and halo mah ( or concentration ) .
the fa - r models provide a lower limit to the scatter around the relation due to these two dispersions , namely @xmath157 ( = 0.05=const . ) , which is already slightly larger than our observational inference . @xmath155 in combination , the slopes and scatters of the baryonic - , - , and - correlations inferred here from observations , as well as the correlations among their residuals , imply that the average value of in normal disk galaxies could not be larger than @xmath158 and that can not depend strongly on @xmath6 ( or @xmath0 ) ; if any , a moderate anti
correlation with @xmath6 is permitted .
a trend of nearly independent on mass , with average values of @xmath159 , has been reported recently by other authors , who extended the analysis to disk dwarf galaxies and inferred halo masses directly from the observational data ( blanton et al .
2007 ; see also baldry et al . 2008
; begum et al .
2008 ) .
@xmath155 the models are able to explain qualitatively the observed ( weak ) anti
correlation among the residuals of the _ baryonic _ - and - correlations and the related finding that the radius ( or surface density ) is a third parameter in the former correlation .
the physical parameter beyond these behaviors is @xmath99 .
model disks with low @xmath99 have high surface density and contribute significantly to , in such a way , that for a given mass , correlates with or . instead ,
disks with high @xmath99 values are of low surface density and have a negligible gravitational contribution to ; then , the dependence of on ( or on for a fixed mass ) tends to disappear .
the models can also explain why the anti correlation among the residuals becomes negligible in the stellar or @xmath14band cases : for a given mass , galaxies not only shift to the high- side in the tfr diagram as the radius is smaller ( is higher ) , but also shift to the high- ( or @xmath65 ) side due to a higher efficiency in transforming gas into stars . as a result ,
the scatter around the stellar tfr becomes independent of or , and smaller than in the baryonic case .
@xmath155 the ( self regulated ) sf efficiency effect just mentioned produces a dependence of the stellar mass fraction on surface density ( or brightness ) .
such a non trivial prediction roughly agrees with the one found for our processed galaxy catalog ( fig . [ stfrac ] ) , and it is associated mainly with the @xmath99 parameter : the smaller the @xmath99 , the higher the disk surface density and the more efficient the transformation of gas into stars . although less relevant , and are also affected by the gas infall history , which is connected to the cosmological halo mah . since the mah plays a role in the scatter of the baryonic tfr , trends among the residuals of this relation and those of the color- and - relations are predicted .
these trends were found for our observational catalog , mainly for the hsb galaxies ( fig .
[ residuos ] ) . for lsb galaxies
, it could be that the sf rate timescale becomes larger than the that of the gas infall rate , the connection between mah and color/ then disappears .
@xmath155 _ potential difficulties.- _ our observational results show that both ( i ) the color@xmath6 and ( ii ) surface density@xmath6 correlations are steeper than the models would predict .
we speculate that these shortcomings suggest that late minor mergers and interaction induced sf , which are more common in massive galaxies , should play some role and/or that both , the galaxy spin parameter and anti correlate with mass ; at least qualitatively , both anti correlations , in combination are not expected to affect the scaling correlations presented here .
( iii ) the fact that the intrinsic scatter around the predicted baryonic tfr appears to be larger than that inferred here , implies that there is no room for other physical or systematical sources of scatter around the tfr such as disk ellipticity . on the other hand , the intrinsic scatter around the stellar tfr becomes smaller in the models than in the observations .
however , external effects , such as the interaction driven sf , work in the direction of adding some scatter to the stellar tfr .
( iv ) the fa - r models ( rescaled to @xmath133 ) agree with the observed baryonic and stellar tfr zero points as well as with the to@xmath0 ratios inferred directly from observations .
the potential difficulty that the sam / hom find in matching both the tfr zero point and the luminosity function could be solved by taking into account the shortcomings we have discussed in 5 .
the results obtained in this work pose relevant questions for future studies .
we showed that the determination of the scaling relations for disk galaxies in different bands and for stellar and baryonic quantities , as well as for the correlations among their residuals , bring valuable information on the nature and evolution of galaxies .
the completion of large galaxy samples , unbiased and broad in properties , with detailed photometric , dynamic , and gas content information is crucial for this kind of study .
the preferred approaches for such an undertaking are : follow up observations in @xmath160 and/or hi emission lines for galaxies from the largest optical / nir band surveys as sdss ( pizagno et al . 2005 ) , and synergy of existing wide area homogeneous surveys conducted at optical / nir bands and hi emission line , for example sdss with hi surveys such as alfalfa ( giovanelli et al .
according to the philosophy followed in this paper , substantial advance comes from the adequate comparison of observations with theoretical models .
thus , in order to confirm , rule out , or perhaps expand , some of the conclusions attained here , a deep model exploration of the formation and evolution of the disk galaxy population should be carried out .
this work was supported by papiit
unam grant in107706 to v.a . and conacyt grant 42810 to h.m.h .
j. z. acknowledges support from dgep - unam and conacyt scholarships , and support by the cas research fellowship for international young researchers .
j. z. is supported by the joint program in astrophysical cosmology of the max planck institute for astrophysics and the shanghai astronomical observatory .
we are grateful to s. courteau for comments to an early version of the manuscript and to the anonymous referee for his / her critizism and suggestions . in the following
we describe how we estimate the uncertainties for the quantities we used to construct the scaling relations presented along the paper .
we assume them to be gaussian distributed , so our estimates refer to the standard deviation .
_ @xmath3 and @xmath4 magnitudes : _ we neglect the contributions to the errors due to uncertainties on the redshift , the extinction in our galaxy , the disk thickness , and the @xmath161 correction . hence , the square of the error in the absolute magnitudes is estimated as the quadratic sum of the measurement error , @xmath162 ( given in the original papers used to compile the sample ) , and the one due to the correction for internal extinction , @xmath163 : @xmath164 to estimate @xmath163 we take into account only errors related to the inclination of the galaxies , @xmath165 , and neglect errors related to the empirical parameter @xmath166 ( recall that @xmath167 , tully et al .
giovanelli et al .
( 1997 ) suggest to add also the error @xmath168 , but this term is in most of the cases sub - dominant compared to the error in the inclination ( see , for instance , fig . 2 of their paper ) ; only for high inclination and high values of @xmath166 both terms may become comparable . thus , we have @xmath169 where we assume @xmath170 , following the estimation for the median error of @xmath165 given by giovanelli et al . (
finally , the error in log@xmath0 is given by @xmath171 , where @xmath172 was given in equation ( [ mag ] ) .
_ stellar mass : _ the error in log is estimated as the quadratic sum of the logarithmic errors in @xmath65 and in the mass to luminosity ratio @xmath27 : @xmath173 where an uncertainty of 25% in the value @xmath27 was considered ( bell et al .
2003a ) . _
baryonic mass : _ the error is calculated as the quadratic sum of the errors in and , @xmath174 ; the corresponding logarithmic error in is then @xmath175 the error in is calculated as : @xmath176 where the measurement error @xmath177 is taken as reported in the hyperleda information system .
scale lengths : _ in order to estimate the fitting error in the scale lengths of the galaxies used here , we have experimented with the fit of the observed sb profiles by using both a marking the disk method and the bulge to disk decomposition method .
we have seen that typical uncertainties in the determination of @xmath2 by both methods are of the order of @xmath178 ( see also macarthur et al . 2003 ; graham 2002 ) .
there is a very weak systematical trend for larger lengths determined by the latter method relative to those determined by the former one from late to early morphological types .
the uncertainty in @xmath2 also depends on the sb profile type and other specific details .
finally we decided to assign a logarithmic error of @xmath179 ( @xmath68 ) to the scale lengths in both the @xmath3 and @xmath4 bands , and @xmath180 to the baryonic case , where the gas disk also contributes to the final surface density profile .
distinct cdm halos those not contained within larger halos can be characterized by the so called virial mass and radius , and , and by a maximum circular velocity , . halos which are contained inside larger ones ( sub halos ) commonly have radii and masses smaller than and , because their growth is truncated or even reversed due to tidal stripping ;
is less affected by these effects .
it is well known that the masses and maximum circular velocities of cdm halos , and even of sub - halos , correlate tightly as @xmath182 with @xmath183 , where is the virial or truncated mass of the halo ( e.g. , navarro et al .
1997 ; avila - reese et al .
1998 , 1999 , 2005 ; bullock et al .
the correlation is basically an imprint of the linear power spectrum of fluctuations , whose variance at the galaxy scales decreases very slowly ( logarithmically ) with the fluctuation mass .
this rough scale invariance naturally produces the @xmath184-@xmath185 relation for the collapsed cdm halos .
the complex non linear non spherical hierarchical halo assembly process introduces only minor deviations upon this relation as well as a scatter .
the cosmological tfr has an intrinsic scatter , , mainly due to the stochastic nature of the mass assembly histories , which also produces a scatter in the halo concentration parameter @xmath186 ( avila - reese et al .
1998,1999 ; fa - r ) . for distinct halos , by definition @xmath187 ;
sub - halos tend to follow a similar relation between and , but with some scatter and with a smaller proportionality coefficient . consequently , and are also correlated , as @xmath188 with @xmath189 , although extra scatter is introduced in this relation due to the variation in for a given mass , as in the case of the @xmath184-@xmath185 relation .
disk galaxies are assumed to form inside the growing cdm halos from the trapped baryons .
therefore , the scaling relations of the halos are expected to be imprinted in the baryonic and stellar scaling relations of disk galaxies .
we describe below the main physical ingredients of the fa - r models ( see for details avila - reese et al .
1998 ; fa - r ; avila - reese & firmani 2000 ) . an extended press
schechter approach is used to generate the mahs of the halos from the primordial density fluctuation field , and a generalized secondary infall model is applied to calculate the time by time virialization of the accreting mass shells .
the evolution and structure of the @xmath10cdm halos calculated in this way agree well with results from cosmological n body simulations ( avila - reese et al .
the halo mass shells are assumed to have aligned rotation axis with specific angular momentum given by @xmath190 , where @xmath191 , @xmath192 , m@xmath193 and @xmath194 are the total angular momentum , mass , and energy of the halo at the shell virialization time @xmath195 .
the halo spin parameter , @xmath196 , is assumed to be constant in time . as a result of the assembling of these mass shells
, a present day halo ends with an angular momentum distribution close to the ( universal ) distribution measured by bullock et al .
( 2001b ) in n body simulations . a fraction of the mass of each shell
is assumed to cool down and form a disk layer in a dynamical time .
the radial mass distribution of the layer is calculated by equating its specific angular momentum to that of its final circular orbit in centrifugal equilibrium ( detailed angular momentum conservation is assumed ) .
the superposition of these layers form the disk , which tends to be steeper in the center and flatter at the periphery than the exponential law .
the disk surface density distribution is mainly determined by the halo angular momentum distribution .
for example , for a given halo of radius and mass , the characteristic size of the baryonic disk ( and its typical surface density ) , described in a first approximation by the scale length , , is determined mainly by the halo spin parameter , @xmath196 .
the distribution of @xmath196 found in analytical and numerical studies is well described by a log normal function , whose median and dispersion almost do not depend on the halo mass .
while we assume here that the pre and post disk formation spin parameters are equal , @xmath197 , in most of our discussions , the quantity @xmath99 refers to the post disk formation spin parameter and it could deviate from @xmath196 if the baryon angular momentum is redistributed ( in the halo or inside the disk ) or if not all the halo gas falls into the disk . the gravitational interaction of disk and inner halo ( important for estimating )
is calculated using an extended adiabatic invariance formalism , which differs from the usual one in that we take into account the ellipticity of the orbits , i.e. , the circular orbit assumption is relaxed .
the disk sf at a given radius ( azimuthal symmetry is assumed ) is triggered by the toomre gas gravitational instability criterion and self regulated by a vertical disk balance between the energy input due to sne and the turbulent energy dissipation in the ism .
this physical prescription naturally yields a schmidt law with an index @xmath198 , varying slightly along the disk .
the sf efficiency depends on the gas surface density determined mainly by @xmath99 , and on the gas accretion rate determined by the cosmological mah .
finally , we estimate the mass of the ( pseudo)bulge as the inner disk mass where the toomre stellar parameter indicates disk instability . in order to pass from the cosmological tfr to the baryonic one , we should pass ( i ) from the halo to , and ( ii ) from the halo mass to .
the latter is given simply by = , where is the ( central ) galaxy mass fraction .
the distribution of and its possible dependence on and other halo properties is not well known .
however , indirect and direct pieces of evidence show that should be much smaller than the universal baryon fraction , @xmath199 ( e.g. , mo et al . 1998 ; fa - r ; smith et al .
2001 ; bell et al . 2003b ; jimnez , verde & oh 2003 ; pizagno et al .
2005 ; hoekstra et al .
2005 ; mandelbaum et al .
2006 ; blanton , geha & west 2007 ) . regarding item ( i ) , the formation of the disk inside the dark halo and the gravitational drag produced on it , redistributes the inner mass profile .
the / ratio increases with the disk surface density and the disk baryon fraction . for a given
, the disk has a larger surface density for a smaller value of @xmath99 .
therefore , the relation between and depends on both @xmath99 and : /=
g(@xmath99 , ) .
the function g(@xmath99 , ) has been approximated in zavala ( 2003 ) from models of disk formation inside cdm halos based on mo et al .
( 1998 ) and fa - r ( note that fa - r used an adiabatic contraction formalism generalized for elliptical orbits ; as a result , the contraction is weaker than in the simpler case of circular orbits ; see also gnedin et al .
if @xmath99 and do not depend on mass , then the function g is almost independent of mass , and then the baryonic tfr is expected to have a slope similar to the cosmological tfr . on the other hand , due to the gravitational disk halo coupling , as decreases
, g also decreases in such a way that the objects in the baryonic tfr diagram shift nearly parallel to the cosmological tfr .
the function g becomes more sensitive to @xmath99 as becomes larger ; for higher values of , a large variation in g is expected .
this has deep implications for the scatter in the baryonic tfr .
the intrinsic scatter in the baryonic tfr has at least three sources : ( i ) the original scatter from the halo cosmological tfr , ( ii ) the scatter in due to the dispersion in @xmath99 , and ( iii ) the scatter in both and due to the average value and dispersion of . as mentioned above , the latter produces a shift of the models nearly along the same relation , in such a way that its effect on the baryonic tfr scatter is expected to be small for reasonably average values of ( @xmath200 ) . the major contribution to
the scatter comes from the dispersion in the spin parameter @xmath99 . for
a given and , different values of @xmath99 produce disks of different surface densities , and therefore different values of ; such an effect is accounted for by the function g. for a given distribution of @xmath99 , the scatter due to @xmath99 , @xmath201 , increases as increases .
the scatter also increases slightly if we allow for a dispersion in around the mentioned values ( gnedin et al .
the scatter in the halo - relation , @xmath202 , arises because halos of a given mass that were formed through a rapid mass assembling process are more concentrated and have a larger than those that assembled slowly ( avila - reese et al .
1998,1999 ) .
the magnitude of this scatter slightly decreases with mass and has a value of @xmath203 for a @xmath204 halo according to cosmological numerical simulations ( avila - reese et al .
1999,2005).according to the fa - r models , the average scatter in the baryonic tfr is @xmath157 ( = 0.05=const . ) .
this value is a lower limit under the assumptions made
. it could be smaller if the distribution of @xmath99 is narrower than the one in the cdm halos .
this could happen if , for example , disk galaxies form only in a subset of cdm halos , biased to have @xmath99 distribution narrower than the overall sample ( e.g. , donghia & navarro 2007 ; dutton et al .
2007 ; gnedin et al . 2007 ) , or if the baryon spin parameter becomes smaller due to angular momentum transport processes ; then the galaxies formed from the low end of the halo @xmath99 distribution would no longer be a disk type . according to the models of disk galaxy formation discussed above , for a given halo radius , the typical size of the disk , for instance its scale length , is mainly proportional to @xmath205 , where @xmath135 is a function that depends weakly on the nfw halo concentration parameter @xmath186 ( mo et al .
1998 ) , given on its own by the halo mah .
recall that detailed angular momentum conservation was assumed . the weak anti
correlation of @xmath186 with introduces a very weak correlation in with .
therefore , if the model parameters @xmath99 and ( @xmath206 ) do not depend on the mass , then the slope of the - correlation is expected to be only slightly steeper than the slope of the - correlation , which is @xmath207 ( appendix b ) .
however , the correlation is predicted to be highly scattered , with a systematical shift in the normalization due to @xmath99 ( recall that is directly proportional to @xmath99 while , in principle , is independent of @xmath99 ) , and also due to and @xmath186 .
the first two parameters determine the surface density of the disk ; therefore , in the - diagram , the scatter should be strongly correlated with the surface density , e.g. .
note that a dependence of @xmath99 or on mass would imply a change in the slope of the - correlation .
for example , if @xmath99 anti correlates with mass , then this slope will be shallower than that of the initial - correlation , but if anti correlates with mass , then the slope will be steeper . in the context of the simple models discussed in this appendix , we have @xmath208 and @xmath209 .
thus , if for the halos @xmath210 , then one obtains that @xmath211/\lambda g[c ] ) \hb$ ] .
as mentioned above , @xmath135 correlates very weakly with mass , and hence with radius .
then , if @xmath99 and are independent of mass , the expected - correlation should be slightly shallower than the cosmological one . on the other hand , if correlates ( anti correlates ) with mass , then the slope becomes steeper ( shallower ) .
the scatter around the - correlation is expected to be the largest one of the three scaling relations .
galaxies are scattered in the - diagram by their associated values of @xmath99 , and @xmath186 .
differences in each of these values spread the galaxies in a divergent way in both axes .
therefore , the scatter of the - relation is expected to correlate significantly with both and galaxy color .
recall that in the - diagram , galaxies are scattered by @xmath99 and @xmath186 only along the axis . in the - diagram
the spread produced by these parameters is also only along one axis ( ) , but furthermore , the spread produced by in both axes is in such a way that galaxies move nearly along the main relation ; that is why the tfr is the least scattered among the three relations . | we explore how the slopes and scatters of the scaling relations of disk galaxies ( -@xmath0[-@xmath1 , @xmath2-@xmath0[-@xmath1 , and -@xmath2 ) do change when moving from @xmath3 to @xmath4 bands and to stellar and baryonic quantities . for our compiled sample of 76 normal , non
interacting high and low surface brightness ( sb ) disk galaxies , we find important changes , which evidence evolution effects , mainly related to the gas infall and star formation ( sf ) processes .
we also explore correlations among the @xmath5 color , stellar mass fraction , mass @xmath6 ( luminosity @xmath0 ) , and surface density ( sb ) , as well as correlations among the residuals of the scaling relations , and among these residuals and those of the other relations studied here .
some of our findings are : ( i ) the scale length is a third parameter in the baryonic tully fisher relation ( tfr ) and the residuals of this relation follow a trend ( slope @xmath7 ) with the residuals of the - relation ; for the stellar and @xmath4 band cases , the scale length is not anymore a third parameter and the mentioned trend disappears ; ( ii ) among the tfrs , the @xmath8band tfr is the most scattered ; in this case , the color is a third parameter , in agreement with previous works ; ( iii ) the low sb galaxies break some observed trends in diagrams that include surface density , color , and , suggesting then a threshold in the gas surface density @xmath9 , below which the sf becomes independent of the gas infall rate and @xmath9 .
our results are interpreted and discussed in the light of @xmath10 cold dark matter based models of disk galaxy formation and evolution .
these models are able to explain not only the baryonic scaling correlations , but also most of the processes responsible for the observed changes in the slopes , scatters , and correlations among the residuals when changing to stellar and luminous quantities .
the galaxy baryon fraction , , is required to be smaller than 0.05 on average .
we detect some potential difficulties for the models : the observed color-@xmath6 and surface density-@xmath6 correlations are steeper , and the intrinsic scatter in the baryonic tfr is smaller than those predicted . |
some of the flare models assign a fundamental role to the high energy particle beams in the flare energy transport . as the beams interact with the ambient plasma , their energy is dissipated and transformed mainly into the thermal energy of the transition region and chromosphere plasma .
several models studied electron and proton beams as heating agents ( e.g. , @xcite ) as well as their influence on spectral line profiles ( e.g. , @xcite ) . the propagation of electron beams is inevitably connected with the so - called return current ( rc , @xcite ) which also contributes to the beam energy dissipation . besides the heating , the beams influence atomic level populations of the ambient plasma via collisional excitation and ionization .
the work presented here concentrates on these two effects which are commonly neglected in the flare modeling and assesses their importance on the formation of h@xmath0 line in solar flares .
the beam propagation and energy deposition is modelled by a test particle approach consistently with the hydrodynamics of the atmosphere and nlte radiative transfer in the transition region , chromosphere , and photosphere .
details of the model and the methods used are described in @xcite .
we study the response of quiet sun atmosphere ( val c from @xcite ) to beam pulses of short duration , 1 s with sinus - like time modulation , and power - law energy spectrum with @xmath1 .
the model takes into account coulomb collisions of the beam with ambient neutrals and electrons , scattering of beam electrons , and optionally rc for the case of an electron beam ( return current is a factor of @xmath2 lower for protons carrying the same power as electrons ( @xcite ) and is neglected ) .
the return current is included in a runaway approximation assuming @xmath3 , i.e. 10% of ambient electrons carry rc . for details and other approximations of rc in solar atmosphere conditions
see @xcite . total energy deposit ( solid ) and energy deposit to hydrogen ( dashed ) at beam flux maximum . _
top _ : electron beam with and without rc included .
_ bottom _ : proton beam with @xmath4 mev and @xmath5 mev .
, title="fig:",scaledwidth=80.0% ] total energy deposit ( solid ) and energy deposit to hydrogen ( dashed ) at beam flux maximum . _
top _ : electron beam with and without rc included .
_ bottom _ : proton beam with @xmath4 mev and @xmath5 mev . ,
title="fig:",scaledwidth=80.0% ] time evolution of the h@xmath0 line center ( left column ) and wing ( right column ) intensities . _
top _ : electron beam for @xmath6 erg@xmath7s@xmath8 and @xmath1 with and without rc .
_ bottom _ : proton beam for @xmath9 erg@xmath7s@xmath8 , @xmath1 , and @xmath10 mev .
the solid curves refer to cases with @xmath11 included , dashed lines to cases without @xmath11 .
, title="fig:",scaledwidth=37.0% ] time evolution of the h@xmath0 line center ( left column ) and wing ( right column ) intensities . _
top _ : electron beam for @xmath6 erg@xmath7s@xmath8 and @xmath1 with and without rc .
_ bottom _ : proton beam for @xmath9 erg@xmath7s@xmath8 , @xmath1 , and @xmath10 mev .
the solid curves refer to cases with @xmath11 included , dashed lines to cases without @xmath11 .
, title="fig:",scaledwidth=37.0% ] + time evolution of the h@xmath0 line center ( left column ) and wing ( right column ) intensities . _
top _ : electron beam for @xmath6 erg@xmath7s@xmath8 and @xmath1 with and without rc .
_ bottom _ : proton beam for @xmath9 erg@xmath7s@xmath8 , @xmath1 , and @xmath10 mev .
the solid curves refer to cases with @xmath11 included , dashed lines to cases without @xmath11 .
, title="fig:",scaledwidth=37.0% ] time evolution of the h@xmath0 line center ( left column ) and wing ( right column ) intensities . _
top _ : electron beam for @xmath6 erg@xmath7s@xmath8 and @xmath1 with and without rc .
_ bottom _ : proton beam for @xmath9 erg@xmath7s@xmath8 , @xmath1 , and @xmath10 mev .
the solid curves refer to cases with @xmath11 included , dashed lines to cases without @xmath11 .
, title="fig:",scaledwidth=37.0% ] h@xmath0 contribution function @xmath12 for an electron beam .
rc and @xmath11 were included .
solid curve : energy deposit on hydrogen .
, title="fig:",scaledwidth=49.0% ] h@xmath0 contribution function @xmath12 for an electron beam .
rc and @xmath11 were included .
solid curve : energy deposit on hydrogen .
, title="fig:",scaledwidth=49.0% ] h@xmath0 contribution function @xmath12 for an electron beam .
rc and @xmath11 were included .
solid curve : energy deposit on hydrogen .
, title="fig:",scaledwidth=49.0% ] h@xmath0 contribution function @xmath12 for an electron beam .
rc and @xmath11 were included .
solid curve : energy deposit on hydrogen .
, title="fig:",scaledwidth=49.0% ] comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) . @xmath11 were included .
, title="fig:",scaledwidth=46.0% ] comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) .
@xmath11 were included .
, title="fig:",scaledwidth=46.0% ] + comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) .
@xmath11 were included .
, title="fig:",scaledwidth=46.0% ] comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) .
@xmath11 were included .
, title="fig:",scaledwidth=46.0% ] + comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) .
@xmath11 were included .
, title="fig:",scaledwidth=46.0% ] comparison of h@xmath0 contribution function @xmath12 for proton beams with @xmath4 mev ( first column ) and @xmath5 mev ( second column ) .
@xmath11 were included .
, title="fig:",scaledwidth=46.0% ] figure [ kasparova - fig : deposit ] shows that rc significantly increases the energy deposit of electron beams at heights @xmath13 km leading to corresponding increase of the temperature and ionization .
the atmospheric response to proton beams was modelled for two different values of the low - energy cutoff @xmath10 mev .
note that deka - mev protons produce approximately the same amount of hard x - rays as deka - kev electrons ( @xcite ) .
figure [ kasparova - fig : deposit ] illustrates that proton beams deposit their energy into deeper layers than electron beams ( @xcite ) .
the energy deposit of proton beams with lower value of @xmath14 peaks at higher atmospheric layers and is larger at heights above in comparison with a proton beam of @xmath5 mev .
consequently , the temperature at these layers is substantially increased .
the hydrogen level populations are affected also by non - thermal collisional ionization and excitation .
the corresponding rates @xmath11 are directly proportional to the beam s energy deposit on hydrogen ( see @xcite for electron beams , and @xcite for proton beams ) .
their influence on the temporal evolution of the h@xmath0 profile was studied for electron beams by @xcite . here , we describe in detail their effect for proton beams and the influence of rc for electron beams .
the temporal evolution of the h@xmath0 line - center ( @xmath15 ) and wing ( @xmath16 ) intensities are shown in fig .
[ kasparova - fig : prof ] .
as discussed in @xcite and @xcite for the case of electron beams , @xmath11 cause a decrease of the line - center intensity at the very start of the beam propagation and enhance the wing intensity later on ( mainly for fluxes @xmath17 erg@xmath7s@xmath8 ) .
the line behaviour can be understood in terms of contribution functions @xmath12 to the outgoing intensity given by @xmath18 , where @xmath19 is the height .
figure [ kasparova - fig : cfel ] shows the evolution of h@xmath0 @xmath12 for the case of an electron beam with rc and @xmath11 included .
a decrease of the line center intensity is caused by an increase of opacity due to @xmath11 ( see @xmath12 at 0.2 s ) .
later on , a new region of wing formation occurs at the layers where the energy deposit peaks ( see @xmath12 at 0.4 s ) .
such a layer is not formed if @xmath11 are not considered . since rc causes heating of the top parts of the atmosphere , it is responsible for the increase of the line center intensity at 0.4 s ( see fig . [
kasparova - fig : prof ] ) forming in a narrow region at @xmath20 km ( see fig .
[ kasparova - fig : cfel ] ) .
the situation for a proton beam is shown in fig .
[ kasparova - fig : cfproton ] . in the case of @xmath5 mev ,
the beam energy is deposited in regions where the line wings are formed .
if @xmath11 are included , they cause increase of opacity which leads to a drop of both the wing and the line - center intensities ( see @xmath12 at 0.2 s ) .
later on , a new wing region occurs .
neglecting @xmath11 , the line intensity does almost not change because the temperature structure is not significantly affected by the beam propagation .
however , for a proton beam of @xmath4 mev , the temperature increase is large enough to form a new region of the line center intensity , as in the case of rc for an electron beam .
also , the energy deposit is large enough to create a new layer of strong wing intensity ( see @xmath12 at 0.4 and 1.1 s and fig .
[ kasparova - fig : prof ] ) .
the second peak of the line center intensity at about 1.1 s is due to broadening of the formation region .
similarly to the electron beams , the line wings are not changed if @xmath11 are not considered .
the h@xmath0 line is influenced by temperature structure resulting from the beam propagation and return current as well as the non - thermal collisional rates .
we propose to use the decrease of h@xmath0 line - center intensity as diagnostic indicating the presence of particle beams .
we also predict that proton beams with deka - mev low - energy cutoffs produce only _ decrease _ of the h@xmath0 in comparison with quiet - sun intensities .
+ this work was mainly supported by the grants 205/04/0358 and 205/06/p135 ( ga cr ) , partially by grants iaa3003202 , iaa3003203 ( ga as ) , and by the key project av0z10030501 ( ai as ) . | using 1d nlte radiative hydrodynamics we model the influence of the particle beams on the h@xmath0 line profile treating the beam propagation and the atmosphere evolution self - consistently .
we focus on the influence of the non - thermal collisional rates and the return current . based on our results ,
we propose a diagnostic method for determination of the particle beam presence in the formation regions of the h@xmath0 line . |
the quest for the direct observation of gravitational radiation is one of the most pressing challenges in the physics of this century .
predicted by einstein shortly after formulating his general theory of relativity , gravitational waves ( gw ) will allow us to probe regions of space - time otherwise unobservable in the electromagnetic spectrum @xcite .
although we have excellent indirect evidence of their existence through their effects on the orbital evolution of binary systems containing pulsars , we have not been able yet to directly detect them .
several ground - based gravitational wave detectors have been operational for awhile , and only recently they have been able to identify the most stringent upper limits to date for the amplitudes of the radiation expected from several classes of sources @xcite .
next generation of earth - based interferometers and pulsar - timing experiments @xcite , as well as newly conceived space - born detectors @xcite are expected to achieve this goal .
ground - based interferometers operate in the frequency band whose lower limit is at about @xmath4 hz , mainly because of the large seismic and gravity - gradient noises below this frequency cut - off . since the mhz region is potentially very rich in gravitational wave sources , the most natural way to observe them is to build and operate a space - borne detector .
the most notable example of a space interferometer , which has been under study for several years jointly by scientists in the united states of america and in europe , is the laser interferometer space antenna ( lisa ) mission . by relying on coherent laser beams exchanged between three remote spacecraft forming a giant ( almost ) equilateral triangle , lisa aimed to detect and study cosmic gravitational waves in the @xmath5 hz band .
although over the years only a few space - based detector mission concepts have been considered as alternatives to the lisa mission ( with astrod @xcite , decigo @xcite , and omega @xcite being the most notable examples ) , starting in 2011 ( with the ending of the nasa / esa partnership for flying lisa ) more mission concepts have appeared in the literature @xcite .
their goals are to meet most ( if not all ) the lisa scientific objectives ( highlighted in the 2010 astrophysics decadal survey _
new worlds , new horizons ( nwnh ) _
@xcite ) at a lower cost . in this context , motivated by the interest of the brazilian space agency to pursue the development and launch of several geostationary satellites , we decided to analyze the scientific capabilities offered by a geostationary gravitational wave interferometer , henceforth called geograwi .
geograwi was one of the alternative concepts to the lisa mission submitted in response to nasa s request for information # nnh11zda019l @xcite .
a detailed analysis of all the submitted projects and a final study report can be found in @xcite . among them , a proposal by sean t. mcwilliams explores a mission concept very similar to the one discussed in this paper @xcite called gadfli .
his concept and the one proposed by us , though strikingly similar , were nonetheless independently developed and submitted .
geograwi , like lisa , has three identical spacecraft interchanging coherent laser beams and forming an ( almost ) equilateral triangle .
however , by being in a geostationary orbit , its accessible frequency band changes to about ( @xmath6 hz ) .
it is worth noting that , although our interferometer reminds that of the omega mission ( proposed by r. hellings @xcite about 15 years ago ) , it is in fact very different from it .
the omega spacecraft were six in total , with two at each vertex of a geocentric equilateral triangle of side equal to one million kilometers .
cost of the launching vehicle as well as challenges in performing the phase measurements associated to the large relative velocities between the spacecraft make omega quite different from geograwi . the astrophysical sources that geograwi is expected to observe within its operational frequency band include extra - galactic massive and super - massive black - hole coalescing binaries , the resolved galactic binaries and extra - galactic coalescing binary systems containing white dwarfs and neutron stars , a stochastic background of astrophysical and cosmological origin , and possibly more exotic sources such as cosmic strings . with geograwi we will be able to test einstein s theory of relativity by comparing the waveforms detected against those predicted by alternative relativistic theories of gravity , and also by measuring the number of independent polarizations of the detected gravitational wave signals @xcite .
since the sensitivities of the geostationary interferometers we considered ( see section [ sensitivity ] below ) are degraded by their shorter armlength in the frequency region ( @xmath7 ) hz , it will be impossible for them to detect the zero - order cyclic spectrum of the white dwarf - white dwarf galactic binary confusion noise .
however , as pointed out in @xcite , the periodic nature of the galactic background signal in the data of an interferometer rotating around the sun will contain higher - order `` cyclic spectra '' , which are in principle not affected by the instrumental noise ( this is stationary ) .
this implies that we could still detect the so - called `` confusion noise '' , and infer properties of the distribution of the white - dwarf binary systems present in our galaxy @xcite . in this article
we pay particular attention to super - massive black holes , which are the main sources of gws for interferometers operating at frequencies below @xmath8 hz .
we show that geograwi will be able to detect a large number of smbhs at very high redshifts , and that a significant fraction of them will have masses smaller than @xmath9 m@xmath3 as a consequence of its improved sensitivity at higher frequencies .
this paper is organized as follows . in section [ sensitivity ]
we provide a description of the mission and a system noise analysis in order to evaluate the sensitivities of the time - delay interferometric ( tdi ) @xcite response @xmath10 .
it is in this section and in [ noise ] where we also emphasize that additional instrumental techniques over those baselined for lisa will need to be implemented in order to achieve the derived sensitivities .
these are estimated for three different on board subsystem configurations of geograwi , which we refer to as geostationary 1 ( geo1 ) , geostationary 2 ( geo2 ) , and geostationary 3 ( geo3 ) .
we note that the geo1 configuration has a reduced sensitivity over that of lisa by a factor of about @xmath11 in the lower part of its accessible frequency band ( @xmath7 hz ) . on the other hand , in section [ science ]
we discuss possible sources of gravitational waves observable by geograwi in the higher - part of its band ( @xmath1 hz ) .
particular attention is devoted to smbhs and we show that the number of observable signals from such systems , with masses smaller than or equal to @xmath12 m@xmath13 , would be significant . finally , in section [ conclusions ] we provide a summary of our work and conclusions .
our proposed space - based detector entails three spacecraft in geostationary orbit , forming an equilateral triangle with armlength of about @xmath14 km .
since the constellation plane coincides with the earth equatorial plane , each spacecraft will need to be spherical and entirely covered with photo - voltaic cells .
this will allow each spacecraft to be continuously powered ( without the need of rotating it as the earth orbits around the sun ) and maintain a high - level of thermal stability .
it should be noticed also that the inclination ( @xmath15 ) of the earth equatorial plane with respect to the ecliptic will prevent ( i ) sun light from contaminating the optics of the spacecraft during most of the interferometer orbital period around the sun , and ( ii ) spacecraft occultation by the earth . in proximity of the equinoxes , however , these properties will no longer be true and , in order to avoid these short - period operation interruptions , onboard electric power and sun - light filters will need to be used .
sun - light filters needed for this purpose have already been developed for the omega mission , which requires their usage at all times in its geocentric trajectory @xcite . as the sun and the moon will exercise gravitational perturbations on each spacecraft ( resulting into a long - term orbital drift ) , in order to maintain orbital stability and small inter - spacecraft relative velocities
, each spacecraft will need to implement `` station - keeping '' @xcite .
this is required not only for keeping the constellation in a stable configuration , but most importantly for relying on a phase - meter design that does not need to accommodate large relative frequency offsets , making it significantly less noisy ( see [ noise ] ) . by taking the nominal value of @xmath16 for the `` east - west '' acceleration perturbation @xcite acting on the spacecraft ( which is responsible for the spacecraft relative velocities ) , it is easy to show that , in order to maintain their maximum velocities smaller than about a few decimeters per second , the station - keeping operation will need to be applied about once per week ( see [ noise ] ) . in order to estimate the tdi sensitivities of our geostationary gw interferometer detector , we have relied on some lisa study documents @xcite .
this allowed us to derive a break - down estimate of the various noise sources affecting the tdi observables and compute the tdi sensitivities of a geostationary interferometer under the following three different on - board subsystem configurations : a. some of the onboard science payload components ( the laser , the optical telescope , the inertial reference sensor ) are assumed to be equal and have a similar noise performance of those planned to be flown onboard the lisa mission .
other subsystems are regarded to have a noise performance that results in a high - frequency noise spectrum that is essentially determined by the photon - counting statistics .
it should be emphasized , however , that our assumed noise performance in the high - frequency region of the accessible band requires additional instrumental developments over those baselined for the lisa mission @xcite . in order to better understand this point we refer the reader to [ noise ] where it is discussed somewhat at length .
we will refer to this configuration as `` geostationary 1 '' ( or geo1 ) .
b. the output power of the onboard lasers and the size of the optical telescopes are assumed to be equal to those of the lisa mission , while the noise performance of the gravitational reference sensors ( grs ) is taken to be a factor of ten worse than that of the grss planned for the lisa mission .
the remaining noises are as in the geo1 case , and this configuration will be referred to as `` geostationary 2 '' ( or geo2 ) . c. the noise performance of each grs is taken a factor of ten worse than that of the grs planned for the lisa mission , the output power of the lasers is assumed to be a factor of @xmath4 smaller than that of the lasers onboard lisa , the diameter of the optical telescopes has been reduced by a factor of @xmath17 over that of the lisa telescopes , and the remaining noises are assumed to be equal to those for the geo1 configuration .
this configuration will be called `` geostationary 3 '' ( or geo3 ) .
the sensitivity of an interferometer detector of gravitational radiation has been traditionally taken to be equal to ( on average over the sky and polarization states ) the strength of a sinusoidal gw required to achieve a signal - to - noise ratio of @xmath18 in a one - year integration time , as a function of fourier frequency .
for sake of simplicity we will limit the derivation of the sensitivities of the three geostationary configurations described above to only the `` x '' tdi combination , as those for the other tdi combinations can be inferred by properly scaling those for @xmath10 . to this end
we will be following the same procedure described in @xcite , and rely on the following expression for the power spectrum of the noises affecting the @xmath10 combination ( see figure [ noisegeo ] ) @xmath19 s^{pm}_y ( f ) + 16 \sin^2(2 \pi f l)\ s^{op}_y ( f ) \ , \label{eq:10}\ ] ] where @xmath20 is the spectrum of the relative frequency fluctuations due to each proof mass , and @xmath21 is the spectrum of optical path ( mainly shot and beam pointing ) noise . both these noises have been regarded as the main limiting noise sources for lisa @xcite , and will be treated as such for our geostationary interferometer configuration ( see [ noise ] for details ) .
it should be acknowledged , however , that the dramatic reduction in optical path noise assumed in this article reflects the implementation of additional instrumental techniques over those baselined for lisa @xcite .
the numerical expressions for the spectra @xmath20 , @xmath22 associated to the lisa mission were provided in @xcite and they are equal to @xmath23^{-2 } \
{ \rm hz}^{-1}$ ] and @xmath24 ^ 2 \ { \rm hz}^{-1}$ ] .
the configurations ( i ) , ( ii ) and ( iii ) discussed above require appropriate scaling factors of these noise levels , and they are given in table [ table1 ] ( also see [ noise ] for detail analysis of the other noise sources ) .
.noise spectra scaling factors as functions of the different instrument configurations analyzed .
see text for more details . [
cols="^,^,^",options="header " , ] the merger phase can be related to the fourier frequency interval associated to the merger ( lower ) frequency @xmath25 all the way to the ring - down frequency @xmath26 . during this evolutionary phase
a significant amount of gw energy , comparable to that emitted during the ring - down phase , is radiated .
one of the main issues regarding the characterization of the merger phase is the determination of the frequency @xmath25 .
we do not address this issue here , and refer the reader to ajith _
@xcite , where it is shown that @xmath27 for equal mass bbh systems .
also in @xcite it is shown that there exist a reliable analytic approximation for the merging - phase waveforms derived numerically . the resulting expression for the radiated energy spectrum is equal to @xmath28 which is valid for two non - spinning and equal mass black - holes . finally , substituting the above equation into equation ( [ snr ] ) we get @xmath29^{5/3}}{30 \pi^{4/3 } f_{m } d_{l}(z)^2 } \ , \int_{f_{min}/(1+z)}^{f_{m}/(1+z ) } \ , \frac{df}{f^{4/3 } s_h(f)}. \label{snr2m}\ ] ] following the same prescription as for the ring - down case , we can then calculate the redshift , @xmath30 , below which a given smbh of mass @xmath31 can be detected with a given snr . in figure [ merg ]
we plot @xmath30 , for the three different geograwi configurations , as a function of the mass @xmath31 of the smbh system . as before we have included the results valid for lisa and lisa cn .
also in this case the geo1 configuration represents an interesting alternative to lisa as merging smbhs could be seen at redshifts as high as @xmath32 . as a function of the smbh mass m for @xmath33 , and null spin , for the different geograwi configurations . included for comparison are the results for lisa and lisa cn.,width=377 ] for sake of curiosity , we compared our results against those corresponding to a `` white '' spectrum @xcite .
we found that the two would agree quite well by assuming a white spectrum corresponding to an energy emission efficiency of about @xmath34 ( i.e. @xmath34 of the total mass @xmath31 of the bbhs is converted to gravitational radiation ) . in this section
we finally analyze the inspiraling phase of a binary system containing smbh . in this case the signals frequency components will fall into the region of the band where @xmath35 , which is accessible to all the geograwi configurations considered here .
the energy spectrum for the inspiraling phase is given by the well known formula @xcite
@xmath36 and , for the particular case of binary systems whose components have equal mass , the snr becomes equal to @xmath29^{5/3}}{30 \pi^{4/3 } d_{l}(z)^2 } \ , \int_{f_{min}/(1+z)}^{f_{m}/(1+z ) } \ , \frac{df}{f^{7/3 } s_h(f)}. \label{snr2}\ ] ] note that now , in order to obtain @xmath30 as a function of @xmath31 , we had to treat the @xmath37 and @xmath38 as free parameters .
as before , we take the @xmath33 .
the value of @xmath38 is chosen in such a way that the signal is integrated ( observed ) over the last year of inspiraling .
for binary systems whose components have equal masses it is easy to show that @xmath38 is given by the following expression @xmath39^{5/3 } t \right\}^{-3/8},\ ] ] where @xmath40 is the time before merging . in figure [ insp ]
we plot @xmath30 as a function of the mass of the system of smbhs for @xmath41 and for one year of observation . as a function of the smbh mass m for @xmath33 for the different geograwi configurations
. also shown for comparison are the curves corresponding to lisa and lisa cn.,width=377 ] although fig .
( [ insp ] ) shows that , at an snr=10 , the maximum redshift achievable by lisa is consistently larger than that for geo1 , also indicates that the maximum redshift at which the coalescence phase can be observed by geo1 is equal to about 5 . in summary , lisa can reach a larger redshift as compared to any geograwi configuration independently of the smbh masses .
the main goal of this article was to analyze the performance and some scientific capabilities of a geostationary gravitational wave interferometer .
we have done so by estimating its sensitivities to gravitational radiation when operated under three different onboard subsystem configurations , and analyzed the kind of gravitational wave sources it will be able to detect and study in the sub - hertz band .
we found our proposed earth - orbiting detector less sensitive than the laser interferometer space antenna mission by about a factor of seventy in the lower part of its accessible frequency band ( @xmath42 hz ) . on the other hand ,
in the higher part of the band ( @xmath43 hz ) , it was shown that geograwi has observational capabilities that are complementary to those of the lisa mission . in relation to this point
, we emphasized that our assumed noise performance in the high - frequency region of the accessible band requires additional instrumental capabilities over those baselined for the lisa mission .
although current experimental evidence @xcite indicates that they will be available by the time such a mission will be considered , a more in - depth analysis of these challenges is beyond the scope of this paper . in the case of binary systems containing smbhs
, our analysis has shown that the geo1 interferometer could observe a great number of events per year .
this is direct consequence of its good sensitivity in the frequency region ( @xmath43 hz ) , where lighter ( and larger in number ) black - holes radiate in the merger and ring - down phases .
consistently we have also shown that these same systems will be detectable by lisa with a snr larger than that achievable by geo1 during the inspiraling phase ( see fig .
[ insp ] ) . since most of the snr from the coalescence phase of a binary system containing smbhs is achieved after about a week of integration time before coalescence , because geograwi will make most of its observations in the high - part of its accessible frequency band , and because during a time period of one week the amplitude modulation of the geograwi response will result into a frequency modulation of the observed gw signal that depends on the source location in the sky , one can argue that geograwi should achieve very good accuracies in the reconstructed gw source parameters of smbh with masses that are in the range @xmath44 .
it is worth noting that mcwilliams @xcite has performed such an analysis for the gadfli mission concept ( which is similar to geograwi ) .
he showed that the precision in the estimation of the parameters achievable by gadfli ( which is analogous to our geo1 configuration ) is somewhat better than that for lisa .
although these results imply that geo1 should achieve comparable accuracies in estimating the parameters characterizing smbhb , it is our intention to quantitatively study this problem in a forthcoming article .
since the high - part of the accessible frequency where geograwi achieves its best sensitivity is at around @xmath8 hz , we will investigate also its ability to perform low - latency searches of burst events . mainly due to its capability of continuously transmitting real - time data to the ground ( by being geostationary )
, it should be able to trigger simultaneous searches for electromagnetic counter - parts .
m.t . acknowledges financial support from fapesp through its visiting professorship program at the instituto nacional de pesquisas espaciais ( i.n.p.e . ) in brazil ( grant # 2011/11719 - 0 ) .
jcna and oda would like to thank cnpq and fapesp for financial support .
mesa would like to thank fapemig for financial support ( grant apq-00140 - 12 ) .
last , but not least , we would like to thank the referee for his ( her ) useful suggestions and criticisms .
in this appendix we provide an analysis of the noise sources affecting the heterodyne measurements performed by the geostationary gw interferometer geo1 .
since some of its onboard instrumentation is similar to that of the lisa mission , it imposes the most stringent noise - performance requirements on the subsystems affecting the one - way heterodyne measurements .
furthermore , as it will be discussed below , the noise performance in the high - frequency region of the accessible band requires additional instrumental capabilities over those baselined for the lisa mission .
although current experimental evidence @xcite induces us to believe that they will be available by the time such a mission will be further analyzed , it is clear that these potential challenges are well beyond the scope of this paper .
our analysis will also rely on two lisa study documents as guiding references @xcite .
there it was shown that there exist two main categories of sensitivity - limiting noise sources : a. acceleration noises b. optical - path noises the _ acceleration noises _ are due to residual forces acting on the proof - masses of the gravitational reduction system ( grs ) , and result into doppler fluctuations into the heterodyne measurements .
their magnitudes are most prominent in the low - part of the accessible frequency band ( @xmath45 hz ) and depend on the ( i ) adopted grs design , ( ii ) the spacecraft design , and ( iii ) the mission trajectory and space environment within which the spacecraft will be operating . for these reasons
the classification of the acceleration noises is a multi - parameters problem and a very challenging one . in the case of the lisa mission
this has already been studied extensively and it will be finalized through the lisa pathfinder experiment @xcite .
this is a esa mission aimed at testing the noise modeling , classification , and performance of the lisa grs system .
the lisa grs was envisioned to have two cubic proof - masses onboard each spacecraft and whose positions relative to the spacecraft were measured with electrostatic ( capacitative ) readouts that were part of their caging systems . in recent years , however , it has been argued that a single , spherical proof - mass design ( whose position relative to its enclosing cage is measured optically ) could be used instead , providing significant simplifications of the grs and of the optical bench where the heterodyne measurements are performed @xcite .
although a single , spherical proof - mass grs implementation might result into a better noise performance than that with two cubic masses @xcite , in our analysis we will assume geo1 to rely on a spherical grs design whose performance is similar to that of the lisa grs , i.e. with a square - root of the acceleration spectral density equal to a value of @xmath46 over the accessible frequency band .
in ultimate analysis it will be the lisa pathfinder experiment that will show us whether the types and magnitudes of the noise sources affecting the grs performance are what we expect them to be .
the primary noise sources within the second category , i.e. _ optical - path noises _ , are much easier to identify and are most prominent in the higher part of the accessible frequency band .
they can be summarized as @xcite * shot - noise at the photo - detectors , * residual laser phase noise in the tdi observables , * laser beam - pointing fluctuations , * phase - meter noise , * master clock noise , * scattered light effects . the _ shot - noise _ is a fundamental noise limitation to the sensitivity of a laser interferometer gw detector in the high - part of its accessible frequency band .
it affects the one - way doppler measurements right at the photo - detector where two laser beams are made to interfere , and it leads to the following spectral density of relative frequency ( doppler ) fluctuations @xcite @xmath47 in eq .
( [ shot ] ) @xmath48 is the planck constant , @xmath49 is the nominal laser frequency , @xmath50 is the fourier frequency , and @xmath51 is the effective power available at the receiving photodetector . by assuming the same optics , optical telescope size , laser power , and photodetector quantum efficiency as those of lisa , from eq .
( [ shot ] ) we conclude that the amplitude of the relative frequency fluctuations due to shot noise affecting the geo1 interferometer scale down linearly with the interferometer armlength .
the _ residual laser phase noise _ represents the `` left - over '' of the laser noise in the tdi observables after the tdi algorithm is applied to the one - way doppler measurements by properly time - shifting and linearly combining them .
it is primarily due to the finiteness in the accuracy by which the armlengths are known , and is proportional to the amplitude of the laser frequency fluctuations .
the relationship between the magnitude of the spectrum of the residual laser frequency fluctuations , @xmath52 , and the armlength accuracies , @xmath53 , is given in @xcite for the unequal - arm michelson interferometer tdi combination and has the following form @xmath54\right\ } \ .
\label{lasertrue}\end{aligned}\ ] ] in the long - wavelength limit it is easy to show from the above equation that the amplitude of the residual laser frequency fluctuations scale linearly with the armlength of the interferometer ( see eq . ( 3.18 ) of @xcite ) .
on the other hand , this scaling no longer exists at higher - frequencies and , in order to maintain this noise source negligible in the geo1 noise budget , a higher level of accuracy in the knowledge of the armlengths is required . since the geo1 configuration will have a shot noise amplitude smaller than that of lisa by about a factor of @xmath11 , by measuring the geo1 armlength with an accuracy that is seventy times better than that of lisa we will make this noise source negligible in the geo1 tdi combination @xmath10 .
given that the required lisa armlength accuracy for @xmath10 has been estimated to be equal to about @xmath55 meters @xcite , we conclude that the armlength accuracy needed for geo1 should be of about @xmath56 centimeters .
such a level of laser ranging accuracy has already been demonstrated experimentally @xcite at a much lower receiving laser power .
this means that , in the case of the geo1 mission , the achievable armlength accuracy will be smaller than @xmath56 cm , further reducing the geo1 residual laser frequency noise .
the _ laser beam - pointing fluctuations _ are due to distortions in the transmitted laser beam wavefront that appear at the receiving spacecraft as additional frequency fluctuations generated by small pointing fluctuations from the transmitting spacecraft . in the case of the lisa mission , whose pointing fluctuations specifications were required not to be greater than @xmath57 , it was shown @xcite that a way for substantially reducing pointing - induced frequency fluctuations was to rely on the light received from the far spacecraft to sense the orientation of the receiving spacecraft . an ingenious way for doing
so @xcite is by sampling some of the incoming light on a ccd array . by detecting the position of the beam on the ccd
it is then possible to deduce the alignment of the spacecraft and implement it by proper control signals .
this measurement is shot - noise limited and , in the case of the lisa mission , it was proposed to implement it by extracting @xmath4 percent of the power of the incoming laser light .
this resulted into a pointing noise of about @xmath58 , more than a factor of @xmath4 better than that specified for lisa .
since the shot - noise level associated with this pointing measurement technique scales down linearly with the armlength ( by being inversely proportional to the square - root of the received laser power ) we conclude that the pointing shot - noise limit for geo1 will scale down linearly with the armlength from the level estimated for lisa .
the _ phase - meter _ is the subsystem that measures the difference between the phase of the incoming laser beam and that from the local laser , and in the process it introduces an additional phase noise in the one - way doppler measurements .
although the noise it generates does not scale with the armlength , it does depend on the magnitude of the relative frequency offset between the transmitted and received frequencies of the two interfering laser beams @xcite .
as an example , it has been shown that the frequency stability of the phase - meter to be flown on the lisa pathfinder mission ( which has an heterodyne frequency range of about a few khz ) will be several orders of magnitude better than that of the phase - meter for the lisa mission @xcite .
it is for this reason that , in order to compensate for the sun and the moon gravitational perturbations on each spacecraft and maintain orbital stability and small inter - spacecraft relative velocities , each spacecraft will implement `` station - keeping '' @xcite .
since this operation will need to be performed about once per week @xcite in order to maintain the spacecraft relative velocities smaller than about a few decimeters per second , it will not disrupt significantly the science data acquisition . as an additional note
, geo1 will also rely on molecular iodine laser frequency stabilization , which has been shown to provide laser frequency stability superior to that achievable by optical cavity stabilization and a laser frequency accuracy at the level of @xmath8 to @xmath59 khz @xcite .
the _ master clock noise _ affects the lisa measurements because it is used for removing the large doppler beat - notes ( as large as @xmath60 mhz ) affecting the doppler measurements .
since geo1 will not be subject to large orbital perturbations as those affecting lisa and it will rely on station - keeping in order to maintain a `` stationary '' configuration , the noise due to the onboard master clock for performing the heterodyne measurements will be negligible as it is proportional to the magnitude of the beat note at the photodetector @xcite .
this will result into a significant simplification of the designs of the phase - meter and transmitting modules as modulations of the transmitting and receiving beam ( for implementing the clock - noise cancellation scheme ) will not be needed .
_ scattered light effects _ are due to spurious light signals generated from the interaction of the received main laser beam with the optical telescope , beam - splitter , and other optical components of the optical bench the beam interacts with before being made to interfere with the light of the local laser .
although this noise source for quite sometimes has been thought to be unavoidable , in recent years new developments in laser metrology have indicated that scattered light noise can be suppressed to sub - picometer levels in the frequency band of interest to space - based gw interferometers .
in one such laser metrology technique , called `` digital interferometry '' ( di ) @xcite , laser light is phase modulated with a pseudo - random noise ( prn ) code that time stamps the light , allowing isolation of individual propagation paths based on their times of flight through the system .
after detection , the prn phase shift is removed by decoding with an appropriately delayed copy of the code . by matching the decoding delay to the propagation delay the signal
is coherently recovered , allowing phase measurements to extract displacement information at resolutions much smaller than the laser wavelength .
this matched - delay filtering of di allows isolation of reflections from an object at a particular distance , such as the optical components giving origin to the scattered light noise . in summary , the magnitude of the relative frequency fluctuations due to the shot - noise and the laser beam - pointing fluctuations scale linearly with the armlength , as it follows from the considerations leading to equation ( 4.4 ) at page 82 of @xcite , and sections 3 at page 1576 of @xcite respectively .
although the frequency fluctuations due to the residual laser phase noise , the scattered - light noise , and the phase - meter noise do not follow a similar scaling , we have shown that they can be made negligible within the geo1 noise budget .
this is because of a demonstrated better armlength accuracy @xcite , of digital interferometry as it suppresses scattered light effects , and adoption of a simpler phase - meter design @xcite that takes advantage of much smaller `` beat - note '' frequencies among the lasers .
the latter is made possible through a combination of laser iodine stabilization @xcite together with spacecraft station - keeping maneuvers @xcite .
in this appendix we discuss the differential coalescence rate per logarithm frequency interval as this can be related to the differential event rate associated , for example , with the ring - down phase shown in fig .
[ differ ] . for a more detailed analysis regarding the coalescence rate and the corresponding cosmological simulations ,
we refer the reader to refs . @xcite and @xcite respectively .
the differential coalescence rate is obtained from an appropriate combination of eqs .
( [ erobs ] ) and ( [ frd ] ) and an integration in a given redshift interval , namely @xmath61 , where @xmath62 refers to a value selected at the beginning of the simulation . fig .
[ dcoalrate ] shows the differential coalescence rate per logarithmic frequency interval as a function of the frequency and for different intervals of integration in the redshift .
we have not included curves corresponding to values of the redshift higher than @xmath60 because they all `` saturate '' to the curve corresponding to this value of @xmath62 .
another interesting information concerning the study by filloux _
@xcite refers to the coalescence rate at the observer frame , which can be obtained by integrating the differential coalescence rate .
after performing such an integration we find a coalescence rate of @xmath63 43 yr@xmath64 ( associated with the ring - down phase ) , which should be regarded as the maximum event rate a given gravitational wave antenna might detect .
our study has actually shown that lisa and geo1 should be able to detect a large fraction of this rate .
implicit in the integration just mentioned , there is an integration in the redshift interval @xmath61 , where @xmath62 was taken to be equal to @xmath65 in the present study . in practice
, however , there is a saturation around @xmath66 , as can be seen in fig .
[ coalrate ] , where the cumulative event rate as a function of @xmath62 is plotted and tabulated .
note that half of the coalescence rate comes from events occurring at @xmath67 .
thorne , k. s. , 1987 . in : _
three hundred years of gravitation _
, 330 - 458 , eds . hawking , s. w. , & israel , w. , ( cambridge university press , new york ) http://www.ligo.caltech.edu/ .
http://www.virgo.infn.it/ .
http://www.geo600.uni-hannover.de/ .
http://tamago.mtk.nao.ac.jp/ .
detweiler , s. , 1979 , _ ap .
j. , _ * 234 * , 1100 - 1104 .
p. bender , k. danzmann , & the lisa study team , 1998 , _ laser interferometer space antenna for the detection of gravitational waves _ ,
pre - phase a report , mpq233 ( max - planck- institt fr quantenoptik , garching , germany .
ni , w. t. , 2002 , _ int .
d _ , * 11 * ( 7 ) , 947 .
seto , n. , kawamura , s. , & nakamura , t. , 2001 , _ phys .
lett _ , * 87 * , 221103 .
hiscock , b. , hellings , r. w. , 1997 , _ bull .
_ , * 29 * , 1312 .
http://pcos.gsfc.nasa.gov/studies/gravitational-wave-mission.php/ .
nrc 2010 , http:// www.nap.edu/catalog/12951.html nasa solicitation # nnh11zda019l titled : _
concepts for the nasa gravitational - wave mission_. document available at : http://nspires.nasaprs.com/external/ .
tinto , m. , de araujo , j. c. n. , aguiar , o. d. , & m.e.s .
alves , 2011 , _ a geostationary gravitational wave interferometer ( geograwi ) _ ( http://arxiv.org/pdf/1111.2576.pdf ) .
mcwilliams , s. t. , 2011 , _ geostationary antenna for disturbance - free laser interferometry ( gadfli ) _ ( http://arxiv.org/ps_cache/arxiv/pdf/1111/1111.3708v1.pdf ) . tinto , m. , & alves , m. e. s. , 2010 , _ phys . rev .
d , _ * 82 * , 122003 .
edlund , j. a. , tinto , m. , krlak , a. , & nelemans , g. , 2005 , _ classic .
quantum grav .
_ , * 22 * , s913-s926 .
tinto , m. , & dhurandhar , s. v. , 2005 , _ living rev .
relativity _ , * 8 * , 4 .
hellings , r. , larson , s. l. , jensen , s. , fish , c. , benacquista , m. , cornish , n. , & lang , r. , _ a low - cost , high - performance space gravitational astronomy mission _ ( http://pcos.gsfc.nasa.gov/studies/rfi/gwrfi-0007-hellings.pdf ) .
soop , e. m. , 1994,_handbook of geostationary orbits _ , kluwer academic publishers ( dordrecht , the netherlands ) .
`` lisa : unveiling a hidden universe '' , esa publication # esa / sre(2011)3 ( february 2011 ) . also available at : http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=48364 leonhardt , v. , & camp , j.b . , 2006 , _
applied optics _ , * 45 * , 4142 .
shaddock , d. a. , 2007 , _ opt . lett .
_ , * 32 * , 3355 .
wuchenich , d. m. r. , lam , t. t .- y . ,
chow , j. h. , mcclelland , d. e. , & shaddock , d. a. , 2011 , _ opt .
_ , * 36 * , 672 .
antonucci , f. , _ et al _ , 2011 , _ class .
quantum grav . , _
* 28 * 094001 .
antonucci , f. , _ et al _ , 2011 , _ class .
quantum grav . ,
_ * 28 * 094002 .
armstrong , j. w. , estabrook , f. b. , & m. tinto , 1999 , _ ap .
j. , _ * 527 * , 814 .
estabrook , f. b. , tinto , m. , & armstrong , j. w. , 2000 , _ phys .
d , _ * 62 * , 042002 . flanagan , e. e. , & hughes , s. a. , 1998 , _ phys .
d , _ * 57 * , 4535 . pretorius , f. , 2005 , _ phys .
letters , _ * 95 * , 121101 .
baker , j. g. , centrella , j. m , .
choi , d .-
i . , koppitz , m. & van meter , j. r. , 2006 , _ phys .
letters , _ * 96 * , 111102 campanelli , m. , lousto , c. o. , marronetti , p. & zlochower , y. , 2006 , _ phys .
letters , _ * 96 * , 111101 .
mcwilliams , s. t. , 2011 , _ class .
quantum grav . , _
* 28 * , 134001 hinder , i. , 2010 , _ class .
quantum grav . , _ * 27 * , 114004 centrella , j. m. , baker , j. g. , kelly , b. j. & van meter , j. r. , 2010 , _ phys . rev .
, _ * 82 * , 3069 .
ajith , p. , babak , s. , chen , y. , hewitson , m. , krishnan , b. , sintes , a. m. , whelan , j. t. , brgmann , b. , diener , p. , dorband , n. , gonzalez , j. , hannam , m. , husa , s. , pollney , d. , rezzolla , l. , santamara , l. , sperhake , u. , & thornburg , j. , 2008 , _ phys .
d , _ * 77 * , 104017 baker , j. g. , mcwilliams , s. t. , van meter , j. r. , centrella , j. m. choi , d .-
, kelly , b. j. & koppitz , m. , 2007 , _ phys.rev .
d _ * 75 * , 124024 .
filloux , ch .
, de freitas pacheco , j. a. , durier , f. & de araujo , j. c. n. , 2011 , _ ijmpd _ , * 20 * , 2399 .
filloux , ch .
, durier , f. , de freitas pacheco , j. a. & silk , j. , 2010 , _ ijmpd _ , * 19 * , 1233 .
echeverria , f. , 1989 , _
d _ * 40 * , 3194 . http://www.apc.univ-paris7.fr/face/en/content/elisango esteban , j. j. , garcia , a. f. , barke , s. , peinado , a. m. , cervantes , f. g. , bykov , i. , heinzel , g. , & k. danzmann , 2011 , _ opt .
express _ , * 19 * , 15937 .
robertson , d. i. , mcnamara , p. , ward , h. , & hough , j. , 1997 , _ class .
quantum grav . , _ * 14 * , 1575 . wand , v. , 2007 , _ interferometry at low frequencies : optical phase measurement for lisa and lisa pathfinder _ , ph .
d. thesis , university of hannover , unpublished .
shaddock , d. , ware , b. , halverson , p. , spero , r. e. , & klipstein , b. , 2006 , _ overview of the lisa phasemeter _ , international lisa symposium , eds . s.m .
merkowitz , & j.c .
livas , springer ( berlin , germnay , pp .
654 - 660 ) .
ke - xun , allen , g. , buchman , s. , debra , d. , & byer , r. , 2005 , _ class .
quantum grav . ,
_ * 22 * , s287 - s296 .
tinto , m. , & armstrong , j. w. , 1999 , _ phys .
d , _ * 59 * , 102003 .
maral , g. , & bousquet , m. , 2009 , _ satellite communications systems _ , fifth edition , john wiley & sons , ( chichester , england , u.k . ) kamel , a. a. , gelon , w. , & reckdahl , k. , 1999 . in proceedings of : _
aas / aiaa astrodynamics specialist conference _ ,
girdwood , alaska , august 16 - 18 , aas paper no .
aas 99 - 388 . tinto , m. , estabrook , f. b. , & j.a .
armstrong , 2002 , _ phys .
d , _ * 65 * , 082003 . | we analyze the sensitivities of a geostationary gravitational wave interferometer mission operating in the sub - hertz band . because of its smaller armlength ,
in the lower part of its accessible frequency band ( @xmath0 hz ) our proposed earth - orbiting detector will be less sensitive , by a factor of about seventy , than the laser interferometer space antenna ( lisa ) mission . in the higher part of its band instead ( @xmath1 hz )
, our proposed interferometer will have the capability of observing super - massive black holes ( smbhs ) with masses smaller than @xmath2 m@xmath3 . with good event rates for these systems
, a geostationary interferometer will be able to accurately probe the astrophysical scenarios that account for their formation .
gravitational waves , black holes physics , gravitational wave detectors |
the determination of the clustering , luminosity , star formation and other properties of galaxies , and the measurement of the evolution of those properties with redshift , are fundamentally important goals of observational cosmology .
the cnoc2 ( canadian network for observational cosmology ) field galaxy redshift survey is explicitly designed to study the evolution of galaxy clustering and luminosity properties over @xmath1 , and currently constitutes the largest galaxy sample at these intermediate redshifts ( see yee et al .
1997 for more details ) .
the survey has a total area of some 1.5 deg@xmath6 , in the form of 4 widely - separated patches on the sky , where each patch is an `` l''-shape that spans about 1.5@xmath7 .
the spectroscopic observations are carried out at the canada - france - hawaii telescope , using efficient mos multi - slit spectroscopy of typically 100 objects at once .
this is made possible with a band - limiting filter that restricts the spectra to 4300 - 6300 , resulting in a redshift range @xmath8 with unbiased coverage of important galaxy absorption and emission features .
redshift errors are about 75 km s@xmath9 in the rest frame .
the survey also includes 5-color @xmath0 photometry for nearly all fields ( plus @xmath10 for some ) , with average 5@xmath11 limits at ( kron - cousins ) @xmath12 and @xmath13 . at present ( 11/97 )
the cnoc2 survey is 80% complete , with 3 of the 4 patches essentially done .
the survey contains some 5000 redshifts , with nearly 4000 for galaxies brighter than the nominal @xmath14 spectroscopic completeness limit , at which the cumulative redshift sampling rate is about 1 in 2 .
figure [ fighist ] shows the redshift histogram for the survey , and figure [ figcol ] plots observed @xmath3 color vs. redshift . in the present paper
we discuss some _ preliminary _ results on the evolution of the galaxy luminosity function , using a subset of about 2000 @xmath2 galaxies from the `` 0223 '' and `` 0920 '' patches , which are the 2 current patches with almost fully reduced redshift and 5-color data .
we adopt @xmath15 km s@xmath9 mpc@xmath9 and @xmath16 throughout , as is typically done in the analysis of other redshift surveys at similar @xmath17 .
the luminosity function ( lf ) of galaxies is a fundamentally important quantity in the study of galaxy populations and their evolution .
accurate determinations of the lf at both low and high redshifts are crucial .
large wide - angle redshift surveys are providing precise measurements of the luminosity function in the local @xmath18 universe ( e.g. loveday et al .
1992 ; marzke et al .
1994b ; lin et al .
1996 ) , and smaller but deeper surveys have begun to measure the lf up to redshifts @xmath19 ( e.g. , lilly et al .
1995 ; ellis et al . 1996 ; cowie et al . 1996
; lin et al .
1997 ) , revealing clearly the rapid evolution of the lf of blue , star - forming galaxies , but relatively little change in the lf of red , more quiescent objects . here
we will examine the luminosity function and its evolution for galaxies of different types in the cnoc2 sample , and we will exploit the large cnoc2 sample size to derive improved constraints .
figure [ figlfs ] shows the @xmath20-band luminosity function for a sample of 2075 galaxies , with @xmath2 and @xmath8 , from the cnoc2 0223 and 0920 patches .
the lf is computed using standard inhomogeneity - independent maximum - likelihood techniques ( e.g. , efstathiou , ellis , & peterson 1988 ) .
the cnoc2 lf appears to be well fit by a nearly - flat schechter function ( @xmath21 ) , and it is also consistent with the @xmath22 lf from the canada - france redshift survey ( cfrs ; lilly et al .
1995 ) , as seen in the comparison of 2@xmath11 @xmath5-@xmath4 error ellipses in the figure [ figlfs ] inset .
note in particular the much smaller lf uncertainties of the cnoc2 sample , which contains about ten times as many galaxies as the cfrs over the same redshift range .
however , as has been observed in many previous galaxy samples , the luminosity function does appear to depend on various galaxy properties , such as rest - frame color , spectral type , or morphology ( see e.g. , lilly et al .
1995 ; lin et al . 1996 , 1997 ;
heyl et al .
1997 ; marzke et al .
1994a ) . to check the type dependence of the lf , we have also divided cnoc2 galaxies into early , intermediate , and late types ( more precisely red , intermediate , and blue ) , based on @xmath23-corrected @xmath3 colors computed from non - evolving coleman , wu , & weedman ( 1980 ; cww ) spectral energy distributions
( sed s ; figure [ figcol ] ) .
figure [ figlfrb ] then clearly shows that the luminosity function depends on galaxy type , with a strong steepening of the faint - end slope @xmath4 as one proceeds from red , early - type galaxies to blue , late - type ones .
these type - dependent differences are also quite significant , as shown in the figure [ figlfrb ] inset , and are also similar to those seen in previous smaller surveys at moderate redshifts , including the cfrs ( lilly et al .
1995 ) and autofib ( ellis et al .
1996 ) samples .
so far the lf s have been fit without regard to evolution , but we know from previous surveys that galaxies do evolve significantly over @xmath24 . as a simple but useful parameterization of luminosity and number density evolution , we adopt the following model : @xmath25 where @xmath26 is the galaxy number density , @xmath27 quantifies the rate of number density evolution , and @xmath28 describes the rate of @xmath5 or luminosity evolution . @xmath27 and @xmath28 are essentially just the linear coefficents in an expansion in @xmath17 of @xmath26 and @xmath5 , respectively . for simplicity ,
@xmath4 is assumed not to evolve so that the lf shape does not change with redshift .
we use maximum - likelihood techniques to derive best - fitting lf and evolution parameters , and figure [ figdep ] shows the resulting 1@xmath11 @xmath27 vs. @xmath28 error ellipses for the three cnoc2 galaxy types , which do indeed show different evolutionary trends . in particular , late - type galaxies show positive density evolution ( @xmath29 so @xmath26 increases at higher @xmath17 ) , early - type galaxies show negative density evolution ( @xmath30 ) , and intermediate types show no density evolution . on the other hand ,
all three galaxy types show positive luminosity evolution , i.e. @xmath5 is brighter in the past , and the derived values of @xmath31 are consistent with those expected from passively evolving galaxy models ( e.g. , bruzual & charlot 1993 ) .
( note that using @xmath32 instead of @xmath16 will roughly decrease @xmath27 by 0.8 and increase @xmath28 by 0.4 . ) also , in our parameterization , the luminosity density varies as @xmath33 , so @xmath34 characterizes the rate of evolution of the luminosity density .
figure [ figld ] plots @xmath35 vs. @xmath17 for the three galaxy types and shows the rapid rise in the luminosity density of late - type galaxies relative to that of early- and intermediate - type galaxies , again consistent with similar trends seen in previous smaller surveys ( e.g. , lilly et al .
1996 ; lin et al .
finally , figure [ figmag ] compares the observed cnoc2 galaxy number counts with those predicted by the lf and evolution parameters fit from the spectroscopic data _
alone_. the observed and lf - predicted counts are in good agreement for all 5 cnoc2 bands , showing that our simple 3-population evolution model is a reasonable description , and that there are no conspicuous unaccounted spectroscopic selection effects
. note also from the @xmath36 counts that our lf evolution model remains valid beyond the adopted @xmath14 limit of the spectroscopic sample .
moreover , as our galaxy classifications are based only on observed @xmath3 colors , the good match to the @xmath37 , @xmath38 , and @xmath39 counts gives us an independent check on the validity of our lf models .
however , an important caveat needs to be kept in mind , namely the sensitivity of the lf evolution results to the precise choice of sed s used to classify galaxies .
in particular , the present choice of non - evolving sed s obviously does not account for the color evolution of galaxies .
if one were to use more physically motivated _ evolving _ sed s , the resulting galaxy classifications will generally be different .
for example , one can eliminate the negative density evolution of early - type galaxies by using simple evolving models with exponentially declining star formation rates ( bruzual & charlot 1993 ) , wherein galaxies are bluer at higher redshifts ; this allows more galaxies to be classified as early - type at higher @xmath17 , and thus increases the value of @xmath27 .
there are obviously many possible combinations of evolving sed s that one may choose , by varying parameters such as star formation rate , age , metallicity , or dust content .
work is ongoing to test particular evolutionary scenarios , for example , to see whether simple pure luminosity evolution models ( no density evolution ) , with various choices of star formation history , age , etc .
, are consistent with both the cnoc2 lf and number count data .
( the availability of 5-color data for all cnoc2 galaxies will be of great help here , as it permits tight constraints on the sed of any particular galaxy . ) the lf evolution model outlined in the present paper should thus be considered as a first step , a _ description _ of galaxy evolution within the framework of non - evolving sed s , rather than an _ explanation _ of galaxy evolution in terms of more physically motivated evolving models
. we will also be able to constrain those physical models using additional tools .
note that there is much spectroscopic information available , which will permit measurements of star formation rates using emission line indices , as well as classification of galaxies via principal component analysis .
moreover , photometric redshifts can be accurately calibrated using the spectroscopic sample , and use of a larger photometric - redshift sample will help to constrain the lf at fainter magnitudes , as well as reduce errors on the @xmath27 and @xmath28 parameters .
measurements of the evolution of the surface brightnesses and morphologies of cnoc2 galaxies will also be possible .
finally , the environmental dependence of luminosity and star formation properties , and the relation to the evolution of galaxy clustering , will be explored using catalogs of groups and pairs of galaxies , and through measurements of the correlation function by galaxy type .
ultimately we hope to put clustering , luminosity , star formation , and spectral properties together into a coherent and self - consistent picture of galaxy evolution at moderate redshifts @xmath40 .
bruzual a. , g. , & charlot , s. 1993 , , 405 , 538 coleman , g. d. , wu , c .- c . , & weedman , d. w. 1980 , , 43 , 393 cowie , l. l. , songaila , a. , hu , e. m. , & cohen , j. g. 1996 , , 112 , 839 efstathiou , g. , ellis , r. s. , & peterson , b. a. 1988 , , 232 , 431 ellis , r. s. , colless , m. , broadhurst , t. , heyl , j. , & glazebrook , k. 1996 , , 280 , 235 heyl , j. , colless , m. , ellis , r. s. , & broadhurst , t. 1997 , , 285 , 613 lilly , s. j. , lefevre , o. , hammer , f. , & crampton , d. , 1996 , , 460 , l1 lilly , s. j. , tresse , l. , hammer , f. , crampton , d. , & lefevre , o. 1995 , , 455 , 108 lin , h. , kirshner , r. p. , shectman , s. a. , landy , s. d. , oemler , a. , tucker , d. l. , & schechter , p. l. 1996 , , 464 , 60 lin , h. , yee , h. k. c. , carlberg , r. g. , & ellingson , e. 1997 , , 475 , 494 loveday , j. , peterson , b. a. , efstathiou , g. , & maddox , s. j. 1992 , , 390 , 338 marzke , r. o. , geller , m. j. , huchra , j. p. , & corwin , h. g. 1994a , , 108 , 2 marzke , r. o. , huchra , j. p. , & geller , m. j. 1994b , , 428 , 43 yee , h. k. c. , et al .
1997 , to appear in the proceedings of iau joint discussion 11 _ redshift surveys in the 21st century _ , ed .
a. fairall , astro - ph/9710356 | the cnoc2 field galaxy redshift survey presently contains some 5000 galaxy redshifts , plus extensive @xmath0 photometry , and is the largest galaxy sample at moderate redshifts @xmath1 . here
we present some _ preliminary _ results on the galaxy luminosity function ( lf ) and its redshift evolution , using a sample of @xmath2 cnoc2 galaxies , subdivided into early , intermediate , and late types based on their @xmath3 colors relative to _ non - evolving _ galaxy models .
we find a significant steepening in the faint - end slope @xmath4 of the lf as one proceeds from early to late types .
also , for all galaxy types we find a rate of @xmath5 evolution consistent with that from passively evolving galaxy models .
finally , late - type galaxies show positive density evolution with redshift , in contrast to negative or no density evolution for earlier types . |
the libra toolkit is a collection of algorithms for learning and inference with probabilistic models in discrete domains .
what distinguishes libra from other toolkits is the types of methods and models it supports .
libra includes a number of algorithms for _ structure learning for tractable probabilistic models _ in which exact inference can be done efficiently .
such models include sum - product networks ( spn ) , mixtures of trees ( mt ) , and bayesian and markov networks with compact arithmetic circuits ( ac ) .
these learning algorithms are not available in any other open - source toolkit .
libra also supports _ structure learning for graphical models _ , such as bayesian networks ( bn ) , markov networks ( mn ) , and dependency networks ( dn ) , in which inference is not necessarily tractable . some of these methods are unique to libra as well , such as using dependency networks to learn markov networks .
libra provides a variety of exact and approximate inference algorithms for answering probabilistic queries in learned or manually specified models .
many of these are designed to exploit local structure , such as conjunctive feature functions or tree - structured conditional probability distributions .
the overall goal of libra is to make these methods available to researchers , practitioners , and students for use in experiments , applications , and education .
each algorithm in libra is implemented in a command - line program suitable for interactive use or scripting , with consistent options and file formats throughout the toolkit .
libra also supports the development of new algorithms through modular code organization , including shared libraries for different representations and file formats .
libra is available under a modified ( 2-clause ) bsd license , which allows modification and reuse in both academia and industry .
libra includes a variety of learning and inference algorithms , many of which are not available in any other open - source toolkit .
see table [ tab : functionality ] for a brief overview .
@xmath0 & bn structure with tree cpds & + & dn structure with tree / boosted tree / lr cpds & + @xmath1 & mn structure from dns & + & mn parameters ( pseudo - likelihood ) & + + @xmath1 & tractable bn / ac structure & + @xmath1 & tractable mn / ac structure & + @xmath1 & mixture of trees ( mt ) & + @xmath1 & spn structure ( id - spn algorithm ) & + & chow - liu algorithm & + @xmath1 & ac parameters ( maximum likelihood ) & + + & gibbs sampling ( bn , mn,@xmath1dn ) & ( dn ) + & mean field ( bn , mn,@xmath1dn ) & ( dn ) + & loopy belief propagation ( bn , mn ) & + & max - product ( bn , mn ) & + & iterated conditional modes ( bn , mn,@xmath1dn ) & + @xmath1 & variational optimization of acs & + + @xmath0 & ac variable elimination ( bn , mn ) & + @xmath0 & marginal and map inference ( ac , spn , mt ) & + libra s command - line syntax is designed to be simple . for example , to learn a tractable bn , run the command : `` libra acbn -i train.data -mo model.bn -o model.ac '' where train.data is the input data , model.bn is the filename for saving the learned bn , and model.ac is the filename for the corresponding ac representation , which allows for efficient , exact inference . to compute exact conditional marginals in the learned model : `` libra acquery -m model.ac -ev test.ev -marg '' . to compute approximate marginals in the bn with loopy belief propagation : `` libra bp -m model.bn -ev test.ev '' .
additional command - line parameters can be used to specify other options , such as the priors and heuristics used by acbn or the maximum number of iterations for bp .
these are just three of more than twenty commands included in libra .
libra supports a variety of file formats . for data instances
, libra uses comma separated values , where each value is a zero - based index indicating the discrete value of the corresponding variable . for evidence and query files , unknown or missing values
are represented with the special value `` ` * ` '' . for model files
, libra supports the xmod representation from the winmine toolkit , the bayesian interchange format ( bif ) , and the simple representation from the uai inference competition .
libra converts among these different formats using the provided mconvert utility , as well as to its own internal formats for bns , mns , and dns ( .bn , .mn , .dn ) .
libra has additional representations for acs and spns ( .ac , .spn ) .
these formats are designed to be easy for humans to read and programs to parse .
libra is implemented in ocaml .
ocaml is a statically typed language that supports functional and imperative programming styles , compiles to native machine code on multiple platforms , and uses type inference and garbage collection to reduce programmer errors and effort .
ocaml has a good foreign function interface , which libra uses for linking to c libraries and implementing some memory - intensive subroutines more efficiently in c. the code to libra includes nine support libraries , which provide modules for input , output , and representation of different types of models , as well as commonly used algorithms and utility methods .
in table [ tab : comparison ] , we compare libra to other toolkits in terms of representation , learning , and inference . .comparison of libra to several other probabilistic inference and learning toolkits . [ cols= "
< , < , < , < , < , < , < " , ] in terms of representation , libra is the only open - source software package that supports acs and one of a very small number that support dns or spns .
libra does not currently support dynamic bayesian networks ( dbn ) or influence diagrams ( i d ) . for factors
, libra supports tables , trees , and arbitrary conjunctive feature functions .
bnt @xcite and openmarkov ( cisiad , 2013 ) also support additional types of cpds , such as logistic regression , noisy - or , neural networks , and algebraic decision diagrams , but they only support tabular cpds for structure learning . opengm2 @xcite supports sparse factors , but iterates through all factor states during inference .
libra is unique in its ability to learn models with local structure and exploit that structure in inference . for exact inference , the most common algorithms are junction tree ( jt ) , enumeration ( e ) , and variable elimination ( ve ) .
libra provides acve , which is similar to building a junction tree , but it can exploit structured factors to run inference in many high - treewidth models . for approximate inference , libra provides gibbs sampling ( g ) , loopy belief propagation ( bp ) , and mean field ( mf ) , all of which are optimized for structured factors .
a few learning toolkits offer likelihood weighting ( lw ) or additional sampling algorithms for bns .
fastinf , libdai @xcite , and opengm2 offer the most algorithms but only support tables . for learning ,
libra supports maximum likelihood ( ml ) parameter learning for bns , acs , and spns , and pseudo - likelihood ( pl ) optimization for mns and dns .
libra does not yet support expectation maximization ( em ) for learning with missing values .
structure learning is one of libra s greatest strengths .
most toolkits only provide algorithms for learning bns with tabular cpds or mns using the pc algorithm .
libra includes methods for learning bns , mns , dns , spns , and acs , and all of its algorithms support learning with local structure . in experiments on grid - structured mns ,
libra s implementations of bp and gibbs sampling were at least as fast as libdai , a popular c++ implementation of many inference algorithms .
the accuracy of both toolkits was equivalent .
parameter settings , such as the number of iterations , were identical .
see figure [ fig : libdai ] for more details .
the libra toolkit provides algorithms for learning and inference in a variety of probabilistic models , including bns , mns , dns , spns , and acs .
many of these algorithms are not available in any other open - source software .
libra s greatest strength is its support for tractable probabilistic models , for which very little other software exists .
libra makes it easy to use these state - of - the - art methods in experiments and applications , which we hope will accelerate the development and deployment of probabilistic methods . | the libra toolkit is a collection of algorithms for learning and inference with discrete probabilistic models , including bayesian networks , markov networks , dependency networks , and sum - product networks .
compared to other toolkits , libra places a greater emphasis on learning the structure of tractable models in which exact inference is efficient .
it also includes a variety of algorithms for learning graphical models in which inference is potentially intractable , and for performing exact and approximate inference .
libra is released under a 2-clause bsd license to encourage broad use in academia and industry .
probabilistic graphical models , structure learning , inference |
the study of large deviation functions has been of increasing importance for the understanding of many - body systems . on the one hand
large deviation functions form the basis of a modern approach to equilibrium statistical mechanics @xcite , on the other hand they are increasingly recognized of being of fundamental interest for the characterization of non - equilibrium systems where they are sometimes considered to play a role similar to that played by the free energy at equilibrium .
important examples are given by the large deviations of the steady - state currents @xcite in systems that are far from equilibrium .
recently , mehl _ et al .
_ @xcite extended the study of large deviation functions far from equilibrium to the steady - state entropy production . studying a one - dimensional system composed of a single particle driven along a periodic potential , they reformulated the problem as a time - independent eigenvalue problem @xcite and showed that for this system the large deviation function of the entropy production exhibits a kink at zero entropy production .
similar kinks in the large deviation function of the entropy production or in related large deviation functions of other quantities can also be found in a range of other systems @xcite .
these are intriguing results that raise the important question whether the presence of this kink is a universal feature of systems with non - equilibrium steady states . in this paper
we study the entropy production in the steady states of different non - equilibrium interacting many - body systems .
on the one hand we study the partially asymmetric simple exclusion process ( pasep ) @xcite , a transport process in an open system where particles can enter or leave only at the system boundaries .
on the other hand we investigate reaction - diffusion systems on a ring where the number of particles can change everywhere in the system due to reactions between the particles .
our main emphasis thereby is on the entropy production in the long - time limit where we study the large deviation function in the same way as done by mehl _
_ in their study of the single particle system
. for all studied many - body systems we find a kink - like feature at zero entropy production , similar to what has been observed for the single particle system .
we show that this kink is a generic feature in non - equilibrium systems obeying a detailed fluctuation theorem for the entropy production .
it is worth noting that large deviation functions of the _ current _ have been studied previously in some related systems .
studies of the totally asymmetric simple exclusion process ( tasep ) @xcite and of the symmetric simple exclusion process ( ssep ) @xcite revealed a non - analytical behavior of the current large deviation function at the value 0 of the parameter that is canonically conjugated to the particle current . using dynamical renormalization group techniques
, it was later shown that the value of the corresponding exponent follows from the noise renormalization @xcite . very recently ,
bodineau and lagouge @xcite studied the current large deviations in a driven dissipative lattice gas model with creation and annihilation processes , whereas simon @xcite investigated the large deviation function of the current for the weakly asymmetric simple exclusion process on a ring .
our paper is organized in the following way . in the next section
we introduce our models , before discussing in section iii our approaches for computing the entropy production at short times as well as the mean entropy production rate and the large deviation function for the entropy production in the long - time limit .
our results are presented and discussed in section iv .
finally , section v gives our conclusions .
in order to elucidate the entropy production in interacting many - body systems , we discuss in the following reaction - diffusion systems on a ring as well as open transport systems where particles move through a system that they can enter or leave only at its boundaries . in the past , due to the combination of their conceptual simplicity and highly non - trivial results , diffusion - limited reaction systems @xcite and simple exclusion processes @xcite have greatly contributed to our understanding of processes far from equilibrium .
for the same reasons , these models are also the natural choices for our study of the entropy production in many - body systems . as reaction - diffusion systems we consider simple cases where particles @xmath0 diffuse on a one - dimensional ring , under the condition that every lattice site can only be simply occupied .
this diffusion process is mimicked by the hopping of a particle to an empty nearest neighbor site with rate @xmath1 . in addition , we also allow for particle creation and annihilation . in the creation process
a new particle can be created at an empty site with rate @xmath2 , whereas in the annihilation process @xmath3 particles on @xmath3 connected sites are destroyed with rate @xmath4 , yielding @xmath3 empty sites . in the following
we characterize our models by the number of particles involved in the annihilation process and call @xmath5 ( with @xmath6 ) the model where @xmath3 particles are destroyed . as a variant , we also study the situation ( we call the resulting model @xmath7 ) where in a two particles reaction only one particle is destroyed .
these models are the same as those studied in @xcite in order to better understand steady - state and transient properties of reaction - diffusion systems .
all these systems have non - equilibrium steady states . in addition ,
some or all reactions do not allow for a direct back - reaction , and microscopic reversibility is broken .
however , as discussed in the next section , the computation of the entropy production requires microscopic reversibility , i.e. for every reaction the direct back - reaction must be possible .
for that reason we are adding to the reaction schemes the direct back - reactions of the reactions just described .
thus , any particle can be destroyed with rate @xmath8 , with @xmath9 , irrespective of whether neighboring sites are occupied or not , and a reversing of the annihilation process yields the creation of particles with rate @xmath10 , where @xmath11 .
it is important to note that even with these expanded reaction schemes our systems are still characterized by non - equilibrium steady states , as long as @xmath12 . in the following we set @xmath13 and focus on the case @xmath14 . for the convenience of the reader
we summarize the different reaction schemes in table [ table1 ] .
.the different reaction schemes discussed in this work .
the back - reactions are taking place with rates @xmath15 and @xmath16 , with @xmath9 and @xmath11 . [ cols="^,^",options="header " , ] as transport process we consider the partially asymmetric simple exclusion process ( pasep ) where every configuration is reversible . in the total asymmetric simple exclusion process ( tasep ) in an open one - dimensional system , particles that are fed into the system at , say , the left end with rate @xmath17 can leave the system at the right end with rate @xmath18 . inside the system particles can only jump to a right neighboring site , provided that site is not occupied .
obviously , microscopic reversibility is always broken for this model . in the pasep ,
however , particles can always jump in both directions , provided that the chosen site is empty , with the probability for a jump to the right being @xmath19 , whereas the probability for a jump to the left is @xmath20 .
in addition , particles can exit the system at the left with probability @xmath21 ( with @xmath22 ) and enter at the right with probability @xmath23 ( with @xmath24 ) .
it follows that microscopic reversibility is always fulfilled . in order to keep the number of parameters as small as possible , we set @xmath25 , with @xmath14 .
whereas our main focus in the following will be on the large deviation function of the steady - state entropy production in the long time limit , we will also briefly discuss the entropy production at short times .
focusing on our lattice models , let us consider a path in configuration space @xmath26 that starts at some configuration @xmath27 and ends at some configuration @xmath28 after @xmath29 diffusion or reaction steps .
every configuration @xmath30 is uniquely characterized by the occupation numbers ( being 0 or 1 ) of all lattice sites .
we will denote by @xmath31 the probability to find the configuration @xmath30 in the steady state . with
every step @xmath32 leading from configuration @xmath33 to configuration @xmath30 we associate a time increment @xmath34 given by @xmath35 where @xmath36 is the rate with which we go from configuration @xmath33 to any other accessible configuration @xmath37 .
the sum in the denominator is thereby a sum over all configurations @xmath37 ( including the configuration @xmath30 in which the system will be at step @xmath32 ) that can be reached from the configuration @xmath33 through diffusion or reaction . with
that we can assign a total time @xmath38 to our trajectory in configuration space . along the same trajectory
the total entropy production is given by @xcite @xmath39 where @xmath40 is the entropy produced in the particle bath connected to our system .
the boundary term , @xmath41 , which can be neglected in the long time limit , needs to be included when investigating fluctuation relations of the entropy production at short times .
it is worth pointing out that in eq .
( [ eq : ent_prod ] ) the transition rates from configuration @xmath33 to @xmath30 and from configuration @xmath30 to @xmath33 are showing up together .
this is the reason why we have added the back - reactions to our reaction schemes . in section
iv.a we will discuss the mean entropy production in the short time regime as a function of time @xmath42 . for very small @xmath42 , i.e. values of @xmath42 where only few configurations are visited , this can be done in an exact way , where , starting from all possible initial states , all possible trajectories are determined and the corresponding entropy changes are recorded @xcite .
this allows us to determine in a numerically exact way the probability distribution of the total entropy production , @xmath43 , after time @xmath42 . for larger ( but still small ) values of @xmath42
, we can compute the mean entropy production through standard monte carlo sampling .
all the data discussed in the following are numerically exact data , with the exception of the data shown in fig .
[ fig2 ] which have been obtained by monte carlo simulations . with the definition ( [ eq : ent_prod ] ) of the total entropy production
the probability distribution @xmath43 obeys a detailed fluctuation theorem for any time @xmath42 : @xmath44 initially @xcite this relation was shown to be valid in the long - time limit when only @xmath40 is considered . if one adds the boundary term , as done in eq .
( [ eq : ent_prod ] ) , then the theorem ( [ dft ] ) holds true for any finite time interval @xmath42 @xcite . in the following much emphasis
will be put on the large deviation function of the entropy production ( also called the rate function ) @xmath45\ ] ] which describes the asymptotic large fluctuations of the entropy production ( in that limit we can neglect the boundary term in ( [ eq : ent_prod ] ) ) : @xmath46 with the normalized entropy production rate @xmath47 here @xmath48 is the mean entropy production rate . in order to compute the large deviation function , we proceed as in ref .
@xcite and consider the generating function @xmath49 \right > = \int ds_m \exp\left [ -\mu s_m(\tau ) \right ] p(s_m,\tau)\ ] ] of @xmath50 .
the time evolution of the generating function is given by the linear operator @xmath51 \nonumber \\ & = & - \left [ \sum\limits_j \omega(c_j \longrightarrow c_i)^{1-\mu } \omega(c_i \longrightarrow c_j)^{\mu } - r(c_i ) \right ] \nonumber \\\end{aligned}\ ] ] where @xmath52 .
the spectrum of @xmath53 is strictly positive for @xmath54 , whereas for @xmath55 the lowest eigenvalue is zero and characterizes the steady state of the system . in the long time
limit the lowest eigenvalue @xmath56 can equivalently be determined as @xmath57 \right > \right ] = \nu(\mu)~.\ ] ] for our finite systems we determine the eigenvalue @xmath56 through the numerically obtained spectrum of the operator @xmath58 which is expressed by a finite matrix .
this is done using standard algorithms .
this lowest eigenvalue allows us to obtain expressions for various quantities of interest .
thus the mean entropy production rate is given by @xmath59 whereas the large deviation function is given by the legendre transform @xmath60 the fluctuation theorem @xcite is revealed through a symmetry relation for the large deviation function , @xmath61 or , on the level of the eigenvalue , through the equivalent symmetry @xmath62
before discussing the long - time behavior of the entropy production , which is the main focus of the paper , we briefly discuss some properties of the entropy production in the short time limit . as an example , we discuss the reaction - diffusion models , but corresponding results are obtained for the transport processes . for all studied cases
the probability distribution @xmath63 for the total entropy production at short times is very irregular , as shown in fig .
[ fig1]a for the model @xmath64 .
similar irregular shapes , dominated by prominent peaks , also show up when computing a related quantity , the so - called driving entropy production , in systems that are driven out of stationarity through a change of system parameters @xcite .
the details of the probability distributions depend on the microscopic details of the studied systems , especially on the values of the diffusion and reaction rates . still ,
even though the probability distributions are very irregular and specific for the studied systems , we recover in all cases the celebrated fluctuation theorem ( [ dft ] ) , see fig .
[ fig1]b , provided that the boundary terms are included in the definition of @xmath65 .
= 4.50 in with the probability distribution of the total entropy production at hand , we can straightforwardly compute the mean value of entropy change @xmath66 during some time interval @xmath42 . as shown for some cases in fig .
[ fig2]a , the total entropy production rapidly yields a linear increase with time .
we also note , see fig .
[ fig2]b , that the entropy change per site already becomes independent of the system size for very small systems . as a consequence of these two observations we can restrict ourselves to small systems when discussing the long - time properties in the following .
= 4.50 in as discussed in section iii the long - time properties of the entropy production are encoded in the eigenvalue @xmath56 of the linear operator @xmath53 . after having determined @xmath56 we can use that quantity in order to discuss the mean entropy production rate , see eq .
( [ eq : mepr ] ) , as well as the large deviation function for the entropy production , see eq .
( [ eq : ldf ] ) . in fig .
[ fig3 ] we show examples of changes of the mean entropy production rate @xmath67 for various systems when varying one of the system parameters . fig .
[ fig3]a illustrates some of our results for the reaction - diffusion systems where the creation rate @xmath2 is changed and all the other system parameters are kept constant .
[ fig3]b is devoted to the transport process with a varying entrance rate @xmath17 .
interestingly , @xmath67 encodes important information on the physical properties of the different steady states . focusing first on the reaction - diffusion systems
, we see that for model @xmath68 the mean entropy production rate has a maximum around @xmath69 and decreases towards zero for larger values of the creation rate . in stark contrast to this
@xmath67 increases monotonously for the other two models @xmath70 and @xmath64 . in order to understand this difference in behavior
, we refer the reader to the reaction schemes listed in table [ table1 ] . noting that for very large values of @xmath2 there is an increased probability for a free site to be surrounded by occupied neighboring sites
, it follows that the creation process @xmath71 is then effectively identical to the process @xmath72 .
this is , however , exactly the reversed reaction to the annihilation process of model @xmath68 , such that this system approaches an equilibrium system for very large values of @xmath2 .
consequently , the mean entropy production rate vanishes in that limit as no entropy production takes place in equilibrium .
this is different for models @xmath70 and @xmath64 , as here the reaction @xmath72 is not the direct back - reaction of the creation process , and the systems remain out of equilibrium for any value of @xmath2 .
similar conclusions are also drawn when varying other system parameters .
for example , when @xmath73 , all reactions become symmetric , and the system approaches equilibrium .
concomitantly , the mean entropy production rate approaches zero in that limit .
it is worth noting that the behavior of the mean entropy production rate is similar to that of a quantity derived from the stationary probability currents @xcite that has been proposed as a possible measure for the non - equilibrium character of a system . for example , both quantities show a non - monotonous change for model @xmath68 when increasing @xmath2 , whereas their increase is monotonous for models @xmath70 and @xmath64 ( see fig . 2 in @xcite ) .
= 4.50 in non - trivial information is also contained in the mean entropy production rate of the transport process . as an example
we show in fig .
[ fig3]b @xmath67 as a function of the entrance rate @xmath17 for a strongly asymmetric case with @xmath74 .
the striking feature in this figure is the appearance of kinks in the curves obtained for smaller exit rates , as for example @xmath75 or @xmath76 .
these kinks , which are located at @xmath77 , yield a sharp peak in the negative of the second derivative of @xmath67 with respect to @xmath17 , see inset of fig .
we checked that these kinks show up for all @xmath78 at the value @xmath77 . for larger values of @xmath18
the kink vanishes and only a small peak remains in the second derivative ( the peak still exists for @xmath79 but is not visible on the scale of the inset in fig . [ fig3]b ) .
interestingly , in this regime the position of the peak is independent of the value of @xmath18 and is given by @xmath80 . in order to understand this result
we need to remind the reader of the phase diagram of the pasep @xcite . for our choice of entrance and exit rates at the boundaries of the system
the phase diagram of the pasep is very similar to that of the tasep and contains a low density phase , a high density phase and a maximum current phase .
the phase boundary between the low and the high density phases , which is discontinuous and characterized by the coexistence of these two phases , is given by @xmath77 , @xmath81 , where @xmath82 depends on the specific values of the different rates .
in addition , one has that the boundary between the low density ( high density ) phase and the maximum current phase is given by @xmath83 and @xmath84 ( @xmath85 and @xmath86 ) .
insertion of the values of the rates used in fig .
[ fig3]b into the exact expressions for the phase boundaries @xcite yields the value @xmath87 . comparing this with the position of the kink in the mean entropy production rate for @xmath78 , we see that the kink coincides with the discontinuous transition separating the low from the high density phase .
in addition , for @xmath84 the extremum of the second derivative of @xmath67 close to @xmath88 indicates the transition separating the low density phase from the maximum current phase . in order to compute the large deviation function for the entropy production ,
we proceed as discussed in section iii and determine the lowest eigenvalue @xmath56 of the linear operator @xmath53 , see eq .
( [ eq : lmu ] ) . the large deviation function
is then obtained through the legendre transform ( [ eq : ldf ] ) .
typical results obtained for our systems are shown in fig .
[ fig4 ] where the left column is devoted to the reaction - diffusion systems and the right one to the pasep . as expected , the eigenvalue @xmath56 fulfills the relation ( [ eq : nu_ft ] ) that follows directly from the fluctuation theorem , see fig .
[ fig4]a and fig .
[ fig4]d . by construction ,
the large deviation function @xmath89 is zero at @xmath90 .
in addition , plots of @xmath89 , see fig . [ fig4]b and fig .
[ fig4]e , reveal for all studied situations a kink at zero entropy production @xmath91 .
this kink is best revealed when studying the derivatives of @xmath92 with respect to @xmath93 , as shown in fig .
[ fig4]c and fig .
[ fig4]e , yielding a step at @xmath91 in the first derivative and a maximum at @xmath91 in the second derivative .
= 4.50 in as stressed in the introduction , a kink has also been observed recently in a study of the steady - state entropy production for a single particle driven along a periodic potential @xcite .
the fact that a similar behavior is observed in our interacting many - body systems suggests that this feature is a universal one . before discussing the origin of the kink further ,
let us briefly pause and discuss how the large deviation function and its derivative depend on the parameters of the system . as an example , we show in fig .
[ fig5 ] @xmath92 and @xmath94 for the reaction - diffusion model @xmath64 for various values of the creation field @xmath2 . in our discussion of the mean entropy production rate we pointed out that for this system @xmath67 increases when @xmath2 increases , resulting in an enhanced non - equilibrium character when taking @xmath67 as a measure .
interestingly , the kink in the large deviation function also gets more pronounced when @xmath2 increases , as witnessed by a larger amplitude of the step in the derivative @xmath94 .
this is the generally observed behavior for all our systems : when the system parameters are varied in such a way that the mean entropy production rate increases ( decreases ) , then the step height at @xmath95 of the derivative of the large deviation function also increases ( decreases ) .
= 4.50 in in fact , the kink at @xmath91 in the large deviation function for the entropy production is a generic feature and immediately follows from the fluctuation theorem @xcite . to see this , we remind the reader that the fluctuation theorem shows up on the level of the large deviation function through the symmetry relation ( [ eq : chi_ft ] ) . taking the derivative of that relation with respect to @xmath93 yields the following relations between the derivatives at both sides of @xmath91 : @xmath96 valid for _ any _ value of the entropy production @xmath97 . in fig .
[ fig6 ] we illustrate for model @xmath64 the relation ( [ eq : chi_ft ] ) for the large deviation function as well as the relations ( [ eq : chip_ft1 ] ) and ( [ eq : chip_ft2 ] ) for its derivative . = 4.50 in the relations ( [ eq : chip_ft1 ] ) and ( [ eq : chip_ft2 ] ) completely characterize the kink .
in fact , one really should describe the observed feature as kink - like , as the derivative of @xmath92 is still continuous at @xmath91 even though its value changes rather dramatically when the entropy production approaches zero .
( [ eq : chip_ft1 ] ) and ( [ eq : chip_ft2 ] ) make obvious the connection between the kink and the mean entropy production rate .
it is the value of @xmath67 that characterizes this intriguing behavior of the large deviation function of the entropy production .
a larger value of @xmath67 yields a more pronounced feature and a sharper and higher peak in the second derivative of the large deviation function .
one might object that in some known cases the distribution of the entropy production has a nearly gaussian form and that for these cases a kink - like behavior is not observed .
an example can be found in ref .
@xcite ( see their figures 1 and 3 ) , where the large deviation function for a particle driven along a periodic potential is nicely fitted to a parabola @xmath98 for small forces . in a similar way , the largest eigenvalue seems to be well described by @xmath99 these results seem to indicate that close to equilibrium systems with a nearly gaussian entropy production can fulfill the fluctuation theorem without exhibiting a kink - like feature in the large deviation function .
= 4.50 in our systems allow us to address this point in a systematic way .
in fact , we always find for systems that are close to equilibrium ( as characterized by a small value of the mean entropy production rate ) that the entropy production approaches a nearly gaussian form and that the kink seems to vanish .
this is illustrated in fig .
[ fig7 ] for the transport process with @xmath100 and various values of @xmath101 . in this case
the system approaches equilibrium when @xmath101 increases , yielding an equilibrium system for @xmath102 . as we see from fig .
[ fig7]a and [ fig7]b , the kink - like feature in the large deviation function gets less and less pronounced when increasing @xmath101 and seems to have vanished completely for @xmath103 .
indeed , in that case one can fit the largest eigenvalue and the large deviation function to the quadratic forms ( [ eq : fit2 ] ) and ( [ eq : fit1 ] ) , thereby obtaining an excellent agreement .
however , a closer look at the second derivative , see fig .
[ fig7]d and [ fig7]e , reveals that qualitatively the case @xmath103 does _ not _ differ from the other cases : the second derivative exhibits a maximum at @xmath91 which is the signature of the kink - like feature .
admittedly , this maximum is rather shallow for a system close to equilibrium , but it persists as long as the mean entropy production rate does not vanish . obviously , it is not good enough to simply fit the data to the quadratic forms ( [ eq : fit1 ] ) and ( [ eq : fit2 ] ) and to conclude from an optically good fit that the kink - like behavior is absent .
the kink - like feature might no longer be easily detectable by simple inspection of @xmath92 in systems where the entropy production approaches a nearly gaussian form , but it still reveals itself in a maximum in @xmath104 . from our study , as well as from the observation of kink - like features in other systems fulfilling the fluctuation theorem , we can therefore conclude that this behavior is generic for non - equilibrium systems .
our study of interacting many - body systems has yielded two main results regarding the entropy production in the steady states of non - equilibrium systems . on the one hand
, we showed that the kink observed in the entropy production large deviation function at zero entropy production is generic and directly follows from the fluctuation theorem .
the well known relation for the large deviation function ( [ eq : chi_ft ] ) implies certain relations between the derivatives on both sides of @xmath91 , see eq .
( [ eq : chip_ft1 ] ) and ( [ eq : chip_ft2 ] ) .
it follows that the kink is characterized by the value of the mean entropy production rate in the steady state . on the other hand ,
our results indicate that the mean entropy production rate is a good measure for quantifying the non - equilibrium character of a system ( or , equivalently , the degree of violation of detailed balance ) .
not only does it characterize the kink - like structure in the large deviation function , it also directly reveals whether a system gets closer to be an equilibrium system when changing some system parameters , as illustrated in fig .
in addition , the mean entropy production rate also reveals the presence of dynamic phase transitions , as seen in our study of the pasep , see fig .
[ fig3]b .
recently @xcite it was proposed that the non - equilibrium character of a system can be captured quantitatively by some global observables derived from the probability currents .
probability currents and entropy production are of course intimately related . they are unique to a non - equilibrium system and are absent in equilibrium systems .
they also show a remarkably similar dependence on system parameters . as an example
, the reader should compare our fig .
[ fig3]a with fig . 2 in @xcite where these quantities are computed for the one - dimensional reaction - diffusion systems discussed in this manuscript . to our knowledge
, however , it has not yet been studied whether the observable derived from the stationary probability currents allows to probe the system for the presence of a phase transition .
a more systematic comparison of the behavior of that quantity with that of the mean entropy production rate seems to be needed in order to further elucidate the commonalities and differences between these two quantities . as a final remark we note that in the past quite some effort has been devoted to the computation and understanding of the current large deviation function in driven systems . in our study
we focused on the large deviation function of the entropy production and characterized its properties both for systems with ( pasep ) and without ( reaction - diffusion systems ) particle currents .
the exact relation between these two large deviation functions in driven systems remains to be better understood .
+ r. s. ellis , _ entropy , large deviations , and statistical mechanics _
( berlin , springer , 1985 ) .
a. dembo and o. zeitouni , _ large deviation techniques and applications _
( berlin , springer , 1998 ) . h. touchette , phys . rep . * 478 * , 1 ( 2009 ) .
l. bertini , a. de sole , d. gabrielli , g. jona - lasinio , and c. landim , j. stat .
p07014 ( 2007 ) .
b. derrida , j. stat .
p07023 ( 2007 ) .
j. mehl , t. speck , and u. seifert , phys .
e * 78 * , 011123 ( 2008 ) .
j. l. lebowitz and h. spohn , j. stat .
phys . * 95 * , 333 ( 1999 ) .
p. visco , j. stat .
p06006 ( 2006 ) .
k. turitsyn , m. chertkov , v. y. chernyak , and a. puliafito , phys .
* 98 * , 180603 ( 2007 ) .
b. cleuren , c. van den broeck , and r. kawai , c. r. physique * 8 * , 567 ( 2007 ) .
d. lacoste , a. w. c. lau , and k. mallick , phys .
e * 78 * , 011915 ( 2008 ) .
n. kumar , s. ramaswamy , and a. k. sood , arxiv:1009.5925 . in most of these papers
the authors did not discuss the kink explicitly .
however , in all cases the existence of the kink can either be derived from the exact expressions given for the large deviation functions or it can be inferred from the numerical data shown in various figures .
r. a. blythe and m. r. evans , j. phys .
a : math . theor . *
40 * , r333 ( 2007 ) .
b. derrida and j. l. lebowitz , phys .
* 80 * , 209 ( 1998 )
. b. derrida and c. appert , j. stat .
phys . * 94 * , 1 ( 1999 ) .
d. kim , phys .
e * 52 * , 3512 ( 1995 ) . v. lecomte , u. c. tuber , and f. van wijland , j. phys .
theor . * 40 * , 1447 ( 2007 ) .
t. bodineau and m. lagouge , j. stat . phys . *
139 * , 201 ( 2010 ) .
d. simon , arxiv:1011.3590 .
m. henkel , h. hinrichsen , and s. lbeck , _ non - equilibrium phase transitions : volume 1 : absorbing phase transitions _
( springer / dordrecht and canopus / bristol , 2008 ) .
g. dor , _ universality in nonequilibrium lattice systems : theoretical foundations _ ( world scientific / singapore , 2008 ) .
b. derrida , phys .
301 * , 65 ( 1998 ) .
g. m. schtz , in _ phase transitions and critical phenomena _ , edited by c. domb and j. lebowitz ( academic press , london , 2001 ) , vol .
o. golinelli and k. mallick , j. phys .
a : math . gen . * 39 * , 12679 ( 2006 ) .
s. dorosz and m. pleimling , phys .
e * 79 * , 030102(r ) ( 2009 ) .
s. dorosz and m. pleimling , phys .
e * 80 * , 061114 ( 2009 ) .
s. dorosz and m. pleimling , physics procedia * 4 * , 107 ( 2010 ) .
u. seifert , eur .
j. b * 64 * , 423 ( 2008 ) .
d. j. evans , e. g. d. cohen , and g. p. morriss , phys . rev .
* 71 * , 2401 ( 1993 ) .
g. gallavotti and e. g. d. cohen , phys .
* 74 * , 2694 ( 1995 ) .
j. kurchan , j. phys .
a * 31 * , 3719 ( 1998 ) .
u. seifert , phys .
95 * , 040602 ( 2005 ) .
m. esposito , u. harbola , and s. mukamel , phys .
e * 76 * , 031132 ( 2007 ) .
r. k. p. zia and b. schmittmann , j. phys .
a * 39 * , l409 ( 2006 ) .
r. k. p. zia and b. schmittmann , j. stat .
p07012 ( 2007 ) .
t. platini , arxiv:1007.0165 .
s. sandow , phys .
e * 50 * , 2660 ( 1994 ) . | entropy production is one of the most important characteristics of non - equilibrium steady states .
we study here the steady - state entropy production , both at short times as well as in the long - time limit , of two important classes of non - equilibrium systems : transport systems and reaction - diffusion systems .
the usefulness of the mean entropy production rate and of the large deviation function of the entropy production for characterizing non - equilibrium steady states of interacting many - body systems is discussed .
we show that the large deviation function displays a kink - like feature at zero entropy production that is similar to that observed for a single particle driven along a periodic potential .
this kink is a direct consequence of the detailed fluctuation theorem fulfilled by the probability distribution of the entropy production and is therefore a generic feature of the corresponding large deviation function . |
* background . *
the graph theory indeed presents relevance with abstract relation among objects .
since author wrote the bots algorithm for graph traversal at april , 2012 , author still studies this problem .
we find the fact that an instance can be computed by bots algorithm without concern the graph classes .
we guess that there is a logic model support this phenomenon , which leads to the approach possess the capacity of data - oriented .
the strategy of equivalent visiting is posed so that we can quantify the process of graph traversal with monotone decreasing function .
the uniform data structure of graph is set up with the method of partition of set , and this core idea may be used to solve other problems of graph , such as graph partition and graph coloring .
it makes these problems may present a quantified model for the abstract relation . and this abstract logic model can guarantee algorithms possess much more general and stronger .
+ * related * work .
we formally state the binary relation on graph
. this relation contribution can be underlying basis for this model .
we repeatedly abstract the basic relation for our new logical mode , and obtain new classes or new properties of relation in new model .
these algorithms we given always may have no associated with weights to each edge or arc .
it makes the new relation is easy to present a practical instance . the theoretical proof and computing of relation
are transformed to algebra of sets .
furthermore , these algorithms present more reliable , intuitive , simple and high precision , although they are heuristic approaches .
first , we introduce bost and obots algorithms , which runtime complexity both are @xmath0
. they can exactly compute any connected graph classes , including difficult mixed graph .
all connected instances may be explored by these algorithms without recursive method as _
_ dynamic programming__@xcite , such that greatly reduce the complexity of program .
we can really and easily achieve the aim of parallel and distributed computing for graph traversal .
graph partition may be independent of weight not like kernighan lin algorithm@xcite , although its runs on time complexity @xmath1 . in this thesis ,
the graph partition actually is a method of cutting graph .
you can arbitrarily choose the nodes on instance for your research of ai , network flow , graph color , physical problems and etc .
it makes the abstract relation among nodes be partition on a sequence of domains for your model of problem .
graph coloring is not a simple labeling each vertex on instance .
it becomes a logical problem for how to cut graph and let those vertices be partitioned to two classes .
author gives two speed - up algorithms bogpc and boerc , which can run in @xmath2 and @xmath3 respectively .
we prove that their precision can be less than and equal to a constant on an instance .
+ * overview .
* author will follow this format : defining objects , exploring the features of objects , proving algorithm , giving pseudocode , computing runtime complexity of approach and finally present exp .
on instance . in these process , we give the discussion or summary to express author s viewpoint with problem .
then this paper is organized as follows .
firstly preliminary knowledge is in section 2 .
section 3 introduces basic definition , properties , method and proofs for graph , including pseudocode of bots and obots , experiments . at the end
give the solving problem of natural number bocps .
the definition and method of graph partition are stated in section 4 . similarly , there are pseudocode , algorithmic complexity and experiment .
section 5 proposes definition of edge and model of graph coloring .
finally we give the formula of graph coloring . of cause
, there show the algorithms bogpc and boerc with concerning complexity .
we will evaluate those algorithms about probabilistic of exploring minimum chromatic value .
the paper is concluded by a summary , a conjecture of _ russell paradox _ and future work in section 6 .
in this paper , we are interested in the connected graph . for each vertex @xmath4 on instance
, there can be at least a path between @xmath4 and the others .
we set each vertex can be labeled with number .
let @xmath5 be a collection of vertices having @xmath6 .
we reserve the letter _
n _ and the term @xmath7 for the number of vertices on an instance
. + * * partition of a set.**@xcite given a no - empty universal set @xmath8 , there exists a family of sets @xmath9 , which is the partition of @xmath8 , if and only if these following conditions hold : 1 .
@xmath10 2 .
@xmath11 3 .
@xmath12 * * equivalent class partition.**@xcite if there is a binary relation @xmath13 on set @xmath8 , then there is a unique partition set @xmath9 of the set @xmath8 . for each component @xmath14 ,
such that there are properties of reflexivity , symmetry and transitivity among all elements in set @xmath15 with respect to @xmath13 .
+ * * cartesian product.**@xcite given @xmath16 sets @xmath17 , there exists a multiplying sets @xmath18 and return _
n - ordered _ vectors set @xmath19 , in which for these members @xmath20 such that @xmath21 .
given a no - empty and connected graph @xmath22 and vertices set @xmath5 .
consider a pair @xmath23 . if there is a binary relation to characterize a behavior of traversal from vertex @xmath4 to vertex @xmath24 .
we define the binary relation as traversal relation , denote by @xmath25 .
we write the form @xmath26 and @xmath27 to represent this relation on pair @xmath28 .
the ordered pair @xmath29 denotes a direction of left to right .
we reserve the notation @xmath30 equal to the first member in ordered pair , and then one equals to @xmath31 .
[ t1 ] if there is a no - empty traversal relation @xmath25 on an instance @xmath32 , then @xmath33 .
given an instance @xmath32 .
let @xmath25 be a traversal relation on instance . as described in definition of cartesian product , for each pair
@xmath34 such that there is @xmath35 .
observe if pair @xmath36 then @xmath37 .
assume that there is a pair @xmath38 .
then we have @xmath39 , a contradiction to definition of traversal relation .
+ it is obviously that there are some properties in traversal relation as follow : 1 .
reflexivity : if @xmath40 , then there may be @xmath41 .
anti - symmetry : if @xmath27 , then there may be @xmath42 .
anti - transitivity : if @xmath43 , then there may be @xmath44 .
let @xmath25 be a traversal relation on instance @xmath32 . for each vertex @xmath40 such that there may be @xmath45 with definition of cartesian product . then observe that there may be pair @xmath46 with @xmath37 as theorem[t1 ] .
hence , @xmath25 may have property of reflexivity .
consider a pair @xmath42 and there is @xmath27 .
observe that we can not say @xmath47 hence , there is no property of symmetry in @xmath25 .
when there are three vertices @xmath48 and pair @xmath44 , similarly we can not say pair @xmath49 with pairs @xmath50 .
we prove there is no transitive relation in @xmath25 and finish this proof .
+ let @xmath51 be a subset of set @xmath25 . if @xmath52 , for two arbitrary components @xmath53 such that @xmath54 .
we call set @xmath51 _ unit subgraph_. @xmath55 denotes the collection of unit subgraphs .
we reserve the subscript of set @xmath51 equal to the one of first member of each component in set @xmath51 .
+ [ t2 ] let @xmath55 be a collection of unit subgraphs on a traversal relation @xmath25 .
then set @xmath55 is the partition of set @xmath25 .
let @xmath55 be a collection of unit subgraphs on traversal relation @xmath25 .
we aim to prove three conditions hold for set @xmath55 on set @xmath25 .
hence , first we can let set @xmath56 .
consider there is an isolated vertex @xmath57 .
it is certainly that pairs @xmath58 with the definition of traversal relation .
hence for each component @xmath59 , there is no such case @xmath60 with @xmath61 by the definition of unit subgraph .
consider a pair @xmath62 . as the definition of unit subgraph
, there naturally may have a component @xmath59 such that @xmath63 , which subscript is @xmath64 .
hence , we have that @xmath65 and @xmath66 .
assume @xmath67 and having a pair @xmath68 .
there certainly may exist a component @xmath69 and introduce pair @xmath70 to @xmath71 , thus observe there may be @xmath72 , a contradiction . hence @xmath73 and @xmath74 .
consider two components @xmath75 with @xmath76 such that @xmath77 .
assume to @xmath78 .
set a pair @xmath79 .
as the definition of unit subgraph , observe there can be @xmath80 , contradicts the given condition of @xmath76 .
hence @xmath81 . to sum up above ,
set @xmath55 is the partition of set @xmath25 .
+ [ t3 ] unit subgraph is a cartesian product set .
let @xmath82 be a unit subgraph .
we have a term to characterize it as follow @xmath83 the form can be written as follow @xmath84 observe that set @xmath82 is a cartesian product set . + as the form
, we call term @xmath85 _ root set _ , denote by @xmath86 . call the right set _ leaf set _ , denote by @xmath87 . therefore , the unit subgraph @xmath82 can be abbr . by @xmath88
. * claim . *
the cardinality of a multiple set is the number of difference members , not be the quantity of members .
we reserve the notation @xmath89 or @xmath90 to represent a component in a multi - set , the @xmath91 is the count of element @xmath92 and @xmath93 .
we call _ group _ for a component containing same and repeated elements .
we define the group minus as that @xmath94 , if and only if @xmath95 .
then the difference value is 0 if @xmath96 . if there is a multiple set @xmath97 and @xmath98 , i.e. for each pairs @xmath99 such that @xmath100 , then we call set @xmath97 _ multiple traversal relation_. for convenience , we use @xmath101 denote each group in set @xmath97 . the notation @xmath91 is the count of the pairs @xmath102
. + let @xmath103 be a subset of multiple traversal relation @xmath97 . if @xmath104 , then for two groups @xmath105 such that @xmath54 .
we call set @xmath103 _ weighted unit subgraph _ , and reserve the subscript of set @xmath103 equals to the one of first element of each pair in set @xmath103 . the collection of weighted unit subgraphs we denote by @xmath106
. + [ t6 ] let @xmath97 be a multiple traversal relation and @xmath106 be a collection of weighted unit subgraphs on set @xmath97 .
then set @xmath106 is the partition of set @xmath97 .
let @xmath106 be a collection of weighted unit subgraphs on multiple traversal relation @xmath97 .
as the definition of weighted unit subgraph , for a group @xmath107 and @xmath108 , there may be @xmath109 such that group @xmath110 may be introduced to set @xmath103 . hence observe @xmath111 if @xmath112 .
if there is @xmath113 , then at least we can have a group @xmath114 . summarizing above ,
it is certainly that there may be a component @xmath109 such that group @xmath115 can be introduced to @xmath103 hence , @xmath116 and @xmath117 .
consider two components @xmath118 with @xmath76 such that @xmath119 .
assume to @xmath120 and at least a pair @xmath121 .
then , we have @xmath80 a contradiction to given condition @xmath76 as described in definition of weighted unit subgraph . hence @xmath122 . for satisfying three conditions for set @xmath106 on set @xmath97 , we understand set @xmath106 is the partition of set @xmath97 .
+ [ d1 ] there is a no - empty traversal relation @xmath25 on graph @xmath22 .
consider each pair @xmath23 and a trail @xmath123 on instance . if the pair @xmath28 lies on trail @xmath123 with a constraint of direction respect to the traversal relation @xmath25
, then we call this constraint _ traversal visiting_.
+ [ t4 ] there is a no - empty multiple traversal relation @xmath97 on an instance g(v ) .
then set @xmath97 characterizes the traversal visiting among all vertices on instance .
given an instance @xmath32 .
let @xmath97 be a multiple traversal relation on it .
consider each pair @xmath124 . if there is a group @xmath125 and no group @xmath126
, it is obviously that there is no traversal visiting on direction @xmath127 , i.e. there is impossible for ordered pair @xmath127 to lie on each trail on instance .
we call this pair _ directed graph_. let @xmath128 .
when @xmath129 , we say there exists a bidirected traversal visiting between pair @xmath130 ; the case is a _
simple graph_. if @xmath131 and @xmath132 , then there are several bidirected and equal visiting to each other
. observe the case is a _ multi - graph_. for @xmath133 and @xmath134 , then there may be unequal visiting opportunities between the pair .
this instance is usually called _
mixed graph_. + [ t5 ] there is @xmath135 and @xmath136 on an instance @xmath32 .
let @xmath97 be multiple traversal relation and @xmath25 be traversal relation on an instance .
as the described in definition of multiple traversal relation , we have @xmath98 . for each group @xmath137
, there is @xmath138 , if and only if @xmath139 .
then @xmath135 , similarly prove @xmath136 .
+ with theorem[t4 ] , consider given a connected graph , there can be these data structures on it as follow : @xmath140 * claim .
* we reserve the abbr .
@xmath141 or @xmath142 to represent a connected instance with no - empty set @xmath25 or @xmath97 respectively
. + * section summary . * in this section , author constructs the basic logic for graph , that is @xmath37 with the property of reflexive . in following
, author will gradually abstract the subset of traversal relation to construct new relation for problems . the new glossary _ unit subgraph
_ indeed is an equivalent class partition in set @xmath25 , because there are three properties of reflexivity , symmetry and transitivity in relation of equal first element of each pair in unit subgraph . in set
theory , the equivalent class also is the partition of set , but it is the unique partition of set respect to certain relation . therefore it is the essential data structure for our research in this paper with the feature of uniqueness .
lemma[t5 ] states a fact that all connected graphs can be viewed as an instance of mixed graph on traversal visiting , such that the exact graph traversal algorithms we will show has to cover all connected graph , which method is data - oriented only .
+ there exist two demands for graph traversal , traversing vertices and traversing edges . in this paper
, author only introduces the problem of traversing vertices .
because the data of traversing edges are huge , and the method is similar to traversing vertices too . + as the definition[d1 ] of traversal visiting , we define a characteristic function @xmath143 as follow @xmath144 because of the case @xmath135 with lemma[t5 ] , therefore we must consider the group weight @xmath145 . then the characteristic function @xmath146 will be converted to map the group with weight to the binary set as follow @xmath147 now when the program enumerates the possible vertices , the program only needs to scan the leaf sets and checks the weights .
the characteristic function @xmath148 provides a method of judgment to ensure enumerating valid vertices .
let @xmath8 be the subset of leaf set .
the weighted unit subgraph is entry parameter .
we define the enumerating operator @xmath149 as follow @xmath150 let @xmath151 be a subset of set @xmath25 . if @xmath152 . for two arbitrary pairs
@xmath153 such that @xmath154 , then we call set @xmath151 _
visiting set_. the collection of visiting sets we denote by @xmath155 ; and reserve the subscript of component in set @xmath155 equals to the one of then element in each pair in native component .
use @xmath156 to denote the set of all first elements in pairs and the then elements set is @xmath157 .
it is easy to prove visiting set is a cartesian product set like unit subgraph .
+ [ a1 ] let @xmath155 be a collection of visiting sets on set @xmath25 . then set @xmath155 is the partition of set @xmath25 .
there is a collection @xmath155 of visiting sets on set @xmath25 . as the described in definition of visiting set , for a pair @xmath62
, then there may exist a visiting set @xmath158 and @xmath159 .
hence , observe that for each component @xmath160 , there may be @xmath161 and @xmath162 .
assume to @xmath163 and at least a pair @xmath164 .
there may be a visiting set @xmath165 such that pair @xmath166 can be introduced to set @xmath167 .
hence there is a contradiction of @xmath168 then , @xmath169 .
consider two components @xmath170 with @xmath76 such that @xmath171 .
assume to @xmath172 and a pair @xmath173 .
with definition of visiting set , we can have @xmath174 a contradiction to given condition of @xmath76 .
hence @xmath175 summarizing , we can understand set @xmath155 is the partition of set @xmath25 . indeed , the proof is in same fashion with unit subgraph , because they both are two equivalent classes on set @xmath25 . + let @xmath176 be a subset of set @xmath97 . if @xmath177 .
for two arbitrary groups @xmath178 , such that there is @xmath154 .
we call the set @xmath176 _ multiple visiting set _ , the collection of multiple visiting sets denote by @xmath179 . we reserve the subscripts of components in set
@xmath179 equals to the ones of then elements in each group in native component .
+ [ a2 ] let @xmath179 be a collection of multiple visiting sets on set @xmath97 .
then set @xmath179 is the partition of set @xmath97 .
indeed we aim to prove three conditions hold for set @xmath179 on set @xmath97 . with the same fashion of proof in theorem[t6 ] , it is clearly for us to prove the fact on set @xmath97 .
here we need not do the repeated work again .
+ let @xmath180 be a collection of cartesian product of set @xmath5 on an instance @xmath181 , having @xmath182 with @xmath183 .
let @xmath184 be a member in set @xmath180 .
if for each ordered pair @xmath185 lies on sequence @xmath184 , such that there is ordered pair @xmath186 , then we call sequence @xmath184 _ connected path_. + [ p ] let @xmath184 be a connected path on graph @xmath181 and @xmath179 be a collection of multiple traversal visiting sets .
if there is such an approach of cutting graph , for each group @xmath187 such that @xmath188 , then we call this approach _ equivalent visiting_. sequence @xmath184 is called _ equivalent visiting path_. + with definition[p ] , we define the equivalent visiting operator @xmath189 as follow @xmath190 [ a3 ] let @xmath184 be an equivalent visiting path on graph @xmath181 .
there is an approach of equivalent visiting on path @xmath184 .
consider a group @xmath191 and @xmath192 .
then the weight @xmath91 may be converged to 0 by invoking equivalent visiting operator @xmath189 if group @xmath193 lies on @xmath184 and @xmath194 .
let @xmath184 be an equivalent visiting path on an instance @xmath181 , on which there is an approach of equivalent visiting .
we set there is a component @xmath195 with @xmath192 .
consider a group @xmath196 .
if @xmath197 , we can understand that equivalent visiting operator @xmath189 do nothing inducing from the inner characteristic function @xmath198 , as definitions of these functions .
+ if @xmath199 , then the @xmath91 may be self - subtract - one and returned by operator @xmath189 inducing from inner function @xmath200 .
consider the vertex group @xmath187 with @xmath201 . as described in definition of equivalent
visiting , this case can lead to @xmath202 , such that there is @xmath203 by iteratively invoking operator @xmath189 .
when @xmath204 , the function @xmath205 would return with nothing , thus we can understand that while for @xmath206 , the function @xmath189 can not continue to compute @xmath91 as the inducing from function @xmath207 .
hence , the @xmath91 can be converged to 0 .
+ [ a4 ] let @xmath179 be a collection of multiple visiting sets and @xmath184 be an equivalent visiting path on instance . if for each component @xmath195 and each group @xmath208 such that @xmath209 , then for each vertex group @xmath210 and @xmath211 , we have @xmath212 .
let @xmath179 be a collection of multiple visiting sets and @xmath184 be an equivalent visiting path on instance @xmath181 .
for each component @xmath195 , we let @xmath213 represent all weights of groups in set @xmath176 .
consider vertex @xmath211 and having group @xmath210 with @xmath201 . as the definition of equivalent visiting
, there is @xmath214 . because the @xmath215 , we have that @xmath216 for @xmath217 .
it implies the fore equation can be view as an iterative equation @xmath218 with @xmath219 .
let @xmath220 . when @xmath221 , for each @xmath222 such that there is @xmath223
. then we have each @xmath224 with lemma[a3 ] , and the number set @xmath225 converges at 0 . for each weight of group
@xmath166 is equal to 0 , the enumerating function @xmath226 can not introduce vertex @xmath4 to path @xmath184 as a valid vertex again .
then for each vertex @xmath4 , such that at most there are @xmath227 possibilities on path @xmath184 .
namely , @xmath212 .
if @xmath228 , we can set @xmath151 is a constant , then @xmath229 with formula @xmath214 .
hence , the case does not exist with respect to invoking enumerating functiuon @xmath226 .
because of this theorem shows the maximum possibility of visiting a vertex , author call it _
equivalent visiting maximum value theorem_. + [ a5 ] let @xmath184 be an equivalent visiting path on a finite and no - empty instance @xmath181 . then path @xmath184 can be convergent . given a finite and no - empty graph @xmath181 .
let @xmath184 be an equivalent visiting path .
consider each group @xmath210 with @xmath201 , then there is @xmath211 such that @xmath230 .
further we can understand the term equals to @xmath231 with set @xmath179 is partition of set @xmath97 as theorem[a2 ] .
let @xmath232 and @xmath151 be a constant .
we can see sequence @xmath184 as a discrete point - sequence for iterated function @xmath233 , and have @xmath234 observe the function @xmath233 is iterative and monotone decreasing with input @xmath151 . because of instance being finite , therefore @xmath233 can be convergent at 0
. then path @xmath184 converges , to which function @xmath226 can not introduce any vertex with all weights equal to 0 .
similarly consider the end - node @xmath235 on current path @xmath184 with @xmath236 .
for each group @xmath237 , if each @xmath91 equals to 0 , even nor @xmath233 converges at 0 , then path similarly converges .
assume the path is infinite .
as the described in theorem[a4 ] , there are infinite weights , and then it implies that at least a number of a pair is infinity .
then , this assumption contradicts the condition of finite instance .
+ [ a6 ] let @xmath184 be an equivalent visiting path on graph @xmath181 and @xmath179 be a collection of multiple visiting sets .
consider each maximum weight @xmath238 in each component @xmath195 . then @xmath239
. given a finite and no - empty graph @xmath181 .
let @xmath184 be an equivalent visiting path on graph @xmath22 . with lemma[a5
] , the path @xmath184 is convergent .
consider each component @xmath195 , in which the maximum weight of group is @xmath238 .
with theorem[a4 ] , we understand that each vertex @xmath57 can lie on path @xmath184 for at most @xmath238 possibilities .
hence there is @xmath240 , if @xmath241 as described in lemma[a5 ] .
+ consider there is such case , of which each weight of group in weighted unit subgraph equals to 0 as input for enumerating function @xmath226 .
the function @xmath226 can return with nothing .
the path @xmath184 can be forced to converge even nor @xmath241 .
its length can be shorter than @xmath242 .
hence , @xmath239 it holds .
+ [ a7 ] if there are self - cycles on an instance @xmath243 , then they are invalid traversal visiting on equivalent visiting . given a no - empty graph @xmath244 .
let @xmath57 and pair @xmath245 .
there is pair @xmath246 and its weight equals 1 as the definitions of unit subgraph and visiting set . if there is an approach of equivalent visiting on instance . when the vertex @xmath235 lies on path @xmath184 , as the definition of equivalent visiting , we have @xmath247 . then the weight of pair @xmath248 will be forced to subtract 1 , such that enumerating operator @xmath226 can not introduce vertex @xmath235 again .
hence , the approach can not traverse self - cycle , i.e. the self - cycle equals to an empty traversal visiting on instance .
+ here author briefly shows the viewpoint about the case of self - cycle on simple graph : the self - cycle at least is an instance of _
russell paradox_. look at the term @xmath246 above .
when we partition the set @xmath25 with the relations of @xmath54 or @xmath154 , we can not say the self - cycle is arriving or leaving on vertex .
thus the method of equivalent visiting on simple graph would filter out self - cycle as the symmetry relation of leaving and arriving . on an instance , if an approach obtains those equivalent visiting paths depend on iteratively and alternately invoking enumerating function @xmath226 and equivalent visiting function @xmath189 to enumerate vertices and modify the traversal relations , we call this approach _ based on table search _ , abbr . by _
bots_. [ a8 ]
the approach bots can enumerate every equivalent visiting path on a simple graph @xmath244 .
given a no - empty graph @xmath244 .
let @xmath184 be a connected path on graph .
consider a pair @xmath249 with @xmath250 .
if vertex @xmath235 lies on path @xmath184 , bots would invoke operator @xmath189 such that @xmath251 with lemma[a5 ] .
if having pair @xmath252 , then there can be @xmath253 , moreover its weight equals to 0 .
but for pair @xmath254 such that it weight is still 1 .
hence , bots can introduce vertex @xmath255 to path @xmath184 by invoking the function @xmath226 to scan the leaf set in subgraph @xmath82 .
hence , function @xmath189 can not affect this work of function @xmath226 . with the weight of pair @xmath256(@xmath257 )
is equal to 0 , while the bots scan the leaf set in subgraph @xmath258 , it can not enumerate vertex @xmath235 as the next valid vertex . the method of equivalent
visiting prevents repeated visiting , but not to block enumerating vertices , which weights are equal to 1 .
+ consider the reasons of search terminating .
let vertex @xmath235 be current node on path @xmath184 .
if @xmath259 , as the described in theorem[a4 ] the enumerating operator @xmath226 can return with nothing , then the length of path @xmath184 may be less than or equal to @xmath260 . if @xmath261 , then the path is a dead - end path . otherwise , the path is a hamiltonian path .
namely , the bots may enumerate all connected paths on graph @xmath22 .
+ author gave two equivalent classes for traversal relation , so that we can define two operators on these classes .
their works can interactively constraint each other , according to the constraint of traversal visiting and method of equivalent visiting .
consequently , the approach of bots has no necessary to be a recursive method . because there is not any demand of traversing a mixed graph or multi - graph , therefore author
will not argument those problems in this paper .
hence all works stop at simple graph traversal . here
we will discuss the problem on the level of program .
summarize the augments above , we can obtain some conclusions as follow : the traversal relation can be organized as a table , in which unit subgraph can be a unit of data .
we can evaluate the longest length of path .
when the enumerating function returns an empty set , we know this search work on current path is over .
hence , we need define three set for approach as follow : first is the set _ stack _ , in which there are the path waiting for search . set _
p _ is second , which is a path containing a sequence of vertices in process of current exploring . and
then set _ r _ stores the final results and returns finally .
the following pseudocode for the approach is given as algorithm [ cols="<",options="header " , ] with rough viewpoint , program need do much more works of comparing among arrays to remove the repeated data .
such that the price of runtime and memory increase quickly .
because of @xmath262 on dodecahedron figure , therefore it must at least enumerate civss for two times .
let @xmath263 represent the first loop , and similarly the second one has @xmath264 .
then there exists a combination formula @xmath265 such that we have a number of combination @xmath266 , i.e. there is a search breadth of 3969 .
we can evaluate the number of combination for a given instance as follow @xmath267 we can let @xmath268 , then complexity is @xmath269 .
it is similar to the current approaches .
* summary . * in this work we studied how to cut graph actually , with such logic structure @xmath270 .
we proved that some problems can be quantified so that it can be a basic relation model for applications .
similarly , we proved some axioms in current theory and show why some things are so hard to us .
the algorithmic contribution focused on the data - structure such that solve the problem of general . in the process , the equivalent class , unit subgraph is the keypoint .
it let us freely choose the method to abstract basic relation for construct new logic model .
for example we abstract the edge relation from it , and finally construct two classes , with those properties of symmetry and transitivity .
in fact , there are more methods to abstract this binary relation for problems such as ai , flow network , tsp . due to limited space of page
, author can not continue to do these works .
* future * work . a wide range of possible future work exists for present abstract relation among those objects , e.g. tsp .
we give the cutting graph by bogpc or boerc . for among each civs
, those edges are the bridges between two arbitrary domains of vertices , and for those domains indeed , there exists the relation of inequivalence - color among them .
it lets us may have a nice condition to use the _ greedy algorithm _ to exactly solve this problem , so that graph coloring is not a pure problem of graph theory .
further for ai , we can use graph partition to characterize the process of solving some problems based on @xmath271 conditions .
finally , we pose the conjecture in following
. + * conjecture .
* there are two binary relation @xmath13 and @xmath272 with @xmath273 on a universal set @xmath274 , which lead to two equivalent classes @xmath275 and @xmath276 on @xmath274 respectively .
if for @xmath277_{\rho}\cap [ a_i]_{\bar{\rho}}$ ] such that @xmath278 , then @xmath279 is _
russell paradox_. + _ reason
. _ self - cycle appears in graph traversal but vanishing in edge relation . and traversal relation justly possesses property of reflexivity without symmetry , to contrary for edge relation . and we can find the equivalent classes for traversal relation in @xmath280 , but not on edge relation .
99 bellman , r. ( 1960 ) , `` combinatorial processes and dynamic programming '' , in bellman , r. , hall , m. , jr .
( eds . ) , combinatorial analysis , proceedings of symposia in applied mathematics 10 , american mathematical society , pp .
kernighan , b. w. ; lin , shen ( 1970 ) .
`` an efficient heuristic procedure for partitioning graphs '' .
bell systems technical journal 49 .
lucas , john f. ( 1990 ) .
introduction to abstract mathematics .
rowman & littlefield .
isbn 9780912675732 .
burnstein , ilene ( 2003 ) , practical software testing , springer - verlag , p. 623
, isbn 0 - 387 - 95131 - 8 hazewinkel , michiel , ed .
( 2001 ) , `` direct product '' , encyclopedia of mathematics , springer , isbn 978 - 1 - 55608 - 010 - 4 | in this paper , author uses set theory to construct a logic model of abstract figure from binary relation . based on the uniform quantified structure ,
author gives two logic system for graph traversal and graph coloring respectively , moreover shows a new method of cutting graph . around this model ,
there are six algorithms in this paper including exact graph traversal , algebra calculation of natural number , graph partition and graph coloring . |
the field of semiconductor spintronics is rapidly developing nowadays . the idea to combine the well established data processing capabilities of semiconductor electronics with ferromagnetism may lead to new functionalities and low power consumption of devices@xcite . one of the main obstacle for spin injection into a semiconductor is the conductivity mismatch at the interface of the ferromagnetic metal and the semiconductor@xcite .
one way to avoid it is to use a thin insulating layer acting as a tunnel barrier between the two materials .
another approach is to design the spin injecting interface with a similar structure and properties by alloying or doping the semiconductor with a magnetic element .
the intermetallic magnetic mn@xmath0ge@xmath1could provide the desired solution as it grows directly onto ge substrate@xcite , therefore being compatible with existing semiconductor technology .
mn@xmath0ge@xmath1shows ferromagnetism with a curie temperature ( t@xmath5 ) around room temperature@xcite and an important spin polarization ( up to 42%)@xcite .
the mn@xmath0ge@xmath1hexagonal cell contains 10 mn atoms which are arranged in two different sublattices ( mn@xmath6 and mn@xmath7 ) due to different coordination .
inserting carbon atoms into interstitial voids of mn@xmath7 octahedra leads to an increase of t@xmath5 up to 450k , supplying a solution for the room temperature spin injection@xcite .
ab - initio _ calculations indicate that the structural distortions have a small influence on the increased t@xmath5 in mn@xmath0ge@xmath1c@xmath8(the lattice is compressed compared to pure mn@xmath0ge@xmath1 ) , with the enhanced ferromagnetism attributed to a 90@xmath9 ferromagnetic superexchange mediated by carbon@xcite .
several preparation methods were used to grow mn@xmath0ge@xmath1thin films .
the most common growth method is the solid phase epitaxy which consists in the deposition of mn or mn and c on a ge(111 ) layer followed by an annealing leading to the formation of the mn@xmath0ge@xmath1or mn@xmath0ge@xmath1c@xmath8films . due to the low mn solubility in ge ,
secondary precipitates or mn - rich regions / clusters frequently appear inside the mn@xmath0ge@xmath1films .
mn atoms also diffuse in the underlying ge(111 ) substrate which deteriorates the interface quality . in this article
, we report on the structural and magnetic properties of thin films c - doped mn@xmath0ge@xmath1epitaxially grown on ge(111 ) by reactive deposition epitaxy ( rde ) at room temperature .
the low growth temperature reduces segregation and allows the formation of thin films of excellent crystalline quality suitable for the determination of various magnetic parameters by fmr : magnetic anisotropy , magnetisation and the _ g_-factor which were quantitatively determined and theirs dependence on carbon content and temperature was identified . from the study of the fmr linewidth , the magnetic relaxation process is investigated and the magnetic relaxation parameters are found .
the main relaxation channels we identify are the intrinsic gilbert damping and the two - magnon scattering .
the intrinsic magnetic damping measured by fmr determines the time scale of the dissipation of magnetic energy into the lattice .
we determine that c - doped mn@xmath0ge@xmath1has a very low damping of spin motions , with an intrinsic gilbert damping constant @xmath4 between 0.005 and 0.01 , which is one of the lowest known value for ferromagnetic mn - based thin - film compounds . in the two - magnon scattering process , the uniform fmr mode ( * * k**=0 ) can scatter to degenerate non - uniform modes ( * * k**@xmath10 0 ) with an effective interaction matrix proportional to the fourier transform of samples inhomogeneities@xcite .
we observe that the role of the scattering centres changes at low temperatures in mn@xmath0ge@xmath1c@xmath3where a superposition of twofold and fourfold symmetry dominates and not the usual six - fold as at room temperature .
the ferromagnetic resonance measurements demonstrate the very good structural quality and the low magnetic damping in the c - doped mn@xmath0ge@xmath1 , paving the way for heterostructure integration and spintronic applications .
the sample preparation as well as the reflection high - energy electron diffraction ( rheed ) measurements were performed in a uhv setup with a base pressure of 2.7@xmath11 pa .
mn@xmath0ge@xmath1c@xmath8layers were grown epitaxially on ge(111 ) substrates@xcite .
these substrates were chemically cleaned before introduction in the uhv chamber .
then we did a degassing of the ge(111 ) substrates by direct heating up to 720 k for 12 h and flashed afterwards at 1020 k to remove the native oxide layer .
after repeated flashes at 1020 k and a cooling down at 770 k , a 15 nm thick ge buffer layer was deposited on the ge(111 ) substrates to make sure that the starting surface of the mn@xmath0ge@xmath1c@xmath8thin films growth is of good quality .
the quality of this starting surface was checked _ in - situ _ by rheed .
eventually the sample was cooled down to room temperature ( rt ) . to form the mn@xmath0ge@xmath1c@xmath8layers we used the reactive deposition epitaxy method@xcite . using this method
the mn@xmath0ge@xmath1c@xmath8layers are created by phase nucleation at the surface of the sample during the epitaxial growth .
no diffusion phenomenon is required for the growth unlike the solid phase epitaxy process which is usually employed to form the mn@xmath0ge@xmath1c@xmath8films on ge(111 ) .
however a good control of the different flows is needed to match the stoichiometry of the desired compound : ge and mn were evaporated using knudsen cells and c atomic flow was obtained thanks to a high purity pyrolytic graphite filament source ( suko ) from mbe - komponenten .
the ge and mn flows were calibrated with a water - cooled quartz crystal microbalance and the c flow was calibrated using the structure transition between the si(001 ) ( 2@xmath121 ) and c(4@xmath124 ) reconstructions which occurs for a c deposited thickness of 0.4 atomic monolayer on si(001 ) surfaces@xcite .
the growth of the mn@xmath0ge@xmath1c@xmath8films was monitored _ in - situ _ by rheed : the mn@xmath0ge@xmath1c@xmath8films growing epitaxially on a ge(111 ) surface exhibit an easily identifiable rheed ( @xmath13)r30@xmath14 pattern which is characteristic of the mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath8compounds@xcite .
the saturation magnetisation and the estimated curie temperatures of all samples were determined by squid measurements . a squid magnetometer quantum design mpmsxl working in a temperature range 1.8k to 300k and in a magnetic field up to 5 t was used .
the fmr measurements were performed with a conventional x - band ( 9.39ghz ) bruker emx spectrometer in the 80k to 300k temperature range .
the samples ( @xmath15mm@xmath16 ) were glued on a quartz suprazil rode and mounted in the centre of a rectangular cavity ( te@xmath17 ) . to improve the signal - to - noise ratio ,
the fmr measurements were carried out using a modulation field of 100khz and 5oe amplitude with a lock - in detection .
the fmr spectra were measured with the applied magnetic field rotated in plane and out - of - plane .
the fmr spectra were fitted to a lorentzian profile and the resonance field and fwhm linewidth were subsequently extracted .
typical spectra at rt are shown in figure [ fig.0](a ) for 12 nm thick films .
the signal - to - noise ratio ( snr ) of mn@xmath0ge@xmath1at 300k ( figure [ fig.0 ] ) is 38 which is the lowest snr of all measured spectra and is due to the proximity to t@xmath5 .
the snrs for the other spectra of figure [ fig.0 ] are 115 and 250 for mn@xmath0ge@xmath1c@xmath3and mn@xmath0ge@xmath1c@xmath2respectively which have higher t@xmath5 .
( color online ) ( a ) typical spectra at rt for 12 nm thick mn@xmath0ge@xmath1 , mn@xmath0ge@xmath1c@xmath2and mn@xmath0ge@xmath1c@xmath3thin films .
( b ) schema of the coordinate system used in the fmr measurements.,title="fig:",width=302 ] +
the fmr spectra were analyzed with the smit - beljers formalism for a thin film with uniaxial ( hexagonal ) symmetry@xcite . for a ferromagnetic film with hexagonal symmetry ,
the free energy density including the zeeman energy , the demagnetizing energy and the anisotropy energy density is written as : @xmath18 \nonumber\\ & - ( 2\pi m^2 - k_2)\sin^2\theta + k_4\sin^4\theta + k_{6\perp}\sin^6\theta \nonumber\\ & + k_{6\parallel}\sin^6\theta\cos6\varphi\end{aligned}\ ] ] where @xmath19 , @xmath20 are the polar and azimuthal angle of the external field with respect to the surface normal of the thin film ( the [ 001 ] direction ) and , respectively , the [ 100 ] direction , @xmath21 and @xmath22 are the polar and azimuthal angle of the magnetisation with respect same directions ( figure [ fig.0](b ) ) and k@xmath23 are the anisotropy constants to sixth order . the resonance condition , neglecting the damping effects and considering the magnetisation at equilibrium under steady field ,
is given by : @xmath24 where @xmath25 and @xmath26 represent the stiffness fields evaluated at the equilibrium angles of the magnetisation : @xmath27 equation is valid for a high - symmetry case , where the mixed second derivative of the free energy is nil .
our experiments were carried out in two distinct geometries : 1 . out - of - plane geometry ( @xmath28 , @xmath19 variable ) .
the stiffness fields are the following : @xmath29 2 . in - plane geometry
( @xmath30 , @xmath20 variable ) .
the stiffness fields are : + @xmath31 here @xmath32 , @xmath33 the angular frequency and @xmath34 = g@xmath35/@xmath36 the gyromagnetic ratio .
the fmr linewidth is analyzed by including the intrinsic and extrinsic damping mechanisms@xcite : @xmath37 in this expression , the intrinsic contribution due to the magnon - electron interaction can be described by the dimensionless gilbert damping parameter @xmath4@xcite : @xmath38 where @xmath39 is the dragging function as the magnetisation * m * is dragged behind * h * owing to anisotropy . when * m * and * h * are parallel , this contribution vanishes . as generally the in - plane and out - of - plane linewidth
are not equal , extrinsic contribution have to be taken into account .
the extrinsic contribution generally include the magnetic relation due to magnon - magnon interaction , the two - magnon interaction , which is given by@xcite : @xmath40 with @xmath41 the two - magnon scattering rate .
the two - magnon contribution usually vanishes for a critical out - of - plane angle @xmath42 .
short - range fluctuations in magnetic properties due to sample imperfections lead to two - magnon scattering .
inhomogeneous broadening effects ( superposition of local resonances ) due to long - range fluctuations of magnetic properties also participate to the extrinsic linewidth , especially at intermediate angles as the resonance local field can vary .
we consider here three types of inhomogeneous broadening : @xmath43 and @xmath44 .
the first term is the mosaicity term due to the distribution of easy axes directions@xcite : @xmath45 with @xmath46 = ( @xmath47 ) .
the second term represents the inhomogeneity of the internal fields in the sample@xcite : @xmath48 finally , the last term which can contribute to the linewidth is a residual frequency and angular independent inhomogeneous linewidth that can not be put in other form and is also due to long - range fluctuations of magnetic properties .
the procedure used to determine the magnetic parameters was as follows : the anisotropy fields were determined using the system of equations - applied at high symmetry directions ( along easy / hard axes ) together with the corresponding measured resonance fields ( fixed frequency ) at a fixed _ g_-factor .
afterwards , the polar and azimuthal angular dependence of the resonance field was fitted to the same equations and the equilibrium condition of the free energy allowing for a variable _ g_-factor as parameter .
the iteration was repeated several times until a good fit was obtained .
this analysis yields the _
g_-factor , the anisotropy constants and the magnetisation direction @xmath21 .
these values serve in the fit of the angular variation of the linewidth which allows the numerical evaluation of @xmath4 , @xmath41 and the inhomogeneous contributions using - .
in this section , experimental results of c - doped mn@xmath0ge@xmath1thin films investigated by ferromagnetic resonance and squid magnetometry are presented . using samples with different carbon content , we determined the magnetic anisotropy energy , the _
g_-factor , magnetisation and magnetic relaxation parameters . to determine the magnetic energy anisotropy ( in absolute units ) ,
fmr measurements were carried out at a frequency of 9.4ghz .
the fmr spectra were recorded as a function of the polar and azimuthal angles of the external magnetic field at different temperatures .
the saturation magnetisation was determined from squid measurements . in figure [ fig.1](d ) , the temperature dependence of the magnetisation up to 300k is shown for mn@xmath0ge@xmath1 , mn@xmath0ge@xmath1c@xmath2and mn@xmath0ge@xmath1c@xmath3 . the curie temperature was estimated from these curves using a mean - field approximation which gives a good approximation of the temperature dependence of the magnetisation .
the full line correspond to a fit with the brillouin function b@xmath49 ( s=3/2 ) and the dotted line to a fit with b@xmath50 ( s=1 ) that follows better the experimental points of mn@xmath0ge@xmath1c@xmath3 .
the estimated values of t@xmath5 are 315k , 345k and 450k .
the error bars correspond to @xmath5110k for mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath2as the experimental points cover a larger temperature range and superpose closely with b@xmath49 .
the experimental points for mn@xmath0ge@xmath1c@xmath3cover only a small part of the temperature range and the error bars are estimated to be of @xmath5130k .
+ the out - of - plane angular variation for the resonance field h@xmath52 is shown in figure [ fig.1](a)-(c ) for mn@xmath0ge@xmath1 , mn@xmath0ge@xmath1c@xmath2and mn@xmath0ge@xmath1c@xmath3at room temperature .
the h@xmath52(@xmath19 ) curves indicate an in - plane easy axis with a minimum resonance field of 1.6koe , 2.3koe and 2.7koe for mn@xmath0ge@xmath1c@xmath3 , mn@xmath0ge@xmath1c@xmath2and mn@xmath0ge@xmath1respectively for @xmath53 .
the hard axis is perpendicular to plane ( [ 001 ] direction ) and has the highest h@xmath52 of 8.6koe , 6koe and 5koe .
the azimuthal angular dependence of the resonance field for mn@xmath0ge@xmath1c@xmath3 , recorded also at 300k is shown in figure [ fig.1](e ) .
the sixfold ( hexagonal ) symmetry in the azimuthal angular dependence indicates that an in - plane hexagonal anisotropy exists with one easy axis along the [ 100 ] direction of the film .
the experimental fmr data of out - of - plane and in - plane dependence of the resonance field can be well simulated with and the anisotropy fields can be extracted .
the anisotropy constants can be found in absolute units by using the sample magnetisation determined from squid measurements . @ l| * 6c sample & t(k ) & 4@xmath54 m@xmath55(koe ) & k@xmath56(erg/@xmath57 ) & k@xmath58(erg/@xmath57 ) & k@xmath59(erg/@xmath57 ) & @xmath60(ghz / koe ) + & 300 & 1.5 & 0.37@xmath12 10@xmath61 & 2832 & & 2.8 + & 200 & 4.0 & 2.01@xmath62 & -2.28@xmath63 & & 2.8 + & 150 & 4.8 & 2.36@xmath62 & -2.93@xmath64 & & 2.8 + & 300 & 2.6 & 1.65@xmath12 10@xmath61 & 3.85@xmath65 & & 2.8 + & 250 & 3.8 & 2.71@xmath12 10@xmath61 & -1901 & & 2.8 + & 200 & 4.4 & 3.37@xmath62 & -5131 & & 2.8 + & 100 & 5.0 & 4.29@xmath62 & 2.58@xmath65 & & 2.8 + & 300 & 5.3 & 4.39@xmath12 10@xmath61 & 4.41@xmath65 & 27 & 2.84 + & 250 & 5.8 & 4.78@xmath12 10@xmath61 & 5.53@xmath65 & 134 & 2.84 + & 150 & 6.6 & 5.19@xmath12 10@xmath61 & 5.35@xmath65 & & 2.84 + & 100 & 7.0 & 5.28@xmath12 10@xmath61 & 4.61@xmath65 & & 2.84 + the resulting anisotropy constants are summarised in table [ table1 ] along with the _ g_-factor at several temperatures .
the positive sign of k@xmath56 indicates that this term favors an out - of - plane easy axis of magnetisation while the shape anisotropy dominates@xcite . in the very thin film limit
, k@xmath56 could overcome the shape anisotropy resulting in an out - of - plane anisotropy axis .
the different k@xmath23 have a different temperature dependence . for mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath2 ,
the sixfold in - plane symmetry is to low to be extracted , therefore only the k@xmath56 and k@xmath58 constants were determined from the angular measurements .
k@xmath56 is positive for mn@xmath0ge@xmath1and c - doped mn@xmath0ge@xmath1at all temperatures and increases at low temperature .
k@xmath58 decreases ( increases in absolute values ) for mn@xmath0ge@xmath1 , but for the c - doped compounds has a minimum or a maximum at an intermediate temperature . the sixfold in - plane anisotropy in mn@xmath0ge@xmath1c@xmath3increases at 250k from the room temperature value , while at lower temperature becomes to small or a transition to a fourfold in - plane anisotropy arises as will be inferred from the linewidth temperature dependence discussed in the next section .
the error bars for the anisotropy constants presented in table [ table1 ] are around 5@xmath66 , which come from the computation of m from squid measurements due to the uncertainty in the volume of our samples .
the _ g_-factor can be estimated from the angular dependence of the resonance field .
its value indicates the influence of the orbital contribution to the total magnetic moment .
the ratio of the orbital to the spin magnetic moment can be inferred from the kittel formula and is equal to the deviation of the _
g_-factor from the free electron value .
the value of the _
g_-factor for mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath2is 2.0005 , while for mn@xmath0ge@xmath1c@xmath3this value increases to 2.0291 meaning an increased orbital contribution with carbon doping ( 1.5% of the spin magnetic moment ) .
the linewidth of the resonant signal @xmath67 is directly related to the magnetic and structural quality of the films and provide information about the different relaxation channels in magnetic damping . in figure [ fig.2 ] , the temperature dependence of the fmr linewidth is shown for the perpendicular to plane direction ( @xmath68 ) for mn@xmath0ge@xmath1and c - doped mn@xmath0ge@xmath1 . a shallow minimum
is observed for all three compounds around 200k and a sharp peak close to t@xmath5 . at lower temperature , the fmr linewidth increases and saturates for mn@xmath0ge@xmath1(measured to 6k ) .
the minimum in the linewidth seems not related with the c - doping .
it occurs around the same absolute value of temperature and could be related with a small in - plane transition to a fourfold anisotropy from sixfold anisotropy ( tetragonal distortion ) or to a constriction by the substrate .
the increase of linewidth at low temperature was explained as an inhomogeneous broadening due to the increase of the anisotropy constants ( k@xmath56 ) with decreasing temperature@xcite .
the error bars for the measured fmr linewidth are less than 2@xmath66 in all cases .
( color online ) temperature variation of the resonance linewidth for mn@xmath0ge@xmath1 , mn@xmath0ge@xmath1c@xmath2and mn@xmath0ge@xmath1c@xmath3.,title="fig:",width=321 ] + figure [ fig.3 ] and figure [ fig.4](a ) show the out - of - plane variation of the fmr linewidth for the c - doped mn@xmath0ge@xmath1at room and low temperatures .
the shape of the curves shows the characteristic dependence for thin films with a maximum of the linewidth at intermediate angles .
our films have an in - plane easy axis at all temperatures , therefore the magnetisation lags behind the applied field when the field is rotated out of the plane .
the peak in the linewidth occurs for @xmath19 between 20@xmath9 at room temperature and 10@xmath9 at low temperature , corresponding to the largest interval between * m * and * h*. from the theoretical fits of the data ( solid lines ) obtained using - , the relaxation parameters are extracted and listed in table [ table2 ] .
( color online ) out - of plane angular dependence of the resonance linewidth for mn@xmath0ge@xmath1c@xmath2at different temperatures .
the full lines represent a fit with intrinsic and extrinsic contributions , the dashed lines indicate the two - magnon contribution while the dash - dotted lines indicate the intrinsic contribution .
the symbols indicate experimental data .
the typical mosaicity and internal fields ( @xmath69 ) contributions are shown at 200k and 100k respectively.,title="fig:",width=264 ] + for all compounds , the perpendicular to plane linewidth is always smaller than the in - plane one indicating the presence of two - magnon scattering and other extrinsic contributions in the samples .
the intrinsic damping alone can not explain the out - of - plane shape of the @xmath70 dependence as observed in figure [ fig.3 ] , where the two - magnon and intrinsic gilbert contribution are shown individually for comparison .
the respective values of the two contributions are further compared in table [ table2 ] by computing the gilbert damping parameter g and the two - magnon scattering contribution @xmath71 in s@xmath72 .
the contribution of the intrinsic linewidth ( gilbert ) to the total linewidth varies between 7@xmath66 and 33@xmath66 .
the mosaicity contribution is low for all cases and its angular dependence induces the v - shaped form around @xmath19=0@xmath9 .
a typical mosaicity contribution is shown in figure [ fig.3 ] for mn@xmath0ge@xmath1c@xmath2at 200k .
the inhomogeneity of the internal field gives mainly a low contribution around @xmath19=0@xmath9 , with the exception of mn@xmath0ge@xmath1c@xmath2at room temperature where it gives a larger contribution for @xmath19 close to the plane .
a typical variation of this contribution is shown in figure [ fig.3 ] at 100k .
the estimated intrinsic damping @xmath4 is considered isotropic and independent of temperature in the considered temperature range ( 100k-300k ) .
we prefer using the dimensionless parameter @xmath4 which varies between 0.005 and 0.01 over the gilbert damping parameter g given by @xmath4=g/@xmath34 m as the latter will imply a strong temperature dependence .
the gilbert damping represents the decay of magnetisation by direct viscous dissipation to the lattice as it is introduced in the landau - lifschitz - gilbert equation@xcite .
the spin - orbit coupling is assumed to be at the origin of spin - lattice relaxation in ferromagnets . _
ab - initio _ calculations that include the spin - orbit coupling explicitly show a weak dependence of @xmath4 with temperature in a large range of temperatures@xcite .
two different mechanisms contribute to the temperature dependence@xcite , one conductivity - like and one resistivity - like with a transition between the two at intermediate temperature .
sometimes these two contributions have an equal influence on the damping .
other models predict an anisotropic tensorial damping constant @xmath73@xcite and its proportionality to 1/m@xcite , leading to a strong variation close to t@xmath5 .
we estimated the value of @xmath4 for each compound by fitting the out - of - plane angular dependence of @xmath67 at a temperature corresponding to the minimum of the curves in figure [ fig.2 ] ( around 200k ) .
for this specific temperature , the estimation corresponds to the maximum possible value of @xmath4 considering small inhomogeneous broadening ( @xmath74 and @xmath75 ) .
although we consider a constant @xmath4 , as it is observed from table [ table2 ] , at room and low temperature the linewidth ( and correspondingly the inhomogeneous residual field ) increases for mn@xmath0ge@xmath1c@xmath2which could be explained by an increase of @xmath4 at least at low temperature .
the room temperature increasing in the linewidth is usually explained as a breakdown of the uniform precession due to thermal excitations ( spin - waves)@xcite .
the increasing of the linewidth at low temperature is smaller for mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath3 in the 100 - 300k temperature range being compatible with a constant @xmath4 as considered for this temperature range .
@ l| * 8c sample & t(k ) & @xmath4 & g(10@xmath76s@xmath72 ) & @xmath77(oe ) & @xmath78(10@xmath76s@xmath72 ) & @xmath79(deg ) & @xmath80m@xmath55)(oe ) & @xmath75(oe ) + & 300 & 0.01 & 0.54 & 150 & 26.38 & 0.05 & 20 & 270 + & 200 & 0.01 & 1.29 & 340 & 59.81 & 0.1 & 10 & 70 + & 150 & 0.01 & 1.41 & 550 & 96.76 & 0.1 & 10 & 10 + & 300 & 0.005 & 0.55 & 210 & 36.94 & 0.05 & 80 & 80 + & 250 & 0.005 & 0.73 & 280 & 49.26 & 0.1 & 5 & 15 + & 200 & 0.005 & 0.81 & 320 & 56.29 & 0.1 & 5 & 5 + & 150 & 0.005 & 0.88 & 400 & 70.37 & 0.1 & 10 & 5 + & 100 & 0.005 & 0.92 & 430 & 75.64 & 0.1 & 15 & 70 + & 300 & 0.01 & 1.92 & 220 & 39.25 & 0.05 - 0.2 & 10 & 5 + & 250 & 0.01 & 2.02 & 300 & 53.53 & 0.05 & 10 & 5 + & 150 & 0.01 & 2.16 & 500 & 89.22 & 0.05 & 10 & 5 + & 100 & 0.01 & 2.21 & 450 & 80.30 & 0.05 & 10 & 5 + _ note _ : the error bars are around 10@xmath66 .
the second relaxation mode that influence the fmr linewidth is the two magnon scattering .
the uniform mode can couple with degenerate spin - wave modes due to fluctuations in the local effective field that can arise from surface defects , scattering centres , fluctuation in the anisotropy from grain to grain or other inhomogeneities@xcite .
the two magnon scattering rate @xmath41 depends on the angle @xmath19 ( out - of - plane geometry ) and on the resonance field h@xmath81 . a detailed analysis based on the effect of the defects on the response functions of thin films was performed in @xcite and @xcite for the case when the magnetisation is tipped out - of - plane .
we consider here the same type of angular dependence of @xmath41 as in @xcite ( see ( 8) ) .
@xmath41 depends on the nature and shape of the defects that activate the scattering mechanism .
the values for the mn@xmath0ge@xmath1compounds , extracted from the fitting of the linewidth curves , are shown in table [ table2 ] as a function of temperature . from the calculated value @xmath77=8h@xmath82
, the exchange spin - wave stiffness d can be inferred if details of the defects are known such as the covered fraction of the surface @xmath83 or the effective height @xmath84 ( h@xmath85 the anisotropy field ) .
atomic force microscopy measurements were performed on the samples , from which the rms surface roughness was determined : for mn@xmath0ge@xmath1the surface roughness was about 1.5 - 2 nm , while for mn@xmath0ge@xmath1c@xmath8it was close to 1 nm . therefore , at room temperature , the spin - wave stiffness was estimated as 0.12@xmath11 g cm@xmath16 for mn@xmath0ge@xmath1 , 0.16@xmath11 g cm@xmath16 for mn@xmath0ge@xmath1c@xmath2and 0.39@xmath11 g cm@xmath16 for mn@xmath0ge@xmath1c@xmath3considering a defect ratio of 50@xmath66 .
these values are only estimates as a precise identification of the defects is difficult to obtain .
as observed from table [ table2 ] , the other extrinsic contributions to the linewidth have only a small impact on the fitted curves .
the mosaicity is very small , inferior to 0.1@xmath9 , being a testimony of the good quality of our samples .
also the inhomogeneity of the internal fields is almost negligible in the majority of cases , only for mn@xmath0ge@xmath1c@xmath2at room temperature it seems to have a larger influence .
the higher values of h@xmath87 are needed to explain the small peak observed around @xmath19 = 0@xmath9 for both mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath2and for the increase of the linewidth at intermediate angles until @xmath19 = 90@xmath9 for mn@xmath0ge@xmath1c@xmath2at room temperature .
the values of the residual inhomogeneous contribution are generally small , the larger values can also be attributed to a temperature dependent intrinsic contribution as discussed above .
+ we now discuss the case of mn@xmath0ge@xmath1c@xmath3 , for which both out - of - plane and in - plane data were fitted , as shown in figure [ fig.4 ] .
the panel ( a ) show the out - of - plane dependence of the fmr linewidth .
the 300k and 250k data are well fitted close to @xmath19 = 0@xmath9 and at larger angles but not at the peaks that correspond to the largest interval between * m * and * h * ( critical angle ) .
the dashed line at t=300k corresponds to a fit with the parameters indicated in table [ table2 ] and @xmath79 = 0.05@xmath9 , while the full line to a fit with @xmath79 = 0.2@xmath9 .
although increasing the mosaicity contribution fits better the peaks , the fitted curve becomes v - shaped between the peaks in total contradiction with the data .
we believe that the mosaicity is small ( 0.05@xmath9 ) and the discrepancy at the critical angle at 300k is due to some other effect ( the fmr line being strongly distorted at this angle ) . we also tried to fit the 300k curve introducing in - plane second and fourth order anisotropy constants ( k@xmath88 and k@xmath89 ) without a better result ( not shown ) .
the low temperature curves are nicely fitted with the presented model for all angles . *
8c t(k ) & @xmath90(oe ) & @xmath91(oe ) & @xmath92(oe ) & @xmath93(oe ) & @xmath94 & @xmath95 & @xmath96 + 300 & 72.75 & 1.5 & & 1.5 & 90 & & 30 + 250 & 97.5 & 1.7 & & 1.5 & 90 & & 30 + 150 & 254.2 & 8.6 & 5.5 & & 57 & 166 & + 100 & 291.4 & 12.4 & 8.6 & & 57 & 167 & + for the in - plane dependence of @xmath67 , the only contributions that were considered were from the isotropic intrinsic damping and the two - magnon contribution which was expressed as follows@xcite : @xmath97 with @xmath98 and @xmath99 characterise the anisotropy of the two - magnon scattering along different crystallographic in - plane directions . at 300k and 250k ( figure [ fig.4](b ) ) , the fmr linewith has the same six - fold symmetry as the angular dependence of @xmath100 ( figure [ fig.1](e ) ) .
if the scattering centres are given by lattice defects ( dislocation lines ) , the azimuthal dependence should reflect the lattice symmetry@xcite .
the angular dependence of the scattering was fitted with @xmath101 at 250k and 300k and with @xmath102 at 150k and 100k .
the parameters @xmath103 and @xmath104 are phenomenologically introduced to account for the observed angular variation .
@xmath105 is expected from the sixfold symmetry .
the in - plane anisotropies are very small as observed form their values in table [ table1 ] , therefore @xmath106 and the dragging function is very close to unity and neglected .
a change of symmetry of the scattering ( or of the dominant contribution to the scattering ) seems to take place around 200k corresponding to the minimum in figure [ fig.2 ] . at lower temperature
a superposition of twofold and fourfold symmetry dominates the angular dependence of the in - plane linewidth .
this can be related to the shape and orientation of the defects ( rectangular ) that cause the two - magnon scattering@xcite .
the two - magnon scattering intensity depends on the direction of the magnetisation in respect to the symmetry axes of the magnetic defects and on the angle between the magnetisation and the crystallographic axes .
one can think that the twofold and fourthfold symmetry can be related to the magnetic symmetry of the defects and to the lattice defect symmetry at low temperature with the defects oriented mostly along the [ 110 ] direction .
this is also consistent with the increase in the two - magnon scattering rate ( see table [ table2 ] and table [ table3 ] ) at lower temperature and with the variation in the two - magnon linewidth contribution to the total linewidth ( from around 30@xmath66 at 300k to 66@xmath66 at 100k ) .
12 nm thick mn@xmath0ge@xmath1and mn@xmath0ge@xmath1c@xmath8films were grown by reactive deposition epitaxy on ge(111 ) substrates .
detailed fmr measurements were performed on the samples at different temperatures .
both mn@xmath0ge@xmath1and c - doped mn@xmath0ge@xmath1show perpendicular uniaxial magneto - crystalline anisotropy and an in - plane easy axis of magnetisation due to the large shape anisotropy . doping with carbon increases the anisotropy fields without diminishing the quality of the sample as demonstrated by the small fmr linewidth of the films . from the angular dependence of the resonance field and of the linewidth , the anisotropy fields , _
g_-factor and magnetic relaxation parameters are obtained .
the contributions to the broadening of the fmr linewidth come primarily from the intrinsic gilbert damping and two - magnon scattering . in the temperature dependence of the out - of - plane
linewidth a very shallow and wide minimum is observed close to 200 @xmath51 40k .
one can suppose that this minimum and the changing of the dominating contribution in the two - magnon scattering from centres with different symmetry are connected one to each other .
nevertheless this supposition requires additional evidences .
this work has been carried out thanks to the support of the a*midex project ( no .
anr-11-idex-0001 - 02 ) funded by the `` investissements davenir '' french government program , managed by the french national research agency ( anr ) .
this work is part of a*midex through the hit project apodise .
we also want to thank the interdisciplinary french epr network renard ( cnrs - fr3443 ) .
26 i. zutic , j. fabian , and s. d. sarma , rev . mod
. phys . * 76 * , 323 ( 2004 ) .
d. d. awschalom and m. e. flatt , nature phys .
* 3 * , 153 ( 2007 ) .
g. schmidt , d. ferrand , l. w. molenkamp , a. t. filip , and b. j. van wees , phys .
b * 62 * , r4790 ( 2000 ) . c. zeng , s. c. erwin , l. c. feldman , a. p. li , r. jin , y. song , j. r. thompson and h. h. weitering , app . phys
. lett . * 83 * , 5002 ( 2003 ) .
m. gajdzik , c. srgers , m. kelemen , and h. v. lhneysen , j. magn .
. mater . * 221 * , 248 ( 2000 ) .
r. p. panguluri , c. zeng , h. h. weitering , j. m. sullivan , s. c. erwin , and b. nadgorny , phys .
status solidi b * 242 * , r67 ( 2005 ) .
s. picozzi , a. continenza , and a. j. freeman , phys .
b * 70 * , 235205 ( 2004 ) . c. srgers , g. fischer , p. winkel , and h. v. lhneysen , phys . rev .
b * 90 * , 104421 ( 2014 ) .
s. olive - mendez , a. spiesser , l.a .
michez , v. le thanh , a. glachant , j. derrien , t. devillers , a. barski , m. jamet , thin solid films * 517 * , 191 ( 2008 ) . m. petit , l. michez , c .- e .
dutoit , s. bertaina , v. o. dolocan , v. heresanu , and v. le thanh , thin solid films * 589 * , 427 ( 2015 ) . l. simon , m. stoffel , p. sonnet , l. kubler , l. stauffer , a. selloni , a. de vita , r. car , c. pirri , g. garreau , d. aubel , and j. l. bischoff , phys . rev .
b * 64 * , 035306 ( 2001 ) . c. zeng , w. zhu , s. c. erwin , z. zhang , and h. h. weitering , phys . rev .
b * 70 * , 205340 ( 2004 ) .
m. farle , rep .
. phys . * 61 * , 755 ( 1998 ) . c. chappert , k. l. dang , p. beauvillain , h. hurdequint , and d. renard , phys .
b * 34 * , 3192 ( 1986 ) .
w. platow , a. n. anisimov , g. l. dunifer , m. farle , and k. baberschke , phys .
b * 58 * , 5611 ( 1998 ) .
s. mizukami , y. ando , and t. miyazaki , jpn .
phys . * 40 * , 580 ( 2001 ) .
t. l. gilbert , ieee trans .
magn . * 40 * , 3443 ( 2004 ) .
k. zakeri , j. lindner , i. barsukov , r. meckenstock , m. farle , u. von hrsten , h. wende , w. keune , j. rocker , s. s. kalarickal , k. lenz , w. kuch , k. baberschke , and z. frait , phys .
b * 76 * , 104416 ( 2007 ) .
r. arias and d. l. mills , phys .
b * 60 * , 7395 ( 1999 ) .
p. landeros , r. e. arias , and d. l. mills , phys .
b * 77 * , 214405 ( 2008 ) .
r. d. mcmichael , d. j. twisselmann , and a. kunz , phys .
lett . * 90 * , 227601 ( 2003 ) .
s. s. kalarickal , p. krivosik , j. das , k. s. kim , and c. e. patton , phys .
b * 77 * , 054427 ( 2008 ) .
k. gilmore , y. u. idzerda , and m. d. stiles , phys .
lett . * 99 * , 027204 ( 2007 ) .
h. ebert , s. mankovsky , d. kdderitzsch , and p. j. kelly , phys .
lett . * 107 * , 066603 ( 2011 ) .
v. l. safonov , j. appl . phys . * 91 * , 8653 ( 2002 ) . c. vittoria , s. d. yoon , and a. widom , phys .
rev b * 81 * , 014412 ( 2010 ) . yi li and k. baberschke , phys .
* 68 * , 1208 ( 1992 ) .
j. lindner , i. barsukov , c. raeder , c. hassel , o. posth , r. meckenstock , p. landeros , and d. l. mills , phys .
rev b * 80 * , 224421 ( 2009 ) . | ferromagnetic resonance ( fmr ) was used to investigate the static and dynamic magnetic properties of carbon - doped mn@xmath0ge@xmath1(c@xmath2 and c@xmath3 ) thin films grown on ge(111 ) . the temperature dependence of magnetic anisotropy shows an increased perpendicular magneto - crystalline contribution at 80k with an in - plane easy axis due to the large shape contribution .
we find that our samples show a small fmr linewidth ( corresponding to an intrinsic magnetic damping parameter @xmath4=0.005 ) , which is a measure of the spin relaxation and directly related with the magnetic and structural quality of the material . in the perpendicular - to - plane geometry
, the fmr linewidth shows a minimum at around 200k for all the samples , which seems to be not correlated to the c - doping .
the magnetic relaxation parameters have been determined and indicate the two - magnon scattering as the main extrinsic contribution .
we observe a change in the main contribution from scattering centres in mn@xmath0ge@xmath1c@xmath3at low temperatures , which could be related to the minimum in linewidth .
_ keywords _ : ferromagnetic resonance , magnetic anisotropy , magnetic damping , thin films |
isobaric spin was introduced by heisenberg in 1932 @xcite in order to explain the neutron - proton symmetry .
subsequently , the isospin symmetry has been widely used in theoretical modeling of the atomic nuclei @xcite . even though it is broken by the electroweak force , isospin is extremely useful for our understanding of nuclear structure and decays . of particular interest for isospin physics
are @xmath0 nuclei . beyond @xmath1ca ,
many of them are located far from the line of beta stability , in close proximity to the proton drip line . because of the similarity of proton and neutron shell - model orbits , the @xmath0 systems exhibit unique phenomena related to the attractive nature of the isoscalar component of the nuclear force .
examples are : superallowed fermi beta decays @xcite , superallowed gamow - teller decays @xcite , superallowed alpha decays @xcite , wigner energy @xcite , isoscalar pairing @xcite , and collective modes @xcite .
the atomic nuclei with enhanced sensitivity to fundamental symmetries are unique laboratories to search for signals of new physics beyond the standard model .
the @xmath0 nuclei are particularly interesting probes as the superallowed @xmath2=0@xmath3@xmath4@xmath2=0@xmath3 beta - decays among the isobaric analogue states in the isospin triplet allow stringent tests of the conserved vector current hypothesis and provide precise values of the strength of the weak force and the leading @xmath5 element of the cabbibo - kobayashi - maskawa matrix .
other examples include fermi- and gamow - teller ground - state beta - decays in @xmath6=1/2 mirror nuclei , which offer alternative tests of the electroweak sector . in all these cases ,
high - fidelity theoretical calculations of radiative corrections and isospin symmetry breaking ( isb ) effects are needed to extract the crucial information from precise measurements .
the theory roadmap for this area involves ab - initio and configuration interaction ( shell model ) approaches , and nuclear density functional theory ( dft ) .
the latter is the tool of choice for open - shell , deformed complex systems .
the focus of this paper is on dft - based frameworks and their extensions , including multi - reference approaches involving symmetry restoration .
such models are able to capture core - polarization effects originating from a subtle interplay between the long - range coulomb interaction and short - range hadronic inter - nucleon forces ; hence , they are particularly useful to study various aspects of electroweak interaction .
the aim of this overview is to present selected aspects of isospin physics in @xmath7 nuclei in terms of theoretical approaches rooted in nuclear density functional theory .
we shall start in sec .
[ sect.02 ] by covering the topic of isospin impurities , which reflect the degree of isospin symmetry breaking .
they are responsible for isospin - forbidden rare decay modes and impact electroweak matrix elements of decays , which probe fundamental symmetries .
two examples of such decays will be discussed : the @xmath8 superallowed beta - decays in sec .
[ sect.03 ] and fermi- and gamow - teller ground - state beta - decays in @xmath6=1/2 mirror nuclei in sec .
[ sect.04 ] .
section [ sect.05 ] is devoted to the mirror and triplet displacement energies sensitive to hadronic interactions , which break charge - symmetry and charge - independence .
the phenomenon of proton - neutron pairing is covered in sec .
[ sect.06 ] in the context of a generalized dft approach that is based on proton - neutron mixed orbitals . finally , perspectives are given in sec .
[ sect.07 ] .
the degree of isospin symmetry violation the isospin impurity is a result of a subtle balance between the attractive short - range strong force and the repulsive long - range coulomb interaction that polarizes the entire nucleus .
consequently , its precise theoretical treatment requires the use of a _ no - core _ framework which , in heavier nuclei , is provided by nuclear dft . the early attempts to evaluate the degree of isospin breaking , measured in terms of the isospin impurity @xmath9 , date back to the 1960s , see ref .
@xcite for a review .
these estimates , based on perturbation theory @xcite or analytically solvable hydrodynamical model @xcite , were able to explain some gross features of @xmath9 such as the steady increase of isospin mixing along the @xmath10 line with increasing @xmath11 or quenching of @xmath9 with increasing @xmath12 .
these early models , however , were not too quantitative .
isospin impurities @xmath9 in even - even @xmath10 nuclei calculated with sly4 edf @xcite . full ( open )
triangles mark the ar ( br ) values , respectively .
note that , the mean - field estimates ( br ) are quenched by @xmath1330% with respect to the results of rediagonalization .
stars mark empirical values in @xmath14ge @xcite ( only the lower bound is known in this case ) and @xmath15zr @xcite .
( taken from ref @xcite . ) ] the accurate calculation of the isospin impurities has been challenging .
this was early realized by engelbrecht and lermer @xcite who pointed out that the self - consistent mean - field ( mf ) approaches can not be directly applied because of the spurious mixing caused by the spontaneous isospin - symmetry - breaking effects . to eliminate the problem of spurious admixtures in the wave function ,
we have developed a no - core mr - dft model involving isospin projection , see also refs .
the model employs self - consistent , isospin - broken mf states @xmath16 .
self - consistency is needed to ensure that the balance between the coulomb force and the strong interaction , represented in our model by the skyrme energy density functional ( edf ) , are properly taken into account .
the mf state can be formally decomposed into good - isospin basis @xmath17 : @xmath18 where @xmath6 and @xmath19 are the total isospin and its third component , respectively , the mixing coefficients @xmath20 can be calculated using the states @xmath21 obtained by isospin projection after variation . to assess the isospin mixing
, one has to rediagonalize the total hamiltonian @xmath22 involving strong interaction plus the coulomb term in the space spanned by the good - isospin basis ( [ tbasis ] ) : @xmath23 enumerates the eigenstates @xmath24 of @xmath22 .
the value of @xmath25 , corresponds to the isospin - mixed ground state ( g.s . ) .
the g.s .
isospin - mixing parameter obtained after rediagonalization ( ar ) is defined as : @xmath26 comparison of @xmath9 with the quantity @xmath27 obtained before rediagonalization ( br ) provides direct information about the spuriosity of mf solutions . as shown in fig .
[ fig.1 ] , the isospin impurities exceed mf values by almost 30% .
the empirical information on isospin impurities in heavier systems is both scarce and uncertain .
the values of @xmath9 extracted from a forbidden e1 transition in @xmath14ge @xcite or from the decay of giant dipole resonance in @xmath15zr @xcite are consistent with the predictions of the isospin - projected dft but the experimental uncertainties are too large to discriminate between different edf parameterizations . indeed , as shown in fig .
[ fig.2 ] , the spread in @xmath28 obtained with nine skyrme edfs , @xmath29 , is relatively small and lies well within the experimental uncertainty limits .
the figure shows the calculated impurities versus excitation energy @xmath30 for the lowest @xmath31 state in even - even @xmath10 nuclei , which is referred to as the _ doorway state_. it is seen that the larger excitation energy of the doorway state the smaller impurity , with exception of the sko result .
calculated isospin impurities in @xmath15zr versus excitation energy of the doorway state .
the results were obtained using the isospin projected dft formalism with nine skyrme edfs . filled ( open ) circles refer to ar ( br ) results , respectively .
the ar results lie within experimental uncertainty limits of ref .
@xcite while the br values are ruled out by the data .
( taken from ref . @xcite . ) ] figure [ fig.3 ] shows the calculated excitation energy of the doorway state for the three skyrme edfs : siii @xcite , sly4 @xcite , and skp @xcite .
the self - consistent values of the excitation energies of the doorway state are compared with the estimate @xmath32mev of ref .
@xcite based on the perturbation theory , and with the hydrodynamical estimate @xmath33mev of ref .
it is seen that the simple estimates of @xmath30 vastly differ from the self - consistent results , both in terms of the magnitude and @xmath11-dependence .
this indicates that the coulomb mixing is a highly non - perturbative effect that requires self - consistent treatment of the interplay between short- and long - range interactions .
that is why back - of - the - envelope predictions of @xmath9 are so unreliable .
excitation energies of the doorway states @xmath34 in the even - even @xmath35=@xmath36 nuclei relative to the g.s .
energies @xmath37 obtained with siii ( diamonds ) , sly4 ( dots ) , and skp ( circles ) skyrme edfs .
horizontal lines mark the mean values .
the estimates based on the perturbation theory @xcite and the hydrodynamical model @xcite are indicated by thick dotted lines .
( adopted from ref .
@xcite . ) ] isospin impurity @xmath38 in @xmath39sn calculated for several skyrme edfs plotted as a function of the isoscalar ( top ) and isovector ( bottom ) effective mass . ]
excitation energies of the doorway states and , in turn , isospin impurities , depend on edf parameterization .
it is , however , not at all obvious what edf components are responsible for the systematic differences seen in fig .
[ fig.3 ] .
indeed , attempts to correlate @xmath28 with various bulk properties of skyrme edfs turned out to be fairly inconclusive .
for example , no correlation has been found with the symmetry energy @xcite , which is the primary quantity characterizing the isovector properties of edfs . as illustrated in fig .
[ fig.4 ] , no clear correlation has been found between @xmath9 and the isovector and isoscalar effective mass .
coulomb impurities in @xmath1ca versus a difference of mean proton and neutron radii .
upper plot shows mean - field values @xmath40 .
lower plot shows true impurities @xmath38 . ]
we do find , however , a strong correlation between @xmath40 and the proton skin , i.e. , the difference between proton and neutron root mean square radii , see fig .
[ fig.5 ] and ref .
the correlation deteriorates for @xmath38 , indicating again that the isospin mixing is a non - perturbative quantity ; hence difficult to estimate .
the formalism discussed above can be extended to incorporate the angular - momentum projection . in this case , the basis is created by applying the isospin ( @xmath41 ) and angular - momentum ( @xmath42 ) projection to @xmath16 : @xmath43 where @xmath44 and @xmath45 denote the magnetic quantum numbers associated with the laboratory and intrinsic @xmath46-axes , respectively @xcite . because @xmath45 is not conserved
, the set ( [ itbasis ] ) is overcomplete .
this problem can be overcome by rediagonalization of the hamiltonian in the so - called _ collective space _ , spanned for each @xmath2 and @xmath6 by the _ natural states _
, @xmath47 @xcite .
the wave functions @xmath48 obtained in the rediagonalization are labeled by the index @xmath49 and conserved quantum numbers @xmath2 , @xmath44 , and @xmath50 .
left : schematic illustration of two possible mean - field g.s .
configurations in an odd - odd @xmath35=@xmath36 nucleus .
upper ( lower ) configuration is called aligned ( anti - aligned ) .
right : the result of the isospin symmetry restoration .
the aligned configuration is isoscalar ; hence , it is insensitive to the isospin projection .
the anti - aligned configuration represents a mixture of @xmath6=0 and @xmath6=1 states .
the projection lifts the isospin degeneracy by lowering the @xmath6=0 level .
( taken from ref .
@xcite . ) ] the double - projected and isospin - projected approaches yield very similar g.s .
isospin impurities in even - even nuclei .
however , there are configurations for which the model solely relying on isospin projection is insufficient . among them
there are the so - called anti - aligned configurations in odd - odd @xmath10 nuclei , which are crucial for calculations of isb corrections to the superallowed beta - decay involving the @xmath51 states in odd - odd nuclei .
these states are not representable by a single mf configuration as shown schematically in fig .
[ fig.6 ] . indeed
, the odd proton and odd neutron can form either the aligned or anti - aligned configuration .
the aligned configuration represents an isoscalar @xmath52 state .
the anti - aligned configuration , however , manifestly breaks the isospin symmetry being an equal mixture of isoscalar and isovector states .
the calculations indicate @xcite that the @xmath6=0 and @xmath6=1 components projected from the anti - aligned configuration strongly mix through the coulomb interaction leading to unphysically large isospin impurities .
double - projected approach is free from such pathologies .
this is demonstrated in the lower part of fig .
[ fig.7 ] for a representative case of @xmath53sc .
isospin impurities in @xmath53sc , calculated for four antialigned configurations @xmath54 obtained by putting the valence neutron and proton in opposite-@xmath45 nilsson orbitals originating from the @xmath55 shell .
open and full dots show the results obtained in isospin - projected and double - projected ( @xmath56 ) variants , respectively .
( adopted from ref .
@xcite . ) ] it can be shown @xcite that the isospin projection technique can be safely used within nuclear dft . augmenting the isospin projection with the angular - momentum projection leads to a number of theoretical issues
. not only the numerical complexity of computations increases but also the uncontrolled singularities appear in the energy kernels @xcite ; this essentially eliminates a possibility to use modern edfs with density - dependent terms @xcite . to overcome this problem ,
regularization schemes have been proposed @xcite but they are difficult in practical implementations . until a workable solution is developed , a practical option is to use hamiltonian - based edfs , such as the sv force @xcite augmented by the tensor terms , or the recently developed slymr forces that include three- and four - body terms @xcite .
all the existing parameterizations of the density - independent skyrme edf are characterized by a nonstandard saturation mechanism driven by a very low isoscalar effective mass @xmath57 .
the low effective mass affects the overall performance of those forces , impairing such key properties as the symmetry energy , level density , and level ordering .
the low value of @xmath58 also impacts the isospin mixing .
in particular , in the case of @xmath15zr discussed above , the sv edf yields @xmath59% , which is considerably smaller than the mean value , @xmath60% , obtained by averaging over the edfs shown in fig .
[ fig.2 ] .
the lack of a reasonable hamiltonian - based skyrme edf is probably the most critical deficiency of the current self - consistent approach .
nonetheless , one has to admit that the edfs with low effective masses perform surprisingly well when used in the context of no - core - configuration - interaction ( ncci ) extensions @xcite
the superallowed @xmath61 beta transitions are of particular importance for testing various aspects of the electro - weak sector of the standard model .
what makes these pure fermi decays so useful is the conserved vector current ( cvc ) hypothesis , which states that the vector current is not renormalized in the nuclear medium .
this implies that the product of the statistical rate function @xmath62 and partial half - life @xmath63 for the superallowed @xmath64 fermi beta - decay should be nucleus - independent and equal to : @xmath65 where @xmath66gev@xmath67s is a universal constant ; @xmath68 stands for the vector coupling constant for semi - leptonic weak interaction ; and @xmath69 is the nuclear matrix element of the isospin operator @xmath70 . in reality , the relation ( [ ft ] ) holds only approximately and must be slightly amended to account for radiative processes and isospin - symmetry breaking effects .
fortunately , the radiative and isb corrections are small , of the order of a few percent , and this allows us to express the rate conveniently as : @xmath71 with the left - hand side being nucleus independent . in eq .
( [ ftnew ] ) , @xmath72% stands for the nucleus - independent part of the radiative correction @xcite , @xmath73 is a transition - dependent ( @xmath36-dependent ) but nuclear - structure - independent part of the radiative correction @xcite , and @xmath74 denotes the nuclear - structure - dependent part of the radiative correction @xcite .
the isb correction @xmath75 is a many - body correction to the nuclear matrix element : latexmath:[\[\label{mfa } isospin - symmetry violation in the atomic nuclei , see refs .
@xcite and references cited therein . in spite of explicit dependence on theoretical input and the related uncertainties in the calculated radiative and isb corrections , the superallowed beta
- decays provide today s most accurate value of the vector coupling constant @xmath68 .
moreover , they enable a very accurate verification of the cvc hypothesis and provide precise information on the leading matrix element @xmath77 of the ckm three - generation quark mixing matrix @xcite .
this is so because the leptonic coupling constant , @xmath78 gev@xmath79 , is well known from the muon decay @xcite .
the uncertainty of @xmath80 extracted from the superallowed beta - decays is almost by an order of magnitude smaller as compared to the values obtained from neutron or pion decays @xcite . in search for physics beyond the standard model ,
precision is of utmost importance .
only those transitions that have @xmath81-values measured with an accuracy @xmath82% are acceptable .
canonical _ set of transitions used over the last decade to test the standard model consists of 13 such cases spreading over a wide range of masses from @xmath11=10 ( @xmath83ca ) to @xmath11=74 ( @xmath84rb ) , see @xcite .
with the cvc hypothesis confirmed , it is possible to extract @xmath80 by averaging over several transitions .
this feature makes the superallowed beta - decay strategy very attractive .
recently , the canonical set has been extended by a measurement of the beta - decay of @xmath85ca @xcite .
moreover , the superallowed decay of @xmath86ne is within experimental reach @xcite .
the isb corrections @xmath75 were computed in diverse nuclear models , see refs .
the standard in this field has been set by hardy and towner ( ht ) @xcite who have used the nuclear shell model to account for the configuration mixing effect in conjunction with the mean - field potential needed to account for a radial mismatch of proton and neutron single - particle ( s.p . )
wave functions due to the long - range coulomb polarization @xcite .
such a description has certain practical advantages .
for instance , it allows for an independent fine - tuning of various model s ingredients .
but it also leads to internal inconsistencies .
in particular , the ht model violates the su(2 ) commutation rules for the bare isospin operators as pointed out in refs .
unfortunately , the associated uncertainty is difficult to quantify .
the self - consistent multi - reference dft involving the isospin- and angular - momentum projections @xcite is an interesting alternative to the fine - tuned phenomenological ht approach .
this is a _ no - core _ theory , which is capable of treating rigorously the rotational symmetry and explicit breaking of isospin symmetry .
the correct balance between the coulomb and hadronic forces is maintained by self - consistency requirements .
the approach allows for a rigorous treatment of the fermi matrix elements using bare isospin operators .
the recently proposed ncci extensions of the framework @xcite allow to take into account more correlations due to inclusion of ( multi)particle-(multi)hole configurations .
as already mentioned , the weakest point of the formalism is the lack of a reasonable hamiltonian - based skyrme edf .
one should stress , however , that @xmath87 does not depend on the absolute magnitude of isospin mixing but rather on a difference between parent and daughter states @xcite , i.e. , it is a less sensitive quantity .
the mr - dft calculations of @xmath75 involving single reference states in parent and daughter nuclei are carried out as follows .
the superallowed @xmath88 fermi beta - decay proceeds between the g.s .
@xmath89 of the even - even nucleus and its isospin - analogue partner @xmath90 in the @xmath10 odd - odd nucleus .
the corresponding transition matrix element is : @xmath91 within the mr - dft , the parent g.s .
is approximated by a projected state @xmath92 where @xmath93 is the g.s . of the even - even nucleus
obtained in the self - consistent hartree - fock calculations . the state @xmath93 is unambiguously defined by filling in the pairwise doubly - degenerate levels of protons and neutrons up to the fermi level .
the daughter state is approximated by @xmath94 where the self - consistent slater determinant @xmath95 ( or @xmath96 ) represents the anti - aligned configuration , selected by placing the odd neutron and the odd proton in the lowest available time - reversed ( or signature - reversed ) s.p .
orbits . the isospin - projected state shown in fig .
[ fig.6 ] , based on the @xmath16 configuration that manifestly breaks the isospin symmetry , is indeed a preferred way to access the @xmath97 states in odd - odd @xmath10 nuclei .
the anti - aligned configurations in odd - odd @xmath10 nuclei are not uniquely defined . in the signature - symmetry - restricted calculations ,
there are in general three anti - aligned slater determinants with the s.p .
angular momenta ( alignments ) of the valence protons and neutrons pointing along the @xmath98 , @xmath99 , or @xmath100 axes of the intrinsic shape defined by means of the long ( @xmath100 ) , intermediate ( @xmath98 ) , and short ( @xmath99 ) principal axes of the nuclear mass distribution , respectively .
thus far , in spite of persistent efforts , no self - consistent tilted - axis g.s .
solutions have been found .
various properties of these linearly dependent solutions , hereafter referred to as @xmath101 , @xmath102 , and @xmath103 , are discussed in detail in ref .
@xcite and will not be repeated here . in a mr - dft double - projection approach based on a single reference state ,
the only way to deal with the shape - current orientation ambiguity is to calculate three independent beta - decay matrix elements corresponding to each orientation and average over the resulting values of @xmath75 .
such an strategy was adopted in ref .
@xcite , which contains the calculated corrections for the canonical set of superallowed transition as well as predictions for yet - unmeasured cases .
the calculated corrections are in reasonable agreement with the ht results as shown in fig .
[ fig.8 ] .
both sets of corrections systematically overestimate the rpa results of ref .
@xcite .
differences between the isb corrections to 12 precisely measured superallowed @xmath8 decays ( excluding @xmath11=38 ) calculated in refs .
@xcite ( sv ) and @xcite ( ht ) .
shaded band shows the combined sv+ht error .
( taken from ref .
@xcite . ) ] with the set of isb corrections obtained in ref .
@xcite , one obtains @xmath104s , which gives @xmath105 .
this value is in excellent agreement with the ht result @xcite , @xmath106 , and the central value obtained from the neutron decay , @xmath107 @xcite .
taking the calculated @xmath108 and combining it with @xmath109 and @xmath110 from the 2010 particle data group @xcite , one obtains @xmath111 which implies that the unitarity of the first row of the ckm matrix is satisfied with a precision better than 0.1% .
a survey of the @xmath108 values obtained using different methods is shown in fig . [ fig.9 ] . a survey of the @xmath108 values obtained using different methods : ( a ) ref .
@xcite ; ( b ) ref .
@xcite ; ( c ) ref .
@xcite with sv ; ( d ) ref .
@xcite with shz2 .
the values obtained form pion - decay @xcite , neutron - decay @xcite , and beta - decays in @xmath6=1/2 mirror nuclei @xcite are shown for comparison .
( taken from ref .
@xcite . ) ] the solutions @xmath101 , @xmath102 , and @xmath103 , differ by at most a few hundred kev in energy .
hence , there is no obvious preference for a choice of reference state .
moreover , the orientation - dependent effects originate , predominantly , from the time - odd fields of the nuclear mf .
hence , the orientation effects are present only in odd - odd nucleus adding up to a difference between parent and daughter nuclei . the averaging procedure applied in ref .
@xcite should be considered as a purely practical solution .
the shape - current - orientation ambiguity has motivated us to extend the formalism to allow for a dynamical mixing of states that are projected from low - lying ( multi)particle-(multi)hole self - consistent mf configurations . in the context of the @xmath75 calculations ,
the idea is to mix @xmath112 states projected from @xmath101 , @xmath102 , and @xmath103 configurations , respectively .
in such an approach , the @xmath2=0 wave functions in an odd - odd @xmath35=@xmath36 nucleus are approximated by : @xmath113 where the coefficients @xmath114 are obtained by solving the hill - wheeler - griffin ( hwg ) equation , and @xmath49 enumerates the hwg eigenstates .
the hwg equation has , typically , three linearly independent solutions which , instead of @xmath49 , can be conveniently labeled by approximate isospin quantum number : @xmath115 .
comparison of the isb corrections to the 12 canonical superallowed @xmath116 transitions calculated in ref .
@xcite ( open symbols ) and in ref .
@xcite ( filled symbols ) . in the latter calculations , the @xmath112 states projected from @xmath101 , @xmath102 , and @xmath103 configurations were mixed dynamically by solving the hwg equation . in the former ,
matrix elements were calculated independently for each orientation and averaged afterwards . ] the new set of the isb corrections calculated using this prescription is displayed in fig .
[ fig.10 ] ; the detailed values can be found in ref .
the improved corrections are somewhat smaller that the values of ref .
@xcite obtained by averaging over orientations .
this difference , however , has almost no influence on @xmath117 and @xmath108 .
indeed the new values @xmath118s and @xmath119 , calculated with the updated set of experimental @xmath81 values taken from ref .
@xcite , almost perfectly match the previous numbers @xcite .
towner and hardy @xcite proposed a confidence level ( cl ) test to check a consistency of the calculated isb corrections .
the underlying assumptions are : ( _ i _ ) validity of the cvc hypothesis up to at least @xmath1200.03% and ( _ ii _ ) validity of the calculated nuclear - structure - dependent corrections @xmath74
these two assumptions allow to calculate _
empirical _ corrections : @xmath121 by treating the value of @xmath117 as an adjustable parameter , one can bring @xmath122 as close as possible to the calculated @xmath75 by minimizing the appropriate @xmath123 . with the new set of corrections calculated in ref .
@xcite , the @xmath123 per degree of freedom ( @xmath124=11 ) drops from @xmath125=10.2 @xcite to 6.3 .
this number is still much larger than the values reported in ref .
@xcite , which are : 1.7 for the damgrd model @xcite , 0.4 for the shell - model with woods - saxon wave functions @xcite , 2.3 for the shell - model with hartree - fock wave functions @xcite , and 2.1 for the relativistic dft+rpa model of ref .
the main contribution to the @xmath123-value in our model can be associated with the sudden increase in @xmath75 due to a single @xmath126ga@xmath4@xmath126zn transition , which contributes more than 62% to the total @xmath123 budget .
the @xmath6=1/2 mirror nuclei offer an alternative way to test the cvc hypothesis @xcite .
these nuclei decay via the mixed fermi and gamow - teller ( gt ) transitions .
hence , apart from the radiative and the isb corrections , the values of @xmath127 and @xmath5 also depend on the ratio of statistical rate functions @xmath128 for the axial - vector and vector interactions , and the ratio @xmath129 of nuclear matrix elements , where @xmath130 denotes the ratio of axial - vector and vector coupling constants .
the cvc hypothesis implies that the vector coupling constant is @xmath131 .
the axial - vector current is partially conserved ; this implies that the axial - vector coupling constant gets renormalized in the nuclear medium .
the effective axial - vector coupling constant , @xmath132 , is quenched by an @xmath11-dependent factor @xmath133 with respect to the free neutron decay value @xmath134 .
quenching factors deduced from comparison of large - scale nuclear shell model ( nsm ) calculations with experiment are : @xmath135 , 0.77 @xcite , and 0.74 @xcite in the @xmath136- , @xmath137- , and @xmath138-shell region , respectively . in the region
@xmath139 , a large value of @xmath140 has been extracted @xcite , see however @xcite . to account for the mass dependence of @xmath133 in shell - model calculations , a phenomenological expression
@xmath141 has been proposed @xcite for @xmath142 .
the origin of the quenching is not fully understood .
it is usually related to missing correlations in the wave function ; truncation of model space ; and in the context of _ ab initio _ models two - body currents @xcite .
much work in this area has been done in relation to the double - beta decay @xcite and wimp scattering @xcite .
recent study of the ikeda sum rule in @xmath143 decays of @xmath144c and @xmath145o performed in ref .
@xcite estimate quenching due to the two - body currents to be @xmath146 .
this range is consistent with experimental data on @xmath147zr @xcite .
recently , a systematic study of both the gt and fermi ground - state transitions in the @xmath6=1/2 mirror @xmath137- and @xmath138-shell @xmath6=1/2 mirror nuclei with @xmath148 have been carried out using the self - consistent ncci approach @xcite . within the ncci model the wave function reads [ cf .
( [ ktmix ] ) ] : @xmath149 when compared to the isospin- and angular - momentum mr - dft wave - function ( [ ktmix ] ) , the ncci wave function contains additional summation over the configurations ( slater determinants ) @xmath150 . in the present implementation of the formalism
, we use the same skyrme interaction to compute the configurations @xmath150 and to mix the isospin- and angular - momentum - projected states @xmath151 .
the model was tested for the @xmath152he@xmath4@xmath152li gt beta decay .
matrix element for this transition , @xmath153=2.1645(43 ) , is precisely known from the recent measurement @xcite .
figure [ fig.11 ] shows a difference between the calculated and experimental gt matrix element as a function of the ncci configuration space considered .
the first point corresponds to the case of no configuration mixing . here ,
the wave functions @xmath154 and @xmath155 are projected from the optimal ( energy - wise ) slater determinants .
next , keeping the parent wave function @xmath154 fixed , we enrich the configuration space of the daughter nucleus by admixing the @xmath156 states projected from the lowest particle - hole ( @xmath157 ) and the two lowest ( @xmath158 ) particle - hole configurations : @xmath159 .
this causes an increase of the matrix element circa 3% above the experimental value .
the extension of the configuration space of @xmath152he by admixing excited @xmath112 states hardly impacts this result , see fig .
[ fig.11 ] .
this test indicates that ncci is capable of capturing the main features of the wave functions that are important for a reliable reproduction of the gt matrix element .
unfortunately , the model underbinds @xmath152li and overbinds @xmath152he ; hence we have not great confidence in its accuracy for this specific gt transition .
this is , however , not surprising as the current dft approaches and their mr - dft and ncci extensions have not been optimized to experimental data in light nuclei .
gt @xmath152he@xmath4@xmath152li matrix element computed in ncci relative to experiment @xcite as a function of ncci configuration space . ]
theoretical ground - state gt matrix elements in the @xmath6=1/2 mirror @xmath137- and @xmath138-shell nuclei ranging from @xmath11=17 till 55 are shown in fig .
[ fig.12 ] .
the ncci results with the sv@xmath160 edf were obtained using the hfodd solver @xcite that has been augmented with the ncci module .
the details of calculations follow ref . @xcite .
top : ground - state gt matrix elements in @xmath6=1/2 medium - mass mirror nuclei calculated using ncci ( filled circles ) @xcite and shell - model ( sm ) approaches ( open circles ) @xcite .
bottom : a correlation between the theoretical and experimental matrix elements .
solid ( dashed ) line represents a linear fit to the ncci ( sm ) results .
] it is seen that the ncci results are fairly close to shell model predictions @xcite .
the consistency between these two theoretical approaches is visualized in fig .
[ fig.12](bottom ) , which shows that both models correlate well with experimental data . as discussed in ref .
@xcite , this similarity is quite surprising since the models differ in many aspects .
top : gt matrix element for the [email protected] .
transition @xmath161mg@xmath4@xmath161na versus the number of particle - hole configurations taken in the ncci calculation involving angular - momentum projection ( dots ) and isospin- and angular - momentum projection ( circles ) .
bottom : the predicted ikeda sum rule for the g.s .
configuration of @xmath161 mg . ]
the model - independent ikeda sum rule @xcite serves as an important indicator of the quality of theoretical models of gt decay . in the @xmath11=39 , @xmath6=1/2 nuclei , which can be viewed as one - hole systems , inclusion of all possible @xmath157 excitations within the @xmath137-shell space exhausts 99% of the sum rule @xcite
recently , we have performed similar calculations for the @xmath11=23 , @xmath6=1/2 systems , which are complex , open - shell deformed nuclei .
the calculations , involving eight deformed particle - hole configurations , were performed using two variants of the ncci model : the full model involving the isospin- and angular - momentum projection and its simplified variant involving only the angular - momentum projection .
the results are shown in fig .
[ fig.13 ] .
as anticipated , both methods give almost identical matrix elements for the gs transition .
this is because the effect of isospin mixing in the ground states of nuclei considered is very small .
note however , that the variant involving the angular - momentum projection only captures mere 40% of the ikeda sum rule . at the same time , the full variant of the model accounts for the 85% of the sum rule .
the example shows that the proper treatment of isospin in excited states is critical for the sum rule evaluation .
isb corrections to the ground - state fermi transitions in @xmath6=1/2 mirror nuclei obtained in ncci @xcite ( dots ) and the sm+woods - saxon model of ref .
@xcite ( circles ) . ]
the full variant of the model involving double - projection is necessary to compute the isb corrections to the fermi branch of beta - decay in the @xmath6=1/2 mirror nuclei .
these nuclei offer the independent way to verify the cvc hypothesis and to study the ckm matrix @xcite .
a set of the isb corrections calculated with the mr - dft technique was published in ref .
@xcite , and the ncci results can be found in table 1 of ref .
the ncci predictions are shown in fig .
[ fig.14 ] and compared to the sm results of ref .
it is seen that the ncci corrections , although slightly smaller on average , are fairly consistent with sm .
figure [ fig.15 ] shows a difference between these two sets of calculations .
the shaded area marks the error band calculated as @xmath162 under the assumption of 10% error on the ncci results due to a basis cut - off .
difference between isb corrections to the g.s .
fermi decays in @xmath6=1/2 mirror nuclei calculated with ncci @xcite and the sm+woods - saxon model of ref .
the shaded area marks the theoretical uncertainty . ]
strong interaction is , predominantly , invariant with respect to rotations in the isospin space .
this fact is rather well established experimentally and confirmed theoretically .
indeed , most nuclear many - body approaches , including shell - model and dft models , use interactions that are scalar in the isospace .
the skyrme interaction is a prime example of an isospin invariant effective force that has long been used to compute various nuclear properties . on the other hand
, there also exists a firm experimental evidence that the strong interaction contains small isospin - breaking components .
for example , the nucleon - nucleon ( nn ) scattering data reveal small differences in phase shifts and scattering lengths . in the @xmath163 channel ,
the scattering lengths for neutron - neutron ( @xmath164 ) , neutron - proton ( @xmath165 ) , and proton - proton ( @xmath166 ) scattering are : @xmath167fm , @xmath168fm , and @xmath169fm , respectively @xcite .
a detailed analysis of scattering data , in particular phase shifts and scattering lengths , shows that the _ nn _ interaction is 1% stronger than _ pp _ interaction and that the _ np _ interaction is 2.5% stronger than the average of _ nn _ and _ pp _ interactions @xcite .
( these figures include corrections for electromagnetic effects and refer solely to the strong nucleon - nucleon interaction . ) on the fundamental level , the isospin symmetry is broken due to different masses and electromagnetic interactions of @xmath170 and @xmath171 quarks @xcite .
albeit small , isospin violation considerably influences global properties of finite nuclei such as binding energies .
it is well known that models including isospin - invariant strong force and the coulomb interaction alone have difficulties in reproducing the mirror displacement energies ( mdes ) , the differences of binding energies of mirror nuclei ( nolen - schiffer anomaly @xcite ) : @xmath172 these models has also problems in reproducing the triplet displacement energies ( tdes ) .
the ted indicator is defined as : @xmath173 and is a measure of the binding - energy curvature within the isospin triplet @xcite .
it is customary to classify components of the nuclear force in terms of the su(2 ) symmetry of isospace .
here , one defines charge independence as invariance under rotation in the isospin space . the charge symmetry , on the other hand , can be defined as an invariance under a rotation by 180@xmath174 about the @xmath175 isospin axis .
violation of this symmetry is referred to as charge symmetry breaking ( csb ) . in the language of nn force
, csb implies a difference between @xmath166 and @xmath164 interactions in the same channel , @xmath176 .
moreover , if in the isospin - triplet @xmath31 channel @xmath177 the force is charge - independence - breaking ( cib ) .
the data on nn scattering lengths indicate that the strong nn interaction must contain both the csb and cib strong components . on the fundamental level
, the csb mostly originates from the difference in masses of the proton and neutron which affects the kinetic energies and influences the boson exchange .
the cib originates mostly from the pion mass splitting , see refs .
@xcite . in nuclear medium ,
the csb and cib components of the effective strong force are accessible through the mdes and tdes , respectively .
a detailed classification of various components of the strong nn force was proposed by henley and miller @xcite .
they divided different components of the strong force into four classes . according to this classification , the isospin invariant ( i.e. , isoscalar )
nn interactions are called the class i forces .
they commute with the total nuclear isospin operator , @xmath178=0 $ ] . the class ii isotensor forces maintain charge symmetry , but break charge independence as @xmath179\neq0 $ ] . the class iii isovector forces break both the charge independence and charge symmetry , but are symmetric under interchange of nucleonic indices in the isospace .
such forces distinguish between the _ nn _ and _ pp _ systems and vanish in the _ np _ system
. forces of class ii and class iii do not mix isospin _ only _ at the two - body level ; hence they contribute to the isospin mixing in finite nuclei .
isovector forces of class iv break both symmetries and mix isospin already at the two - body level .
those forces do not influence the _ nn _ and _ pp _ systems , but induce a spin - dependent isospin mixing effects in the _ np _ system .
the isospin symmetry breaking components of nn interactions were studied in _ ab inito _ approaches @xcite .
hadronic isb terms of _ effective _ nn interactions were investigated in the context of nuclear shell model @xcite and mean - field approaches @xcite .
a note of caution is in order here .
first , any attempt to extract effective nn interactions from spectroscopic data should first account for the coupling to the many - body continuum @xcite in the presence of isospin - conserving nuclear forces .
if neglected , or not treated carefully , the continuum effects can alter the results of such analyses @xcite .
second , in - medium nuclear effective interactions ( g - matrix ) contain contributions from the coulomb force .
if the coulombic contributions are not treated precisely during the renormalization procedure , they can result in the presence of csb and cib components , which can then be incorrectly labelled as hadronic " . to investigate the effect of hadronic csb and cib terms and their possible influence on isospin mixing and isb corrections
, one can apply the local skyrme - dft approach . to this end
, one needs to generalize the formalism to the case of @xmath180-mixed quasiparticle states @xcite .
such an extension leads to isospin - invariant edfs that depend explicitly on local @xmath180 densities and currents .
recently , this formalism was applied to the hartree - fock case by admitting @xmath180 mixing in the particle - hole channel @xcite . within this framework ,
the explicit isb comes entirely from the electrostatic interaction .
in the next step , the @xmath180-mixed formalism can be extended to accommodate the csb and cib hadronic components .
the @xmath180-mixing is a necessary prerequisite that allows to study these terms in a fully self - consistent manner . since the discrepancies between the experimental and theoretical mdes and tdes are small @xcite they can be modelled in terms of class ii ( cib ) and class iii ( csb ) zero - range corrections to the conventional skyrme force : @xmath181 @xmath182 where @xmath183 and @xmath184 are the isb force parameters that are adjusted to reproduce empirical tdes and mdes .
the spin - exchange has been omitted as it leads to a trivial rescaling of the parameters .
the associated isb contributions to the edf are : @xmath185 note , that class ii forces depend on the @xmath180-mixed particle @xmath186 and spin @xmath187 densities , respectively .
hence , these forces can be included only within the @xmath180-mixed dft formalism .
the class iii forces , on the other hand , depend only on the standard @xmath166 and @xmath164 densities and can be , therefore , treated within the conventional skyrme - dft approach . in order to control the total isospin of the nucleus in the @xmath180-mixing calculations
, we use a three - dimensional isocranking approach @xcite .
this technique is analogous to the well known cranking method in real space , which is commonly used in high - spin physics .
it is realized by adding the isocranking term to the mean - field hamiltonian @xmath188 , @xmath189 where @xmath190 stands the s.p .
isospin operator and @xmath191 is the isocranking frequency . by adjusting the frequencies @xmath191 , one can control both the length and direction of the isospin vector . in practical applications
, @xmath191 is parameterized as follows : @xmath192 the offset @xmath193 is introduced to facilitate calculations with the coulomb interaction @xcite . by choosing the offset properly , one can compensate for the effect of the electrostatic interaction on the third component of the isocranking term . in this way
, @xmath194 helps to avoid s.p .
level crossings while tilting the isocranking axis and , consequently , keeps the expectation value of the total isospin fixed @xcite . in order to study the influence of the isb forces on nuclear binding energies
, we have performed the isocranking calculations for the @xmath11=34 isospin triplet . in these test calculations ,
the coulomb interaction has been switched off .
the results with the cib force ( [ eq : class2 ] ) are shown in fig .
[ fig.16 ] . binding energy for the @xmath11=34 isospin triplet calculated using the isocranking approach with the isoscalar sv skyrme force without coulomb , augmented by the cib ( class ii ) interaction ( [ eq : class2 ] ) with @xmath195=20mev . ]
as anticipated , this force impacts the binding energy of @xmath196 system without affecting the binding energies of @xmath197 triplet members .
hence , it changes the tde indicator without affecting mde . on the other hand , as shown in fig .
[ fig.17 ] , the class iii force ( [ eq : class3 ] ) changes mdes without affecting tdes .
this implies that the simplest strategy it to adjust the coupling constants @xmath195 and @xmath198 to tdes and mdes , respectively .
similar as in fig .
[ fig.16 ] but for the csb ( class iii ) interaction ( [ eq : class3 ] ) with @xmath199mev . ]
figures [ fig.18 ] and [ fig.19 ] show preliminary results of the calculated mdes and tdes in the isospin triplets ranging from @xmath11=22 to 55 .
the calculations have been done using the sv skyrme edf with coulomb , augmented by the cib and csb forces with @xmath200mev and @xmath201mev , respectively .
it is seen that a generalized @xmath180-mixed skyrme dft approach is able to reproduce experimental data on teds and meds .
top : experimental and calculated ( with and without hadronic isb terms ) mdes in the isospin triplets ranging from @xmath11=22 to 55 .
bottom : mde residuals pertaining to isb calculations . ] similar to fig .
[ fig.18 ] but for tdes . ]
the @xmath180-mixed isb dft formalism can address the evolution of isb effects with angular momentum in rotational bands of @xmath202 and @xmath31 nuclei .
such effects has been investigated within shell - model framework @xcite .
the quantities of interest are mirror energy differences ( meds ) and triplet energy differences ( teds ) , similar to mdes and tdes but defined through excitation energies @xmath203 of rotational states at a given angular momentum @xmath2 : @xmath204 @xmath205 in general , while shell - model studies indicate that the isb hadronic forces are as important as the coulomb interaction for the explanation of angular momentum dependence of meds and teds , the overall picture is not fully understood , especially in the upper @xmath206 shell .
one of the key features of the ncci model based on skyrme forces with low effective mass is its ability to reproduce rotational bands .
the preliminary calculations performed for @xmath207cr , a key nucleus in the @xmath62-shell concerning collective properties , indicate that our model is capable of reproducing its collective g.s .
band fairly well @xcite .
this opens an opportunity to address med and ted effects in a non - perturbative way , as a function of angular momentum , incorporating the isb effects self - consistently along the rotational path .
superfluidity and superconductivity belong to the most spectacular examples of emergent phenomena in many - body systems .
they appear at different physical scales and in different environments in atomic , condensed matter , nuclear , and elementary particle physics .
the bcs mechanism @xcite behind the pair condensate works irrespective of details of the underlying interaction that couples fermions into the cooper pairs at the fermi surface . in low - energy nuclear physics ,
nucleonic pairing affects many properties of finite nuclei and nucleonic matter @xcite .
nucleonic cooper pairs can exist in many flavors . in terms of isospin quantum number , one can talk about isovector triplet @xmath208 and isovector singlet @xmath209 pairs .
the conventional @xmath164 and @xmath166 pairing has isovector character , while the @xmath180 pairing can be either isovector or isoscalar .
the first attempts to incorporate the @xmath180-pairing into the independent quasi - particle approach date back to the mid - sixties @xcite .
these models were further developed into a consistent formalism allowing for simultaneous treatment of isovector and isoscalar pairs @xcite .
the formalism was further extended to include stretched @xmath6=0 pairs ( @xmath210 @xmath180-pairing ) in ref .
@xcite . these early approaches encountered difficulties in predicting coexisting @xmath211 and @xmath31 pairing - phases .
the solutions of the early models based on separable seniority - type interactions could be conveniently classified in terms of a single parameter , the value of isoscalar - to - isovector matrix element ratio @xmath212 . in particular , for @xmath10 nuclei ,
the solution is of a pure isovector type for @xmath213 ; for @xmath214 the isovector and isoscalar phases are degenerate ; and for @xmath215 there appears a pure isoscalar solution .
these models were later extended @xcite to include particle - number projected wave functions . for @xmath216 ,
the extended models have produced coexistence of @xmath211 and @xmath31 phases .
it was also predicted that for @xmath217 the isoscalar pairing component could be the source of the wigner energy @xcite . following these developments ,
various models have been proposed to look for signatures of the isoscalar pairing phase and explain the wigner energy @xcite . in spite of many efforts , however ,
a comprehensive theory of nucleonic pairing still eludes us .
this negatively impacts our understanding of nuclei in the vicinity of the @xmath10 line , where the @xmath180-correlations are expected to be strongest .
indeed , a consistent approach to the the @xmath180-pairing problem requires implementing the @xmath180-mixing on the mean - field level , whereby the s.p .
wave functions are combinations of proton and neutron components .
basic self - consistency principles require such a mixing to accompany any hypothetical @xmath180-mixing in the pairing channel .
moreover , the stability and existence of the @xmath180-pairing condensate may critically depend on the restoring force related to the @xmath180-mixing on the mean field level , and thus both must be simultaneously included in the theoretical description . as discussed above ,
the dft frameworks incorporating the @xmath180-mixing have been developed recently @xcite .
this constitutes the first step towards building a consistent symmetry - unrestricted superfluid dft approach with the complete @xmath180-mixing .
the experience gathered so far indicates that in order to capture structure of @xmath0 nuclei , the @xmath180-mixed formalism should be further extended to include the restoration of number symmetry and proper treatment of isospin @xcite .
experimental ( circles ) and calculated ( ncci ; dots ) excitation energies of the @xmath211 states in @xmath53sc relative to the @xmath112 ground state . ] a need for the enhancement of the deuteron - like @xmath218 pairing channel is apparent in the present skyrme - ncci approach , which accounts relatively well for low - spin states but has systematic problems with the @xmath219 matrix elements in @xmath10 nuclei .
this is illustrated in fig .
[ fig.20 ] showing the calculated and experimental @xmath211 multiplet in @xmath53sc .
it is seen that the antialigned state @xmath218 is strongly underbound and the stretched @xmath220 level is overbound .
a similar situation has been encountered in @xmath221li where the ncci model strongly underbinds @xmath219 matrix elements @xcite .
in this paper , we overview selected topics related to isobaric spin in @xmath7 nuclei . we demonstrate that no - core " and self - consistency aspects of nuclear dft are essential for our quantitative understanding of isospin breaking .
the approximate nature of isospin poses many theoretical challenges as the underlying symmetry is broken both explicitly and spontaneously , and the isospin mixing is highly non - perturbative due to a mismatch between the ranges of electrostatic and strong forces .
to handle this situation theoretically , one needs to go beyond the usual single - reference dft formalism .
the proposed multi - reference dft framework , based on isospin- and angular - momentum projected wave functions , provides a very satisfactory description of isospin breaking effects in beta decay .
the future developments will undoubtedly utilize the newly developed isospin - invariant density functional framework @xcite .
the @xmath180-mixed dft formalism needs to be extended to the particle - particle channel by including pairing interaction of both isoscalar and isovector types .
this will enable us to study the importance of the isoscalar pairing densities and fields on the structure of @xmath0 nuclei and the impact of @xmath180-mixing on single- and double - beta decays .
it seems , however , that in order to capture key structural aspects of @xmath10 nuclei , the @xmath180-mixed formalism should be further extended to include the restoration of particle number , isospin , and angular momentum .
this will require an appropriate ncci extension . in quantitative calculations ,
quality input is crucial .
a construction and optimization of a realistic hamiltonian - based edf is among the most urgent needs .
this development will not only improve the predictive power of the model but will also help addressing the burning question pertaining to the role of isb interaction components . useful discussions with jacek dobaczewski are gratefully acknowledged .
pawe baczyk is thanked for providing material presented in sec .
[ sect.05 ] prior to publication @xcite .
this article is based upon work supported by the polish national science centre ( ncn ) under contract no .
2012/07/b / st2/03907 and by the u.s .
department of energy , office of science , office of nuclear physics under award numbers no.doe-de-sc0013365 ( michigan state university ) and no .
de - sc0008511 ( nuclei scidac collaboration ) .
the csc@xmath222it center for science ltd , finland , is acknowledged for the allocation of computational resources . | this paper overviews various phenomena related to the concept of isospin symmetry .
the focus is on @xmath0 nuclei , which are excellent laboratories of isospin physics .
the theoretical framework applied is nuclear density functional theory and its isospin- and angular - momentum projected extensions , as well as symmetry - projected multi - reference models .
the topics covered include : isospin impurities , superallowed beta decays , beta - transitions in mirror nuclei , isospin breaking hadronic interactions , mirror and triplet binding energy differences , and isoscalar pairing . |
absorption processes and their signatures ( absorption lines ) imprinted on the spectra of bright background sources ( quasars , gamma - ray bursts , etc . ) are one of the main sources of information about the physical and chemical properties of intervening systems .
it is well known that information about their temperature , density , chemical abundances , and kinematics can be extracted from the analysis of these absorption lines .
for instance , a detailed insight into the physical state of the intergalactic medium ( igm ) is provided by the analysis of the absorption lines found in the spectra of distant quasars ( qsos ) ( see _ e.g. _ * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
these lines are due mainly to absorption by neutral hydrogen ( hi ) present in a class of low column density absorbers generally known as ly@xmath3 forest , and due to other elements in low ionisation stages ( cii , civ , si ii , mgii , feii , oii , etc . ) , which arise in higher column densities absorbing systems associated with galaxies , such as the lyman limit systems ( llss ) and damped ly@xmath3 absorbers ( dlas ) . a wealth of information about the distribution , density , temperature , metal content , etc .
of these systems is now available as a result of exhaustive and extensive studies of qso absorption lines ( see _ e.g. _ * ? ? ?
* ; * ? ? ?
* for a review on ly@xmath3 absorbers , and on metal systems and dlas , respectively ) . in this type of analysis , and within the realm of a given cosmological model , the number and observed central wavelength of the absorption lines provide information on the spatial distribution of the absorbing systems .
furthermore , knowledge about the actual physical state of these systems can be obtained basically from the line profiles .
both line counting and line profile measurement are tricky tasks though , since the accuracy with which they can be performed highly depends on the resolution of the observed spectra .
for instance , depending on the spatial distribution of the absorbing systems , lines can appear very close to each other or even superpose ( line blending ) , and a low spectral resolution may lead to the misidentification of the resulting composite profile as being a single , complex one . because of this same reason , the determination of the exact shape of each individual absorption profile is far from being trivial , and misidentified profiles may lead to wrong conclusions about the properties of the absorbing systems .
if one assumes that the physical state of the absorbing medium is uniquely defined by its temperature and column density , single absorption line profiles are ideally described by voigt profiles .
mathematically , a voigt profile is given in terms of the convolution of a gaussian and a lorentzian distribution function , known as voigt - hjerting function @xcite , and a constant factor that contains information about the relevant physical properties of the absorbing medium ( cf .
[ sec : abc ] ) .
any departure from a pure voigt profile in the observed lines is expected to yield information about the kinematic properties ( non - thermal broadening , rotational or turbulent macroscopic motions ) , as well as spatial information ( clustering ) of the absorbing systems . the voigt - hjerting function has long been known and , consequently
, various numerical methods to estimate and tabulate this function have been developed and presented @xcite .
also , a great effort has been done in order to derive a semi - analytic approximation to this function that reproduces its behavior with high accuracy ( see _ e.g. _ * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , being the latter by far the one with the highest accuracy . with the help of these methods ,
computational subroutines have been developed that make it nowadays possible to numerically integrate this function for a wide parameter space ( see _ e.g. _ * ? ? ?
the aim of this work is to make a further contribution to the practical handling of the voigt - hjerting function in order to compute voigt profiles . starting with an exact expression for this function in terms of harris infinite series ,
we argue why this series may be truncated to first order in @xmath4 in the context of intergalactic hi absorption lines .
we then show that the second term of this series can be approximated with a non - algebraic polynomial function , which is mathematically simple to handle in the sense that is does not contain singularities .
such an analytic , well - behaved expression in terms of simple functions as presented here is very attractive , since it allows one to replace the many steps and operations needed for numerical integration , or to read from look - up tables of values , by a single line with simple operations .
it is also extremely flexible to implement in various types of codes and machines , and is particularly useful for computational routines in higher - level programming languages ( _ e.g. _ idl , _ mathematica _ , _ maple _ , etc . ) , in which numerical integration or look - up table reading is cumbersome , especially if absorption line profiles have to be calculated many times with moderate precision and relative high speed .
for instance , such an analytic expression should be very useful to synthesise ly@xmath3 absorption spectra as in _ e.g. _
@xcite , or in line - fitting algorithms like autovp @xcite or fitlyman @xcite , used to obtain line parameters such as redshift , column density , and doppler width from absorption voigt profiles imprinted on observed spectra . in the next section we briefly outline the origin of the voigt - hjerting function in the physics of absorption processes , and define the context in which the desired approximation of this function is of interest to us . in sections [ sec : dawson ] and [ sec : analyt ] we derive this approximation , and in section [ sec : ana ] we compare the accuracy and speed of a numerical method for computing voigt profiles based on our approximation to other existing methods . in section
[ sec : app ] we present an application of our method to model voigt profiles , and in section [ sec : summ ] we summarise our main results .
the probability of a photon with an energy @xmath5 to be absorbed within a gas with column density n and kinetic temperature @xmath6 , also known as absorption coefficient , is given by @xmath7 where @xmath8 is a constant for the @xmath9th electronic transition caused by the photon absorption . here
@xmath10 is the electron mass , @xmath11 is the oscillator strength , and @xmath12 the damping constant or reciprocal of the mean lifetime of the transition .
the function @xmath13 is the so - called * voigt - hjerting function * and is given by @xmath14 let @xmath15 be the resonant wavelength of the corresponding transition , and @xmath16 the thermal or doppler broadening , which defines a doppler unit . here
, the doppler parameter @xmath17 is related to the kinetic temperature of the gas via @xmath18 , where @xmath19 is the boltzmann constant and @xmath20 is the proton mass .
it follows from these definitions that the damping parameter @xmath21 quantifies the relative strength of damping broadening to thermal broadening , and that the variables @xmath22 and @xmath23 are just the wavelength difference relative to the resonant wavelength in doppler units , and the particle velocity in units of the doppler parameter , respectively .
the particular form of the absorption coefficient ( [ eq : tau ] ) induces a characteristic absorption feature known as voigt profile .
hence , the voigt profile , and consequently the voigt - hjerting function , naturally arise in the process of absorption - line formation , when one assumes that the physical state of the absorbing medium is uniquely defined by its density and kinetic temperature . generally speaking , it is the physical conditions what determines the shape of the absorbing features , _
i.e. _ the line profiles .
conversely , it is true that line profiles give information about the physical state of the absorbing medium . in particular
, line profiles other than voigt profiles give information about the departure of the physical conditions assumed here . the class of hi absorbers present in the igm ( ly@xmath3 forest clouds ) and associated with galaxies and larger structures ( llss , dlas ) can be characterised by their column density @xmath24 and kinetic temperature , and consequently their absorption features observed on _
e.g. _ qso spectra are well described by voigt profiles .
their observed column densities span a range of ten orders of magnitude , approximately from @xmath25 , and have temperatures that correspond to doppler parameters approximately in the range @xmath26 , with a median value around @xmath27 that decreases with redshift @xcite .
for such a range in @xmath17 , the damping parameter @xmath4 for the lyman transitions of intergalactic hi spans a range of @xmath28 . in this case
, high values of @xmath4 correspond to ly@xmath3 , while lower values are typical for higher order lyman transitions . transitions of other elements , such as c , si , mg , fe , o , etc . , and their various ionisation stages , are also typically found on qso spectra , and their damping parameters cover a range which strongly overlaps with that of hi .
this can be seen in figure [ fig : adis ] , where we show the distribution of @xmath4 for different elements ( including hi ) in different ionisation stages , for a doppler parameter @xmath29 for hi , and assuming that the doppler parameter for other elements is related to @xmath30 via @xmath31 , where @xmath32 is the mass of element @xmath33 . for clarity
, we have grouped all different ionisation stages of a given element under a single label .
the values of the atomic constants ( central wavelength of the transition and @xmath34-value ) have been taken from @xcite , is available in plain - text format at www.astro.physik.uni-goettingen.de/~tepper/hjerting/damping.dat
. please consult this list in order to know the exact value of @xmath4 for a given element in a given ionisation stage . ] .
the shaded area contains the values of @xmath4 above the range characteristic to intergalactic hi for which our approximation to the voigt - hjerting function derived in the next sections
cannot be applied or should be applied with caution .
note , however , that the region spanned by @xmath4 for intergalactic hi , _ i.e. _ the region underneath the shaded area , contains most of the @xmath4-ranges spanned by all other elements , especially those corresponding to c , o , mg , and fe .
for other elements , such as cr or zn , the damping parameter has values right at the upper limit of this range .
for this reason and for the sake of simplicity , in the following we will constrain our discussion to the lyman absorption lines of intergalactic hi , but the reader shall bear in mind that the the discussion and method to synthesise voigt profiles presented in this work can be directly applied to the transitions of other elements associated with hi absorbing systems , according to figure [ fig : adis ] .
following @xcite , it is true that for @xmath35 , _
i.e. _ when doppler broadening dominates over damping broadening , the voigt - hjerting function ( [ eq : vhf ] ) can be expressed as @xmath36 where the functions @xmath37 are defined by @xmath38 these functions are bounded with respect to @xmath39 and @xmath40 , and they have values of the order of unity .
indeed , taking the absolute value of the integral , neglecting the cosine , which takes values of the order of unity , and computing the resulting integral one can show that @xmath41 for all @xmath42 and @xmath43 . from this
it follows that if @xmath44 , the voigt - hjerting function can very well be approximated to zeroth order in @xmath4 by the first term of the series ( [ eq : vhf2 ] ) , _ i.e. _ @xmath45 as first noted by @xcite .
note that this result is _ exact _ in the limit @xmath46 .
taking the definition ( [ eq : vhf3 ] ) , it follows that @xmath47 and thus @xmath48 for @xmath44 .
we call this the voigt - hjerting function to _ zeroth order_. but what actually means that the condition @xmath44 be satisfied , so that this zeroth order approximation can be safely used to model absorption line profiles ? in order to address this , we compute for the extreme values of @xmath4 for intergalactic hi , @xmath49 and @xmath50 , the departure of the voigt - hjerting function from a pure gaussian function . this
is shown in figure [ fig : hvsh0 ] as the logarithmic difference between ) . ]
@xmath13 and @xmath51 relative to @xmath13 as a function of @xmath40 , _
i.e. _ the quantity .
note how the zeroth order approximation completelly differs from the actual voigt - hjerting function at @xmath52 for @xmath53 , and at @xmath54 for @xmath55 .
thus , even a value of @xmath4 as small as @xmath56 is not a sufficient condition for the zeroth order term to be a good approximation of @xmath13 for an arbitrary range in @xmath40 . in the context of the hi absorption coefficient
, the condition on @xmath4 for which the approximation to zeroth order of the voigt - hjerting function is valid , translates into a restriction on @xmath24 , and more specifically on the quantity @xmath57 .
this can be seen by taking a glance at equation ( [ eq : tau ] ) : despite the fact that the condition @xmath44 holds , the product @xmath58 can be very large for high enough is satisfied . ]
column densities , and since the functions @xmath59 are of the order of unity , such terms can have a significant contribution to the absorption coefficient . along this line of reasoning , and taking into consideration that the constant @xmath60 in equation ( [ eq : tau ] ) is of the order of @xmath61 for all lyman transitions , it is clear that the departure of the voigt - hjerting function from its zeroth order approximation becomes significant in the wavelength ranges @xmath62 and @xmath54 ( in doppler units ) at column densities @xmath63 for a damping parameter @xmath64 , and at column densities @xmath65 for @xmath66 . since intervening hi absorbers typically have column densities in the range @xmath67 , a gaussian approximation to @xmath13 for modeling intergalactic hi absorption line profiles is only suitable for the low end of the column density range .
in addition to the factor @xmath58 being large and even more decisive for the uselessness of the zeroth order approximation for an arbitrary range in @xmath40 , is the fact that the zeroth order term , @xmath68 , rapidly decreases for large values of @xmath40 and is therefore overwhelmed by higher order terms in harris expansion , which hence dominate the behavior of the absorption coefficient , as already seen in figure [ fig : hvsh0 ] . to shed some light on this ,
consider the following numerical example : out to @xmath69 ( in doppler units ) , and for @xmath64 , the voigt - hjerting function is of the order of @xmath70 .
the zeroth order term in the series ( [ eq : vhf2 ] ) satisfies @xmath71 , whereas the first order term @xmath72 .
thus , at large enough @xmath40 , the behaviour of @xmath73 is evidently governed by the terms of order @xmath74 in the series ( [ eq : vhf2 ] ) . due to the arguments stated above , and even though the damping parameter satisfies @xmath44 , it is clear that the absorption coefficient of intergalactic hi can not simply be approximated by a constant times @xmath68 for an arbitrary wavelength range and column densities grater than @xmath75 .
one actually has to take into account terms of higher order in the series ( [ eq : vhf2 ] ) , at least to first order in @xmath4 , _
i.e. _ @xmath76 in fact , one should take into account all terms up to @xmath77th order for values of @xmath78 that are nearly equal or greater than the absolute difference between the sum @xmath79 and the exact voigt - hjerting function .
however , as we shall show next , the approximation to first order in @xmath4 in harris expansion is enough to model absorption line profiles with moderate to high accuracy for the range of parameters @xmath80 characteristic to intervening hi absorbers . in order to prove the above statement , we look at the relative contribution of the zeroth and first order terms to the voigt - hjerting function for @xmath81 and an arbitrary range in @xmath40 .
this is achieved , for example , by computing the logarithm of the quantity @xmath82 as a function of @xmath40 for the two extreme values of @xmath4 , as shown in figure [ fig : hvsh0h1 ] .
here we have advanced the function @xmath83 which is being handled in the next section .
note that the greater the contribution from the zeroth and first order term to @xmath13 , the smaller the quantity @xmath84 . in this case , as can clearly be seen , @xmath84 takes on values of the order of @xmath85 or less over the whole wavelength range shown here and for the whole range in @xmath4 for intergalactic hi .
if one takes into account that @xmath86 for all @xmath40 , it is obvious that the relative difference is equal or greater than the _ absolute _ difference , _
i.e. _ @xmath87 .
thus , in the case of @xmath50 and @xmath88 , the departure of @xmath13 from its first order approximation becomes significant at column densities @xmath89 , and at even larger @xmath24 for @xmath90 and/or smaller @xmath84 .
hence , the first two terms of harris expansion dominate the behaviour of @xmath13 over the whole wavelength range shown here , with an accuracy of @xmath85 or greater , for the range in @xmath4 characteristic to intergalactic hi . on this basis
, we consider thar an approximation to first order in @xmath4 of the voigt - hjerting function in terms of the functions @xmath51 and @xmath83 is suitable to model voigt profiles .
since the function @xmath51 is known and simple _ per se _ , we now turn to the task of finding an approximation to the function @xmath83 in terms of a simple , analytic expression .
according to definition ( [ eq : vhf3 ] ) we have @xmath91 integrating this equation partially , and computing the resulting sinus transform of a gaussian it follows @xcite @xmath92 \ , , \ ] ] where we adopt the notation first introduced by @xcite @xmath93 @xmath94 is known as the _ dawson function _ @xcite .
it is evident that finding an approximation for @xmath83 translates into the same problem for @xmath94 .
we thus want to show that a simple analytic expression can be found which approximates the dawson function , and hence the function @xmath83 , accurately enough , in order for an approximation of the voigt - hjerting function in terms of these functions to be useful for synthesising voigt profiles .
we briefly want to state some important properties of the dawson - function .
first , this function is antisymmetric , _ i.e. _ @xmath95 for all @xmath96 .
because of this , from now on we restrict our analysis to @xmath97 .
besides , it has no roots in the positive semi - axis , and @xmath98 , as can easily be seen from its definition ( fundamental theorem of calculus ) .
furthermore , @xmath94 is bounded , since @xmath83 is bounded itself ( cf .
[ sec : low ] ) .
indeed , differentiation with respect to @xmath40 gives @xmath99 hence , the upper bound is given by @xmath100 , where @xmath101 is defined by equating @xmath94 s derivative to zero and solving for @xmath40 .
actually , @xmath94 has its maximum at @xmath102 with @xmath103 ( see _ e.g. _ * ? ? ?
from this last equation it follows also that equation ( [ eq : h_1 ] ) can be rewritten as @xmath104 we want to know how the function @xmath94 behaves asymptotically , _
i.e. _ for @xmath105 as well as for @xmath106 . using the power series of the exponential we get @xmath107 in this case , the sum and the integral operator commute ,
since the power series of the exponential converges uniformly in any interval @xmath108 $ ] , particularly for @xmath109 $ ] ( see _ e.g. _ * ? ? ?
expressing the term @xmath68 by its corresponding power series and rearranging terms , this last equation reads in explicit form @xmath110 thus , for @xmath111 the dawson function behaves asymptotically up to third order as @xmath112 in order to investigate how @xmath94 behaves for @xmath106 , we first rewrite the function ( [ eq : daw ] ) with the replacement @xmath113 and the aid of the power series of the exponential as @xmath114 with the definition @xmath115 it is not hard to see that every term of the series ( [ eq : daw4 ] ) is separately bounded with respect to @xmath40 as well as @xmath39 . making the replacement @xmath116 in eq .
( [ eq : i_n ] ) , and integrating for @xmath117 we get for @xmath118 @xmath119 where @xmath120 is a rather cumbersome polynomial function .
now , for @xmath121 , we may drop all terms which contain an exponential factor and in this way we get the asymptotic form @xmath122 it can be seen from this expression that @xmath94 vanishes as @xmath123 for @xmath124 , and that the first derivative ( [ eq : daw2 ] ) , and thus the function @xmath83 , also vanish in this limit as @xmath125 . from equation ( [ eq : daw2 ] )
it is also true that @xmath126 converges to unity for @xmath127 and that @xmath83 converges to @xmath128 in this limit .
since we want our approximation to @xmath94 , and consequently to @xmath83 and @xmath13 , to be valid in the whole range @xmath129 , we require it to fulfill both these conditions as well .
equation ( [ eq : daw4 ] ) , together with eq .
( [ eq : i_n ] ) , represent indeed an exact expression for the dawson function
. however , these expression are not suitable for practical computation .
we therefore explore the possibility of finding an analytic expression which is easy to handle and which can be used to compute the value of @xmath131 for @xmath129 .
in particular , we shall see if it is possible to truncate the series ( [ eq : daw4 ] ) in order to find an approximation to @xmath94 , which has all its properties ( antisymmetry , boundedness , etc . ) , which converges for @xmath132 as well as for @xmath133 , and which is well defined in the whole range @xmath134 .
for instance , equations ( [ eq : fap1 ] ) and ( [ eq : fap2 ] ) do not fulfill these requirements .
nevertheless , they show us how our desired function has to behave asymptotically .
let us define @xmath135 where the @xmath136 s are given by eq .
( [ eq : i_n ] ) . using this definition
we get @xmath137 - \mathrm{e}^{-2x^2 } \cdot \left[\frac{x}{2 } + \frac{1}{2x } \right ] \ , .\ ] ] it is easy to show that this function behaves qualitatively in the same way as @xmath94 does , _
i.e. _ it is antisymmetric , bounded , and has no roots in the positive semi - axis .
furthermore , both these functions have the same asymptotical behavior . indeed , up to third order we have for @xmath111 @xmath138 as can be shown by expanding the exponentials in eq .
( [ eq : d1 ] ) in terms of their power series .
a glance at eq . ( [ eq : fap1 ] ) makes the similarity between @xmath94 and @xmath130 evident in this limit .
for @xmath106 we get from eq .
( [ eq : d1 ] ) , neglecting the exponentials , @xmath139 that is the same as for @xmath94 ( see eq .
[ eq : fap2 ] ) .
furthermore , the first derivative of @xmath130 with respect to @xmath40 is unity at @xmath140 and vanishes as @xmath125 for @xmath132 .
however , the maximum of @xmath130 is at @xmath141 with @xmath142 , _
i.e. _ at slightly different values from those of @xmath94
. the magnitude and range of the error in approximating the dawson function by the function @xmath130 can be estimated , at least qualitatively , in the following way : from equations ( [ eq : daw4 ] ) and ( [ eq : fap3 ] ) it follows that @xmath143 the sum in this last equation , _ i.e. _ the absolute error in our approximation @xmath144 , is a positive semi - definite quantity , since each term of the sum has this property .
hence , it is true that @xmath145 for @xmath129 .
furthermore , since both @xmath94 and @xmath130 converge to zero for @xmath127 as well as for @xmath132 , and both these functions are bounded , it follows from eq .
( [ eq : fap4 ] ) , that the error also vanishes for @xmath127 as well as for @xmath132 , that it is also bounded , and that its maximum value is less equal than @xmath146 . from this , and since their respective maxima are slightly shifted with respect to each other , one is led to the conclusion , that the error is constrained to a narrow range in @xmath40 and that the maximum error in our approximation to @xmath94 occurs near the maxima of these functions , _
i.e. _ in the vicinity of @xmath147 . we wo nt further try to quantify the actual error in our approximation to @xmath94 .
it shall be enough to know , for our purposes of finding an approximation to the voigt - hjerting function , that the error in the approximation @xmath144 is bounded and constrained to a narrow wavelength interval around @xmath147 .
besides , we will indirectly estimate the error in this approximation when quantifying the error in our approximation to @xmath13 in section [ sec : ana ] .
once we have found an approximation to the dawson function , we can use it to give the desired expression for the voigt - hjerting function using harris expansion to first order in @xmath4 . replacing in equation ( [ eq : h_12 ] ) the function @xmath94 by our approximation @xmath130 ( eq . [ eq : d1 ] ) ,
taking the corresponding derivative , and rearranging terms we get @xmath148 where we have defined @xmath149 \ , .\ ] ] we want to highlight the fact that equation ( [ eq : h_13 ] ) is well defined , _
i.e. _ it has no singularities in the whole interval @xmath134 .
furthermore , it converges to the correct value in the limits @xmath127 and @xmath132 .
indeed , it is easy to show that @xmath150 and @xmath151 whether one uses for @xmath83 the exact expression ( [ eq : h_1 ] ) or the approximation ( [ eq : h_13 ] ) .
in contrast , in approximations to the voigt - hjerting function to model voigt profiles often used in the literature ( see _ e.g. _ * ? ? ? * ; * ? ? ? * ) and given in the form @xmath152 where the @xmath153 s are constants , the second term which represents the lorentzian damping clearly diverges for @xmath127 , and one has to artificially define the wavelength range in which this second term is used .
it is customary to neglect this term for low column densities and in the vicinity of @xmath140 , but how to exactly choose the radius of the vicinity is not clear and completely arbitrary .
however , with an expression like equation ( [ eq : h_13 ] ) at hand , no such assumptions have to be made .
taking into account that @xmath44 in order to neglect terms of order @xmath154 in the series ( [ eq : vhf2 ] ) , we get , using the expressions for @xmath51 ( eq . [ eq : h_0 ] ) and @xmath83 ( eq . [ eq : h_13 ] ) , that the voigt - hjerting function to first order in @xmath4 is given by @xmath155 \ , .\ ] ] this expression is symmetric in @xmath40 , as it should be , and thus it is valid for @xmath43 and @xmath44 . according to this equation
, the voigt - hjerting function can be regarded as a corrected gaussian function , where the correction term depends on the parameter @xmath4 . in the context of the absorption coefficient of hi ,
this correction term also depends on the column density @xmath24 , of course , via the quantity @xmath156 .
in order to quantify the quality of our approximation to @xmath13 , we perform a test on speed as well as on precision , comparing a numerical method to compute @xmath13 , based on our approximation , to other standard , available methods to numerically compute this function . for this purpose ,
we use the approach and corresponding computational routine developed by @xcite , which consists of the numerical implementation in fortran of four different methods to compute @xmath13 : harris @xmath157 and @xmath158 , huml@xmath159ek s , and monaghan s . in murphy
s notation , harris @xmath157 and @xmath158 correspond to the voigt - hjerting function approximated by the first three and five terms of the series expansion ( [ eq : vhf2 ] ) , respectively .
s optimized algorithm and monaghan s differential approach to approximate @xmath13 are explained in detail in @xcite and @xcite , respectively . our method to compute @xmath13
consists simply in the numerical implementation in fortran ) and ( [ eq : vhf4 ] ) that leads to the smallest number of operations reads , in code syntax , @xmath160 \ , , \ ] ] where the terms @xmath161 , @xmath162 and @xmath163 have to be computed just once . ] of equations ( [ eq : h_14 ] ) and ( [ eq : vhf4 ] ) . following murphy s approach ,
the relative speed of all five methods are determined by calculating the time that a routine based on each method requires to compute the voigt - hjerting function for and damping parameters @xmath4 in the range @xmath164 for a total of @xmath165 runs . in this way
, we get that the relative speed of each method in the order corresponds to , independent of @xmath4 .
according to this result , our method is second fastest .
the precision of our method relative to the other methods mentioned above is determined in the following way : first , the values of the voigt - hjerting function computed using monaghan s algorithm for @xmath166 $ ] and @xmath167 $ ] are taken as fiducial .
then , the value of @xmath13 for the same range in @xmath40 and @xmath4 is computed using each of these methods , and the logarithmic difference between each method and its fiducial value , _
i.e. _ the quantity @xmath168 with @xmath169 , is calculated as a function of @xmath40 for each different @xmath4 .
the result is shown in figure [ fig : comp ] .
it can be seen from this figure that harris @xmath158 is the second most precise method to compute @xmath13 , if one takes monaghan s algorithm as fiducial , but is six times slower than harris @xmath157 , and one - and - a - half times slower than our method , as stated in the previous section .
our method has a precision of @xmath0 or better for @xmath170 . for all values of @xmath4 , the difference peaks around @xmath171 to a value of the order of 0.01 , and
the precision increases again for values of @xmath172 .
the precision is better for smaller @xmath4 , as expected , since the zeroth order term gains in importance in our approximation for decreasing @xmath4 . for @xmath173 and @xmath174 ,
our method is more precise than harris @xmath157 or huml@xmath159ek s , and , as seen above , 1.5 times faster than the latter .
we now turn to analyse how accurate is our method in order to model hi absorption profiles . taking the whole range in column density @xmath175 $ ] dex and doppler parameters @xmath176 \ , \mathrm{km \ , s^{-1}}$ ] characteristic to intergalactic hi , we synthesise for each pair @xmath177 ( with a resolution of @xmath178 dex , and @xmath179 ) a single ly@xmath3 absorption profile in the range @xmath180 $ ] with a resolution of @xmath181 .
the absorption profile is synthesised according to equations ( [ eq : tau ] ) and ( [ eq : vhf ] ) , using both our method and monaghan s to compute @xmath13 .
we then compute for each pair @xmath177 the absolute value of the difference between the profiles generated using these two methods relative to monaghan s , _
i.e. _ @xmath182 , with @xmath183 , as a function of wavelength for the whole wavelength range , and pick the _ maximum _ value of this difference in the _ entire _ range . we choose to do so in order to pick up the worst cases possible , _
i.e. _ those with the lowest accuracy , and put in this way a stringent lower limit to the accuracy of our method .
the result is shown as a contrast diagram on the @xmath184-plane in figure [ fig : map ] .
note that , in this case , it is not the logarithmic , but the _ linear _ difference which is shown .
the highest precision is of the order of @xmath185 or even better .
however , for the sake of simplicity , any value below @xmath0 has been coded as zero . the largest discrepancy between both methods , _
i.e. _ the lower limit in the precision of our method if one takes monaghan s as fiducial , amounts to 0.01 , in agreement with the result shown in figure [ fig : comp ] .
as can clearly be seen , the ( lower limit in the ) precision of our method depends both on @xmath17 and @xmath24 . while the dependence on @xmath17 extends to the whole range @xmath186 , the dependence on @xmath24 is limited to the range @xmath187 dex . for a fixed @xmath17
, there is a regime of values around 0.01 with a width of nearly 2 dex , which gives as a result a narrow region of low accuracy all across the plane . along this stripe
, the difference reaches its highest value of the order of 0.01 or smaller for the combinations ( low @xmath17 , low @xmath24 ) or ( high @xmath17 , high @xmath24 ) , in the @xmath188 and @xmath189ranges stated above . outside this stripe ,
the accuracy increases dramatically to values of the order of @xmath0 or even better .
the origin of any inaccuracy in our method is obviously the fact that terms of order @xmath190 have been neglected in the series ( [ eq : vhf2 ] ) , and furthermore , that the second term is this series has been approximated as well .
in particular , the origin of the low - accuracy stripe on the @xmath191-plane can qualitatively be understood in terms of the functional dependence of @xmath192 on @xmath17 and @xmath24 .
considering that both our method and monaghan s take the zeroth order term exactly into account , it is legitime to state that the quantity @xmath193 has a least a dependence of first order on @xmath4 , _
i.e. _ @xmath194 where @xmath195 is a function that may be of zeroth order in @xmath4 . hence
, using equation ( [ eq : tau ] ) and this last expression , it follows that @xmath196 as can be seen , @xmath73 strongly depends on the ratio @xmath197 .
therefore , an increase in @xmath17 of one order of magnitude ( from 10 to 100 @xmath198 ) is nearly compensated ( in the sense that the value of @xmath73 remains nearly constant ) by an increase in @xmath24 of two orders of magnitude , which accounts for the stripe of 2 dex in column density seen on the @xmath184-plane .
why this happens precisely between @xmath199 dex , as well as the shape of this stripe , is non - trivially related to the exact value of the constant @xmath200 , the fact that @xmath195 may depend also on @xmath17 through the damping parameter @xmath4 , and the fact that the @xmath201 depends effectively not on @xmath202 , but on @xmath203 .
it is worth mentioning that , for higher - order lyman transitions , the precision of our method to compute voigt profile has the same behaviour on the @xmath184-plane , and is the same as or even better than the precision of the ly@xmath3 transition shown here .
the reasons for this are , first of all , that the functional form of the absorption coefficient is obviously the same , irrespective of the transition .
in addition , the lowest precision possible of 0.01 is the same for the whole range in @xmath4 spanned by the lyman transitions , according to figure [ fig : comp ] .
furthermore , higher transitions have lower damping parameters and our approximation is better the lower @xmath4 , as already mentioned . finally , since the constant @xmath200 is smaller the higher the order of the transition , the critical range of lowest precision is shifted to higher column densities and higher doppler parameters . since the ranges in @xmath24 and @xmath17 are fixed for intergalactic hi , this has the net effect of increasing the high - precision region on the @xmath204-plane_i.e .
_ the region to the left of the stripe for higher order transitions . in other words , the precision of our method improves from ly@xmath3 to higher lyman transitions . even though a discrepancy of the order of 0.01 when using our method to model voigt profiles may seem large , we want to emphasise again that this is merely an _ lower _
limit for the precision in the _ entire _ wavelength range @xmath180 $ ] , in the case of ly@xmath3 .
it turns out that the range in wavelength for which the accuracy is lowest is negligible for practical purposes . in order to show this
, we first choose three points on the @xmath184-plane along the stripe of lowest precision , _
i.e. _ for which the maximum difference is largest . using these parameters , we synthesise ly@xmath3 absorption profiles using our method and monaghan s , and compute again for each of these worst scenarios the quantity @xmath192 for the whole wavelength range .
the result is shown in figure [ fig : dla ] .
each row corresponds , from top to bottom , to the parameter pairs @xmath205 , @xmath206 , and @xmath207 chosen along the low - precision stripe .
the upper panel of each row shows the ly@xmath3 absorption profile synthesised using our method , and the corresponding lower panel gives the logarithmic difference @xmath208 between our method and monaghan s as a function of wavelength .
as can be seen , the smallest discrepancies are given at the line cores , as expected , since in this regime the zeroth order term dominates and both our approach and monaghan s exactly take this term into account .
the largest discrepancies , of the order of 0.01 , are present in an extremely narrow range of @xmath209 for @xmath210 , of @xmath211 for @xmath212 , and of @xmath213 for @xmath214 .
this discrepancies are found at the boundaries between gaussian core and lorentzian wings , due to the fact that our method neglects terms of order @xmath190 , which dominate the behaviour of @xmath13 in that regime .
note , however , that the difference rapidly drops with increasing distance ( in ) from the line center to values of the order of _
e.g. _ @xmath215 at a distance @xmath216 for the first two rows , and @xmath217 for last row .
hence , the effective accuracy of our method is far better than 0.01 in the wavelength range shown here . for the same reason mentioned in the last paragraph , the precision for higher order lyman transitions is of the same order as or even better than for the ly@xmath3 transition shown here
as a further test of the quality of our approximation to model voigt profiles , and to show its accuracy in a less academic situation as in the last section , we consider fitting a synthetic spectrum to a real quasar absorption spectrum with a population of intergalactic hi absorbers spanning a representative range in @xmath17 and @xmath24 along a random line - of - sight .
for this purpose we use the observed spectrum of the quasistellar source qso j2233 - 606 , a relatively bright ( @xmath218 ) quasar at an intermediate redshift @xmath219 . the spectrum of the source qso j2233 - 606 , centered at the hdf - s ,
was obtained during the commissioning of the uves instrument at the vlt kueyen telescope and reduced at the space telescope european coordinating facility .
the high - resolution spectroscopy ( r @xmath220 45000 ) was carried out with the vlt uv - visual echelle spectrograph ( uves ) .
the data were reduced in the echelle / uves context available in midas .
the final combined spectrum has constant pixel size of 0.05 and covers the wavelength range 3050 - 10000 .
the s / n ratio of the final spectrum is about 50 per resolution element at 4000 , 90 at 5000 , 80 at 6000 , 40 at 8000 .
the data used here are publicly available and were retrieved from www.stecf.org/hstprogrammes/j22/j22.html , in its version of november 23 , 2005 . with help of the midas package fitlyman @xcite , which performs line fitting through @xmath221 minimization of voigt profiles , @xcite determined the redshifts , column densities , and doppler widths of the identified absorption features imprinted in the spectra of qso j2233 - 606 . in this way
, they found that the line of sight to qso j2233 - 606 intersects a total of 270 ly@xmath3 forest clouds , and identified other 24 absorption systems containing metal lines .
the ly@xmath3 absorbers span a range in column density of @xmath222 , and a range in doppler parameters of @xmath223 .
the distribution of the parameter pairs @xmath177 for these systems for @xmath224 \ , \mathrm{km \ , s^{-1}}$ ] is shown in figure [ fig : nhi_b ] . using this list of ly@xmath3 line parameters @xmath225 ,
we generate a synthetic spectrum of the qso j2233 - 606 in the wavelength range @xmath226 with a higher resolution than that of the observed spectrum of @xmath227 , using eqs .
( [ eq : tau ] ) , ( [ eq : h_13 ] ) , ( [ eq : h_14 ] ) , and ( [ eq : vhf4 ] ) .
we synthesise a second spectrum using monaghan s algorithm , and compute the logarithmic difference @xmath192 between these two synthetic spectra in the same fashion as in the previous section . in this way
, we test again our method against the highest - precision method available , for a typical range in column densities and doppler parameters present in a qso spectrum . the result is shown in figure [ fig : specdiff ] .
the upper panels of each row show the observed spectrum and the spectrum synthesised using our method , whereas the lower panels show the logarithmic difference between both synthetic spectra .
we choose to cut off the logarithmic difference at @xmath185 , since differences smaller than these are out of the range of the highest available numerical precision .
it can be seen again , as in figure [ fig : dla ] , that the smallest discrepancies are given at the line cores , and the largest , of the order of @xmath0 , are given at the wings ( cf .
discussion of figure [ fig : dla ] , section [ sec : mod ] ) . as can be seen from the column density and doppler parameter distribution in figure [ fig : nhi_b ] ,
the largest discrepancies in this wavelength range are consistent with the maximum absolute differences shown in figure [ fig : map ] .
note that in the spectral regions where no apparent absorption features are found , the logarithmic difference does _ not _ fall to @xmath228 , as one would naively expect .
these features are present in figure [ fig : dla ] as well .
the reason for this valley - shaped features is that , even though having small values away from the line center , the function @xmath13 does _ not _ fall to zero , and thus in these regions the wings of two or more lines overlap . strictly speaking
, in these regions there is always some absorption left , _
i.e. _ @xmath229 , and different methods to compute @xmath13 will account for this effect differently .
again , since our method neglects terms of order @xmath190 , the absorption in these regimes differs from its fiducial value .
the lack of this terms in our approximation to @xmath13 is also pointed out pictorially by the local maxima symmetrically placed around the deeps corresponding to the logarithmic difference at the line cores .
the absorption lines imprinted in the spectra of background sources yield a wealth of information about the physical and chemical properties of the intervening absorbing material , as is the case of intervening neutral hydrogen ( hi ) systems embedded in the intergalactic medium ( igm ) and associated with galaxies and larger structures . in order to extract the desired information from these absorption lines , their profiles have to be modeled in a proper way . in the case of absorption features found on qso spectra ,
absorption line profiles are best modeled by voigt profiles , which are mathematically given in terms of the voigt - hjerting function . in this work
, we presented a simple analytic approximation to the voigt - hjerting function with which voigt profiles can be modeled for an arbitrary range in wavelength ( or frequency ) , column densities up to @xmath230 , and for damping parameters satisfying @xmath231 .
starting with an exact expression for the voigt - hjerting function in terms of harris expansion that is valid for @xmath35 , we showed that the zeroth order term of this series , a gaussian function , is suitable for modeling absorption line profiles emerging in a medium with low column density @xmath232 .
however , for higher column densities , terms of higher order have to be taken into account . a key point leading to this conclusion
is the fact that it is not the damping parameter alone , but rather the factor @xmath156 that determines to which extent terms of order higher than zeroth in harris expansion may or may not be neglected .
we showed that the departure of the actual voigt - hjerting function from the first two terms in harris expansion is of the order of @xmath85 or less for an arbitrary wavelength range and @xmath2 .
hence , we concluded that with an approximation to first order in @xmath4 to the voigt - hjerting function voigt profiles can be modeled with moderate to high accuracy . on this basis , we obtained a simple analytic expression for the voigt - hjerting function and consequently for the absorption coefficient of intergalactic hi , in terms of an approximation for the second term @xmath83 of harris expansion .
the main advantages of the analytic expression we presented here are , first , that it is valid for an arbitrary wavelength range , in the sense that is has no singularities .
in addition , it is simple and flexible to implement in a variety of programming languages to numerically compute voigt profiles with moderate speed and moderate to high accuracy . as a matter of fact ,
our method to compute the voigt - hjerting function is faster with respect to other known standard methods , for instance , huml@xmath159ek s or monaghan s algorithm .
furthermore , our approximation reaches an accuracy of @xmath0 or better in a wide wavelength range , and of the order of than @xmath233 only a negligible wavelength interval , for values of @xmath4 and @xmath24 characteristic to intergalactic hi absorbers .
our method thus offers a great compromise between speed , accuracy , and flexibility in its implementation .
even though we did not extend our discussion in this work to other transitions typically present in quasar absorption spectra and associated to hi absorbers , such as metal lines , our method to synthesise voigt profiles can certainly be applied to most of these elements as well , since their column densities are obviously the same , and their ranges in @xmath4 strongly overlap with the range of @xmath4 for intergalactic hi , for which our approximation to the voigt - hjerting function is valid . as a matter of fact , our approximation is valid to model absorption voigt profiles found in any type of spectrum ( stellar , solar , etc . ) , which arise in a medium whose damping parameter and column density satisfy @xmath2 and @xmath234 , respectively .
i thank uta fritze-v.alvensleben and the gttingen galaxy evolution group for encouraging comments .
special thanks to the referee m.t .
murphy for providing his computational routine to compute voigt profiles , for pointing out some additional references , and for very helpful suggestions which helped to significantly improved the presentation of our results .
this project was partially supported by the _
mexican council for science and technology _ ( conacyt ) , the gttingen graduate school of physics ( ggsp ) , and the _ georg - august - university of gttingen_. | the voigt - hjerting function is fundamental in order to correctly model the profiles of absorption lines imprinted in the spectra of bright background sources by intervening absorbing systems . in this work we present a simple analytic approximation to this function in the context of absorption line profiles of intergalactic hi absorbers . using basic calculus tools
, we derive an analytic expression for the voigt - hjerting function that contains only fourth order polynomial and gaussian functions . in connection with the absorption coefficient of intergalactic neutral hydrogen ,
this approximation is suitable for modeling voigt profiles with an accuracy of @xmath0 or better for an arbitrary wavelength baseline , for column densities up to @xmath1 , and for damping parameters @xmath2 , _
i.e. _ the entire range of parameters characteristic to all lyman transitions arising in a variety of hi absorbing systems such as ly@xmath3 forest clouds , lyman limit systems and damped ly@xmath3 systems .
we hence present an approximation to the voigt - hjerting function that is both accurate and flexible to implement in various types of programming languages and machines , and with which voigt profiles can be calculated in a reliable and very simple manner .
[ firstpage ] methods : analytic , quasars : absorption lines , line : formation , line : profiles , line : identification |
it is widely accepted that the largest fraction of matter in the universe is the dark matter ( dm ) @xcite .
its nature is still unclear , but most likely it consists of yet undiscovered cold ( non - relativistic ) particles , which have very low interaction cross - sections with each other and with baryonic matter @xcite .
so the main physical mechanism governing the evolution of dark matter is gravitation . the dark matter is responsible for formation of large - scale structure of the universe , as well as extended dm haloes of galaxies @xcite .
however , the very centers of galaxies ( except low surface brightness galaxies ) are dominated by baryons , which form galactic bulges .
their influence on dm haloes consists of at least two effects .
firstly , as the baryons cool and settle down in the center of a potential well created by dm , they change the common gravitational potential , which leads to compression of dm halo .
this effect is called baryonic compression and increases dm density several times @xcite . on the other side
, dark matter particles are scattered off by bulge stars and captured by supermassive black holes ( smbhs ) , which reduces their density in the inner parts of bulges @xcite . additionally , a non - zero annihilation cross - section of dm particles also leads to decrease of dm density in the very centers of galaxies and gives possibility of indirect detection of dm through its annihilation radiation @xcite .
all these issues are addressed in the present paper .
the paper is organized as follows . in the second section
we describe previous investigations on this subject and explain motivation of this study . in the third section we present our class of initial dm halo models , which is somewhat broader than in previous papers .
then the kinetic equation governing the evolution of dm is derived , along with its coefficients and boundary conditions .
the fourth section is devoted to solution of the above equation , given for a series of initial halo models with inclusion of different physical processes and approximations .
this helps us to clarify validity of some commonly used assumptions , and determine parameters on which the evolution depends mostly .
predictions for annihilation radiation are also given in this section .
finally , the conclusions are presented .
the question about dark matter content in central areas of galaxies has long been studied .
first of all , it depends on initial structure of dm halo .
there are two principal classes of methods of halo structure modelling .
analytical methods generally consider the nonlinear stage of collapse of initial overdensity region .
these include kinetic theory of phase mixing @xcite or variants of self - similar infall models @xcite . in general , they predict a power - law density profile in the center of a dm halo , @xmath5 where the power - law index @xmath6 is between 1 and 2 .
a significant drawback of this approach is very limited ability to handle mergers , especially major ones . alternatively ,
numerical simulations of dissipationless _
n_body gravitational interaction have been largely used in recent years .
they also generally predict density profiles that have power - law cusps in central areas ( e.g. navarro , frenk , white ( nfw ) profile @xcite with @xmath7 or moore profile @xcite with @xmath8 ) .
other models were proposed in which the density slope tends to zero at @xmath9 ( einasto profiles @xcite or burkert profiles @xcite ) .
there is still no consensus on the question whether dark matter halos are cuspy or cored in the centers , which means that central density profiles obtained in simulations may equally good be described with different fitting formulae , although the inferred densities in the central parsecs would vary orders of magnitude .
the largest simulations of a galaxy - sized halo ( e.g. @xcite ) provide measurements of density no closer to the galaxy center than @xmath10 pc , which is still many orders of magnitude greater than the distances relevant to annihilation and scattering . however , there are other processes that modify dark matter distribution in the galactic centers after halo formation .
first of all , it is now well established that virtually all galaxies harbour supermassive black holes ( smbhs ) in their centers @xcite , whose masses @xmath11 range from @xmath12 to more than @xmath13 .
a black hole ( bh ) immersed in a stellar cluster ( or bulge as a whole ) modifies gravitational potential at distances smaller than the gravitational influence radius of the black hole @xmath14 , defined as containing stellar mass equal to @xmath15 .
if the black hole grew adiabatically , the surrounding matter ( stars or dm ) forms a `` spike '' with power - law index @xmath16 , @xmath2 being the power - law index of density profile before black hole formation @xcite .
the above argument applies for spherically symmetric evolution .
if the black hole formed off - center and then spiralled in , the resulting profile should be shallower @xcite .
additionally , in the hierarchical merging scenario , binary smbhs should be ubiquitous . a binary smbh ejects matter from @xmath17 due to slingshot effect , and after coalescence the stellar and dark matter density around it is reduced @xcite .
however , it is unlikely that a major merger took place in the recent @xmath18 yr of milky way history @xcite .
if the dm consists of self - annihilating particles , the product of annihilation cross - section @xmath19 and relative velocity @xmath20 being weakly dependent on @xmath20 , then the maximum density after a period of time @xmath21 is believed to be set at the so - called annihilation plateau : @xmath22 ( @xmath23 is the particle mass ) @xcite . however , this result is exact only for circular particle orbits . if we consider arbitrary velocity anisotropy @xmath24 ( particulary , isotropic velocity corresponding to @xmath25 ) , then we obtain a weak cusp instead of a plateau : its density varies as @xmath26 and approximately equals corresponding plateau density @xmath27 at the radius where plateau itself should transform into initial density profile .
this issue is explained in detail in @xcite . finally , a very important process that modifies the dark matter distribution is the gravitational scattering of dm particles by stars in the vicinity of the smbh and capture of particles by the smbh .
the detailed analysis of this process is the main subject of the present investigation . the stellar distribution around a massive black hole was much studied in 1970s @xcite .
the usual method includes representation of the distribution function ( df ) @xmath28 in terms of energy @xmath29 and angular momentum @xmath30 per unit mass ( for spherically symmetric problem these are the only two independent variables ) and consideration of diffusion in the @xmath31 phase space . in these early studies
it was shown that diffusion along @xmath30 axis leads to formation of logarithmic dependence of @xmath28 on @xmath30 , so that there is continuous flux of stars towards low values of @xmath30 , where they are captured by black hole .
the diffusion along @xmath29 axis drives the df to a quasi - stationary state @xmath32 , when the loss of stars is balanced by a inward flux from higher values of @xmath29 ( outside the bh sphere of influence ) . the corresponding density profile @xmath33 is called bahcall - wolf cusp , and is indeed observed in the center of our galaxy @xcite , though the density slope is somewhat lower .
it is important that the stationary solution applies only to systems older than two - body relaxation time @xcite @xmath34 where @xmath35 is 1-d velocity dispersion , @xmath36 stellar density , @xmath37 coulomb logarithm .
it appears that in the center of our galaxy @xmath38 and weakly depends on radius at @xmath39 .
( we adopt @xmath40 for consistency with other papers , though recent estimates are somewhat higher @xcite .
this yields @xmath41 pc ) .
dark matter dynamics is governed by the same mechanism of relaxation on stars , but due to negligible particle mass the effect of dynamic friction is unimportant , and resulting density profile should be @xmath42 .
this simple argument was considered in @xcite .
a more elaborate treatment involves time - dependent solution of diffusion equation .
this was first done in @xcite for a constrained initial dm distribution function using only diffusion along angular momentum . in @xcite
the problem was treated as one - dimensional diffusion for energy with inclusion of loss - term due to angular momentum diffusion and capture by bh , which was taken from @xcite . in subsequent papers @xcite the effect of self - annihilations
was also included .
these studies have led to the conclusion that the scattering off by stars drives dm distribution towards a quasi - stationary state with diminishing amplitude of density profile due to combined effect of heating by stars and capture by the smbh .
this density profile is regenerated within one @xmath1 even after a cusp has been destroyed by a binary smbh @xcite .
we want to draw attention to several important issues that were not thoroughly investigated in previous studies . 1 .
the relaxation time actually depends on angular momentum @xmath30 , as will be shown later . for small values of @xmath30
the diffusion along energy goes faster , so that the two - dimensional problem can not be reduced to one - dimensional .
2 . the black hole mass might have grown in time , which moves up the radius of bh influence and changes relaxation timescale inside this radius .
3 . the initial dm distribution function need not be isotropic , i.e. might depend on angular momentum .
all these issues are examined in the present paper .
additionally , the following features are discussed : 1 . a broad class of initial distribution functions
is considered .
this enables direct comparison between different studies that took different initial conditions , and helps to disentangle the effect of initial conditions from other effects investigated .
the diffusion is considered both inside and outside the smbh sphere of influence .
values of diffusion coefficients are different in these two domains .
analytical approximations of evolution of dm density profile and distribution function are made for limiting cases of @xmath43 and @xmath44 .
they are in good agreement with the solution of exact equation .
the two - dimensional diffusion equation is solved directly .
the results are compared to the simplified case of one - dimensional treatment , to validate the possibility of such simplification .
the effect of self - annihilations is included as well .
the proportion of captured , evaporated and annihilated dark matter mass is calculated .
having stated these important goals , we proceed to the investigation itself .
plenty of models have been suggested for dm haloes .
as we are interested in dark matter dynamics in galactic centers , we shall consider only internal part of halo , which can be reasonably approximated by a power - law density profile ( [ rho_d ] ) .
we operate with a distribution function @xmath45 so we need to transform @xmath46 to @xmath45 .
this transformation is essentially non - unique one can obtain different dfs for the same density profile , which differ in dependence on angular momentum @xmath30 .
it is convenient to choose a factorized family of models @xmath47 , where we have changed @xmath30 for a dimensionless scaled quantity @xmath48 $ ] : @xmath49@xmath50 being angular momentum of a circular orbit with given energy .
( this has been suggested in @xcite ) . in the case of power - law density @xmath51
should also be power - law @xcite .
we consider two family of dfs , which differ in dependence on @xmath30 : @xmath52 model a has been proposed in @xcite as a result of analytical treatment of dm halo collapse and phase mixing .
the quantity @xmath53 is related to orbital eccentricities of particles and is small , @xmath54 .
( further discussion can be found in @xcite ) .
model b is the simplest generalization of isotropic distribution for the case of arbitrary radial velocity anisotropy @xmath55 , considered in @xcite . here
@xmath24 is binney s anisotropy parameter @xcite : systems with @xmath25 are isotropic , with @xmath56 have radially biased velocities . for realistic dm haloes ,
@xmath55 in the center is zero or slightly positive @xcite , as indicated also by a density slope
anisotropy relation proposed in @xcite . concerning the density slope
@xmath6 , we shall adopt values @xmath7 ( nfw profile @xcite ) , @xmath8 ( moore profile @xcite ) and @xmath57 for the model b and @xmath58 for model a @xcite .
we use nfw or moore profiles for most of our variants of calculation , since these are the most frequently used in the literature .
these density profiles refer to pure dark matter haloes .
the central part of a halo is likely compressed in the process of galaxy formation and baryonic contraction @xcite , which increases the slope of density profile as well as its normalization . to account for adiabatic contraction
, we transform @xmath45 to @xmath59 , where @xmath60 is radial action a quantity that is conserved during slow change of potential .
hence @xmath59 is conserved during compression . as the initial condition for diffusion equation we take @xmath61 converted back from @xmath59 using now the potential of stars in the bulge and the smbh .
( in what follows we omit primes in @xmath62 and @xmath63 ) .
all calculations are performed for the case of milky way .
the normalization of density profile is taken from the condition that @xmath64 , where @xmath65 kpc is the distance to galactic center , and @xmath66 @xcite . at 1 pc before contraction .
] we summarize initial model data in table [ table_init_models ] .
.initial halo models used in calculations .
@xmath6 is the slope of density profile in the center before baryonic compression , @xmath55 is the velocity anisotropy parameter , @xmath67 and @xmath68 are density values at @xmath69 pc before and after compression , in @xmath70 . [
cols="<,<,<,<,<",options="header " , ] we now turn to one - dimensional diffusion for energy in the absence of loss terms .
first we consider the evolution of @xmath28 in the bulge outside smbh domain of influence . in this region @xmath71 ,
the initial condition has the power - law form ( [ f_init_qr ] ) : @xmath72 ( @xmath73 ) , and the one - dimensional equation for diffusion along energy can be solved analytically .
the solution outside smbh domain of influence may be approximated as @xmath74 @xmath75 is defined to match these two asymptotes .
this means that diffusion effectively blurs @xmath28 for low @xmath76 , driving it towards plateau with ever - diminishing amplitude ( as a power - law of @xmath21 ) .
then we consider the vicinity of smbh , where @xmath77 .
the equation then admits solution in the form @xmath78 , where @xmath79 is the relaxation time ( as said above , it is independent of energy in the case considered , but still depends on @xmath80 : relaxation goes on faster for smaller angular momenta ) .
if we neglect the dependence of @xmath79 on @xmath80 , we may come to the following conclusion : as long as @xmath81 , the evolution of @xmath28 ( and hence density ) is only in amplitude which drops exponentially , not the shape .
when the exponent becomes significant ( or @xmath82 ) , the function tends to constant value which is set by the outer boundary condition ( where it is determined by the solution outside smbh domain of influence described above ) . since the relaxation time in the inner area is always shorter than in bulge , this constant ( with respect to @xmath76 ) value is determined solely by the diffusion in bulge and is decreasing in time .
the constant value of @xmath28 implies the density profile @xmath83 , as mentioned in @xcite .
( this is different in the case of stars , where the effect of dynamic friction should be included in the equation , and the corresponding solution has the bahcall - wolf density profile @xmath33 ) .
however , the profile is not stationary and its amplitude decreases in time ( as noted in @xcite ) , which is attributed to ongoing heating of dark matter particles by stars , continued mainly outside @xmath14 after one relaxation time .
the amplitude then decreases as power - law of time : @xmath84 .
we want to draw attention to the importance of considering both regions ( bulge and smbh sphere of influence ) in order to obtain the density profile and annihilation signal evolution for @xmath85 . for both cases @xmath44 and @xmath86 analytical estimates can be obtained , but for transition period @xmath87 exact calculation is required .
specifically , galactic center is just in this category .
thus far we have considered only one - dimensional approximations , with the diffusion along angular momentum caused by capture of dm particles by the black hole , and diffusion along energy caused by heating on stars . a common approach to solution of the two - dimensional equation
is to consider one - dimensional diffusion along energy with additional loss term due to capture of particles by black hole , caused by diffusion along momentum .
the loss term is taken from the stationary solution of diffusion equation for momentum , and introduced into the equation for energy as an additional term @xmath88 .
the equation itself is written for averaged @xmath89 .
the magnitude of the loss term @xmath88 is taken from the boundary condition at the black hole : @xmath90 , where @xmath91 is related to @xmath92 by ( [ f_near_rg ] ) .
the diffusion coefficient @xmath93 also should be averaged over @xmath80 , despite the fact that it significantly varies with @xmath80 ( see ( [ diff_coefs]a ) ) .
the validity of this reduction should therefore be verified by comparison with solution of full two - dimensional equation , which was one of aims of the present study . in order to address different issues mentioned in section [ sec_overview ]
, we have performed several variants of calculation , which differ in initial conditions and in account for distinct factors of evolution .
all of them are listed in table [ table_variants ] .
the full two - dimensional diffusion equation ( [ diff_eq ] ) was integrated on rectangular grid with variable space- and time - steps .
the boundary condition at @xmath94 was treated in the same way as in @xcite ( section vc ) , i.e. for @xmath95 less than @xmath96 ( the size of first cell ) , flux @xmath88 through absorption boundary was considered as flux through @xmath97 , which was related to the value of @xmath28 in the center of the cell via ( [ f_near_rg ] ) with @xmath98 .
the spatial density profile was recalculated every few timesteps and used to compute orbit - averaged annihilation rate at each @xmath99 in variants 2 and 5 .
the variants are chosen to investigate the relative role of different factors .
five different initial models were chosen ( sec .
[ sec_initmodels ] ) .
the model b3 ( nfw profile with isotropic velocity distribution ) is a basic model to study the influence the one - dimensional approximation ( var.6 ) and growing bh mass ( var.7 ) , which are to be compared with reference var.4 . from the existence of quasars at @xmath100 @xcite
we know that smbhs were already massive at very early times , but the situation is also possible when the black hole grows slowly from small initial seed due to accretion of stars and gas .
the growth law is taken to be @xmath101 @xcite ; the smbh radius of influence and the total gravitational potential are changed accordingly . in other variants @xmath11
was held constant , and gravitational potential remained fixed .
the influence of velocity anisotropy is studied in var.3 for moore profile ( from the @xmath102 relation of @xcite it is naturally to expect more radial anisotropy for steeper profiles ) , which should be compared to var.1 .
additionally , annihilation was accounted for in vars.2 and 5 , with maximal annihilation cross - section of @xmath103 and particle mass @xmath104 gev , same as in @xcite .
these should be compared to vars.1 and 4 , respectively .
in other variants annihilation was disregarded , but predictions of @xmath2-ray flux were made ( see below ) .
the prospects of indirect dm detection are related to registration of @xmath2-rays from the dm annihilation in the most dense regions .
one of such sites is the galactic center , where indeed a @xmath2-ray source was detected by several ground - based cherenkov telescopes such as h.e.s.s .
@xcite , veritas @xcite , cangaroo @xcite and magic @xcite , as well as space telescope egret @xcite .
the @xmath2-ray flux is usually split into two factors , one of them depending on particle physics parameters and the other ( @xmath105 astrophysical factor ) is determined by dm density distribution @xcite .
@xmath106 where @xmath107 , and @xmath108 is dimensionless integral of squared density , averaged over a solid angle @xmath109 : @xmath110 @xmath111 , and integral is taken over line of sight and the solid angle corresponding to telescope resolution .
( for simplicity , the point - spread function is taken as heaviside step function ) .
we calculate values of @xmath112 and @xmath113 corresponding to @xmath114 and @xmath115 .
the former is the approximate resolution of glast @xcite and modern cherenkov telescopes such as hess , the latter is the resolution of egret @xcite .
the spatial resolution is 15 and 150 pc , correspondingly .
it is obvious that for a power - law density profile @xmath116 with @xmath117 the main contribution in ( [ j_av ] ) comes from the inner boundary , and for @xmath118 the situation is inverse . since we have adiabatically compressed dark matter halo , its density slope @xmath119 after contraction is always greater than 1.5 ( @xmath120 , where @xmath2 is initial slope )
therefore , the flux is determined by dark matter distribution in the vicinity of black hole , and the additional contribution from the dark matter in bulge ( @xmath121 ) is negligible .
this is in contrast with the conclusion of @xcite , where the authors did not use adiabatically contracted dark matter profile and hence the contribution from bulge was significant because @xmath118 .
another observational constraint on dark matter distribution may come from the study of stellar orbits which measure the enclosed mass and may help to determine the fraction of dark mass near the smbh @xcite .
@xmath122 @xmath123 @xmath124 the results of calculation are presented in table [ table_variants ] .
the five initial models differ in normalization , i.e. in the initial dark matter mass @xmath125 within @xmath14 .
since the dynamics of dark matter is governed by linear equations ( except for annihilation ) , the results may well be scaled to match each other .
the similarity between different models then becomes apparent .
approximately 1015% of @xmath125 is captured by black hole by @xmath126 yr ( @xmath127 ) ; half of this mass comes from the smbh sphere of influence ( @xmath128 ) .
about 40% of @xmath125 has been evaporated from the central region .
the difference between variants with the same initial model are significant only for @xmath129 yr , which is by coincidence the total time of evolution .
@xmath1 ( [ t_relax ] ) is the relaxation time inside @xmath14 and is independent of radius .
so the current state ( and further evolution ) is basically independent of initial conditions , although if @xmath1 was , say , twice as large , then the difference would be much more prominent by present time .
first we consider basic variant 4 without annihilation .
since the diffusion along energy goes on faster for low angular momenta , the particles with both low energy and low angular momenta are quickly heated up or captured by black hole , therefore the density in the immediate vicinity of the black hole quickly drops .
however , by one relaxation time it decreases roughly by the same factor of few in the whole region @xmath130 ( fig .
[ fig_dens_annihil ] , upper solid ) , except for the very inner radii . by this time
the density profile is still steeper than @xmath131 inside @xmath14 , and hence the integral @xmath132 which determines the @xmath133-factor in ( [ j_av ] ) depends on the low-@xmath134 density behavior .
then the heating starts to dominate , and within few @xmath1 the distribution function and density tend to nearly universal form : for @xmath135 density @xmath136 , near @xmath137 density profile is a little bit shallower , and further away it is identical to initial profile .
so most of @xmath133 is gained at @xmath137 , and hence the difference between variants @xmath138 almost vanishes . the density profile
inside @xmath14 then drops nearly self - similarly ( if we continue integration to @xmath139 yrs , density is reduced about an order of magnitude everywhere at @xmath39 while retaining its form ) .
it is important to note that after this regime has been established , diffusion _ outside _ smbh sphere of influence plays the key role . in the models with initial dominance of low angular momentum orbits ( b2 and a1 )
the initial flux of dark matter into the smbh is larger , but then as the @xmath80-diffusion establishes near - logarithmic profile ( at @xmath4 ) , the memory of initial conditions is erased .
@xmath140 @xmath141 the inclusion of self - annihilation ( in vars .
2 and 5 ) affects only the inner region in the manner that the density does not exceed the `` annihilation weak cusp '' ( not plateau ) .
consequently , the @xmath133 value diminishes faster than without annihilation ( fig .
[ fig_j ] ) , but after @xmath4 the heating starts to prevail over annihilation even in the very inner region , and everything again goes as in basic variant .
the variant 7 which features a black hole growing from small initial seed differs in the way that relaxation time inside @xmath14 is less and a nearly self - similar relaxed solution is established earlier . in this case
the total dark matter mass captured by the black hole is also less than in the case of constant black hole mass ( fig .
[ fig_mcapt ] ) . in all variants
the total captured mass is much less than @xmath11 , so the neglect of the absorption on the black hole mass is justified .
finally , comparison of one - dimensional solution ( var.6 ) to the two - dimensional one ( var.4 ) reveals very little difference in either final result or dynamics .
this is because @xmath80-profile of distribution function at fixed @xmath76 is reasonably well described by a logarithm , failing only at very early ( @xmath43 ) or intermediate ( @xmath142 ) times ( fig .
[ fig_fr ] ) .
therefore , the commonly used approach of reduction to one - dimensional diffusion equation is a good approximation , at least for the class of initial distribution functions that have weak dependency on @xmath80 ( the case of velocity isotropy model b3 being in this class ) .
moreover , the evolution of @xmath80-averaged distribution function is qualitatively consistent with simple analytical description of section [ sec_onedim ] ( fig .
[ fig_fq ] ) .
we have considered the evolution of dark matter distribution in the galactic center caused by gravitational scattering on stars of galactic bulge , absorption by the supermassive black hole and annihilation .
the spherically symmetric evolution is described by two - dimensional fokker - planck equation in phase space of energy and angular momentum .
one of main goals was to investigate how the difference in initial conditions and in various dynamical factors affects the final result .
it turns out that at least for our galaxy there has been enough time to eliminate most of the differences and to establish a nearly universal distribution and density profile with the following features : for @xmath143 pc the density is close to power - law @xmath144 with slope @xmath145 or less , farther out the slope becomes shallower , and even farther it joins the initial profile which is already steeper than @xmath131 because of adiabatic compression in the bulge .
therefore , as regarding the possible annihilation signal , after approximately four relaxation times it is determined by dm distribution near @xmath14 and depends only on initial profile normalization , not the details of evolution .
this explains the low scatter in @xmath133 values in the last column of table [ table_variants ] , the most probable range for @xmath146 being @xmath147 .
( this is in moderate agreement with results of @xcite , though in our calculations the difference between models with no annihilation and maximal annihilation is much smaller ) .
the low scatter in @xmath105 means that constraints on the microphysical nature of dark matter following from eq.([phi_signal ] ) will be more robust , if the @xmath2-ray flux detected from the galactic center is at least partially attributed to dark matter annihilation @xcite
. the different initial models also demonstrate similar proportions of dark matter mass that has been captured by the black hole ( @xmath148 ) and evaporated from the central 2 pc ( @xmath149 of the initial dm mass within 2 pc ) .
a number of previous studies suggested that the dm may constitute a significant fraction of the total smbh mass .
however , in these papers either the loss cone was assumed to be always full @xcite , or the diffusion along energy axis ( heating ) was disregarded @xcite , which led to higher estimate of the captured mass compared to the present study .
it remains in question whether a non - spherically - symmetric initial distribution could sustain a full `` loss cone '' due to chaotic randomization of orbits @xcite .
this issue will be addressed in a future paper .
comparison with paper @xcite where 1-d equation was solved shows qualitatively similar behavior of @xmath80-averaged distribution function and density evolution : for @xmath150 the density drops self - similarly in the smbh region of influence , and for @xmath151 it tends to @xmath152 form .
the quantitative difference might be attributed to different stellar density profile outside @xmath153 : the evolution in the transition spatial ( @xmath154 ) and temporal ( @xmath155 ) regions essentially depends on the magnitude of diffusion coefficients which are determined by stellar density .
there are other targets of indirect dark matter search , except the galactic center , such as nearby galaxies or globular clusters @xcite .
the importance of the processes discussed in this paper on the @xmath2-ray flux from these other targets depends on whether they are dynamically old , i.e. is their relaxation time in the very center less than hubble time . for example , dark matter dominated dwarf spheroidal galaxies such as draco have relaxation times well exceeding @xmath18 yr , so the scattering off by stars is unimportant , and the flux is determined by initial conditions and annihilation cross - section @xcite . on the other hand ,
globular clusters such as g1 in andromeda galaxy are dynamically old , and are expected to be on the relaxed stage of evolution and therefore have less scatter in @xmath105 , hence they are also promising in constraining dark matter parameters @xcite .
( however , in this case one should account for evolution of stellar distribution as well ) .
finally , galaxies on the intermediate stage of evolution such as m32 which has central relaxation time of order @xmath156 yr @xcite may exhibit very different values of @xmath105 ( a drop by many orders of magnitude during several @xmath1 is seen on fig . [ fig_j ] ) .
the authors thank v. l. ginzburg for countenance of this research and a. ilyin , v. sirota and k. zybin for fruitful discussions .
this work was supported by russian fund for basic research ( grant no .
07 - 02 - 01128-a ) and president s grant council ( grants no .
nsh-438.2008.2 and mk-446.2008.2 ) .
j. silk , int .
a * 17 * suppl.1 , 167 ( 2002 ) .
g. bertone , d. hooper , and j. silk , phys . rept . * 405 * , 279 ( 2005 ) .
v. springel , s. white , a. jenkins et al . ,
nature * 435 * , 629 ( 2005 ) .
g. blumenthal , s. faber , r. flores , and j. primack , astrophys .
j. * 301 * , 27 ( 1986 ) .
o. gnedin , a. kravtsov , a.klypin , and d. nagai , astrophys .
j. * 616 * , 16 ( 2004 ) .
j. sellwood and s. mcgaugh , astrophys .
j. * 634 * , 70 ( 2005 ) .
e. vasiliev , jetp lett . *
84 * , 2 ( 2006 ) .
a. ilyin , k. zybin , and a. gurevich ; j. exp .
phys . * 98 * , 1 ( 2004 ) . m. zelnikov and e. vasiliev , int
a * 20 * , 4217 ( 2005 ) .
g. bertone and d. merritt , phys .
d * 72 * , 103502 ( 2005 ) . ya .
zeldovich , a. klypin , m. khlopov , and v. chechetkin , sov .
* 31 * , 664 ( 1980 ) .
p. gondolo and j. silk , phys .
lett . * 83 * , 1719 ( 1999 ) .
g. bertone and d. merritt , mod .
a*20 * , 1021 ( 2005 ) .
y. mambrini , c. munos , e. nezri et al .
, jcap * 0601 * , 010 ( 2006 ) .
a. gurevich and k. zybin , phys .- usp . * 38 * , 687 ( 1995 ) .
a. nusser , mon . not .
r. astron .
* 325 * , 1397 ( 2001 ) .
g. mamon and e. okas , mon . not .
. soc . * 363 * , 705 ( 2005 ) .
c. austin , l. williams , e. barnes et al .
, astrophys .
j. * 634 * , 756 ( 2005 ) .
j. navarro , c. frenk , and s. white , astrophys .
j. * 490 * , 493 ( 1997 ) .
b. moore , t. quinn , f. governato et al .
, mon . not .
. soc . * 310 * , 1147 ( 1999 ) .
j. navarro , e. hayashi , c. power et al .
, mon . not .
349 * , 1039 ( 2004 ) .
d. merritt , j. navarro , a. ludlow , and a. jenkins , astrophys .
j. * 624 * , l85 ( 2005 ) .
g. gentile , a. burkert , p. salucci , u. klein , and f. walter , astrophys .
j. * 634 * , l145 ( 2005 ) .
j. diemand , m. kuhlen , and p. madau , astrophys .
j. * 657 * , 262 ( 2007 ) .
l. ferrarese and h. ford , space science reviews * 116 * , 523 ( 2005 ) .
p. ullio , h. zhao , and m. kamionkowski , phys .
d * 64 * , 043504 ( 2001 ) .
d. merritt , m. milosavljevi , l. verde , and r. jimenez , phys .
88 * , 191301 ( 2002 ) .
e. vasiliev , phys .
d * 76 * , 103532 ( 2007 ) .
j. bahcall and r. wolf , astrophys .
j. * 209 * , 214 ( 1976 ) .
a. lightman and s. shapiro , astrophys .
j. * 211 * , 244 ( 1977 ) .
h. cohn and r. kulsrud , astrophys .
j. * 226 * , 1087 ( 1978 ) .
r. genzel , r. schdel , t. ott et al .
, astrophys .
j. * 594 * , 812 ( 2003 ) .
a. ghez , s. salim , d. hornstein et al .
, astrophys .
j. * 620 * , 744 ( 2005 ) .
r. schdel et al . , astron .
astrophys . * 469 * , 125 ( 2007 ) .
j. binney and s. tremaine `` galactic dynamics '' , princeton university press ( 1987 ) .
f. eisenhauer , r. genzel , t. alexander et al .
, astrophys .
j. * 628 * 246 ( 2005 ) .
o. gnedin and j. primack , phys .
lett . * 93 * , 061302 ( 2004 ) .
d. merritt , phys .
lett . * 92 * , 201304 ( 2004 ) .
d. merritt , s. harfst , and g. bertone , phys .
d * 75 * 043517 ( 2007 ) .
j. an and n. evans , astrophys .
j. * 642 * , 752 ( 2006 ) .
s. hansen and b. moore , new astron . * 11 * , 333 ( 2006 ) .
s. tremaine , k. gebhardt , r. bender et al .
, astrophys .
j. * 574 * , 740 ( 2002 ) .
m. zelnikov and e. vasiliev , jetp lett .
* 81 * , 85 ( 2005 ) .
x. fan , m. strauss , g. richards et al .
j. * 131 * , 1203 ( 2006 ) . v. sirota , a. ilyin , k. zybin , and a. gurevich , jetp * 100 * , 294 ( 2005 ) .
f. aharonian et al . , astron . astrophys . * 425 * , l13 ( 2004 ) . k. kosack et al . , astrophys.j . *
608 * l97 ( 2004 ) .
k. tsuchiya et al . , astrophys .
j. * 606 * , l115 ( 2004 ) .
j. albert et al .
, astrophys . j. * 638 * , l101 ( 2006 ) .
d. hooper , b. dignus , phys .
d * 70 * , 113007 ( 2004 ) .
l. bergstrm , p. ullio , and j. buckley , astropart . phys .
* 9 * , 137 ( 1998 ) .
g. zaharijas and d. hooper , phys .
d * 73 * , 103501 ( 2006 ) .
j. hall and p. gondolo , phys .
d * 74 * , 063511 ( 2006 ) .
f. aharonian et al . , phys . rev .
lett . * 97 * , 221102 ( 2006 ) .
h. zhao , m. haehnelt and m. rees , new astron .
* 7 * , 385 ( 2002 ) .
j. macmillan and r. henriksen , astrophys .
j. * 569 * , 83 ( 2002 ) .
d. merritt and m. poon , astrophys .
j. * 606 * , 788 ( 2004 ) .
m. wood , g. blaylock , s. bradbury et al .
, astrophys .
j. * 678 * , 594 ( 2008 ) .
m. sanchez - conde , f. prada , e. okas et al .
d * 76 * , 123509 ( 2007 ) .
g. zaharijas , phys .
d , * 78 * , 027301 ( 2008 ) . t. lauer , s. faber , e. ajhar , c. grillmair , and p. scowen , astron .
j. * 116 * , 2263 ( 1998 ) . | the evolution of dark matter in central areas of galaxies is considered ( the milky way is taken as an example ) .
it is driven by scattering off of dark matter particles by bulge stars , their absorption by the supermassive black hole and self - annihilation .
this process is described by diffusion equation in the phase space of energy and angular momentum .
the equation was integrated for several different models of initial dark matter distribution and using various assumptions about the dynamical factors .
it turns out that because the milky way center is rather dynamically old ( @xmath0 relaxation times @xmath1 ) , the difference in initial conditions almost vanishes .
the density attains a nearly universal profile , and the @xmath2-ray flux from dark matter annihilation lies in rather narrow range , which enables more robust determination of the dark matter parameters . by present time
the mass of dark matter inside the black hole sphere of influence ( @xmath3 pc ) has been reduced approximately twice , mostly because of heating by stars .
it is shown that the dynamics of dark matter for @xmath4 is determined mainly by stars outside the sphere of influence . |
we are deeply grateful to wojciech czaja for his generous help on this article . research presented in this paper
was supported in part by laboratory of telecommunication sciences .
we gratefully acknowledge this support . 1 a. barrlund , perturbation bounds for the ldlh and lu decompositions . bit ,
358 - 363 , 1991 .
e. bura , r. pfeiffer , on the distribution of the left singular vectors of a random matrix and its applications .
statistics and probability letters , vol .
78 , pp . 2275 - 2280 , 2008 .
d. chandler , w. m. kahan , the rotation of eigenvectors by a perturbation .
siam j. numer .
, vol . 7 , pp . 1 - 46 , 1970 . w. czaja , r. wang , mpsk classification by pca . in preparation
. f. dopico , a note on sin@xmath225 theorems for singular subspace variations . bit ,
395 - 403 , 2000 . c. huang , a. polydoros , likelihood methods for mpsk modulation classification .
ieee transaction on communications , vol .
43 , no . 2 - 4 ,
1493 - 1504 , 1995 .
s. s. soliman , s .- z .
hsue , signal classification using statistical moments .
ieee trrans . commun .
908 - 916 , 1992 .
stein , r. shakarchi , real analysis , measure theory , integration , & hilbert spaces .
3 , p.91 , 2005 .
a. swami , b. m. sadler , hierachical digital modulation classification using cumulants .
ieee trans .
3 , pp 416 - 429 , 2000
. s. j. szarek , condition numbers of random matrices .
j. complexity , vol . 7 , no .
131 - 149 , 1991 .
v. vu , singular vectors under random perturbation .
random struct .
algorithms , vol .
526 - 538 , 2011 .
p. a. wedin , perturbation bounds in connection with singular value decomposition .
bit numer .
99 - 111 , 1972 .
h.weyl , das asymptotische verteilungsgesetz der eigenwerte linearer partieller differentialgleichungen ( mit einer anwendung auf die theorie der hohlraumstrahlung ) .
441 - 479 , 1912 . | we perform a non - asymptotic analysis on the singular vector distribution under gaussian noise . in particular , we provide sufficient conditions of a matrix for its first few singular vectors to have near normal distribution .
our result can be used to facilitate the error analysis in pca .
startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ introduction]introduction the singular value decomposition ( svd ) lies in the heart of various dimension reduction ( dr ) techniques , such as pca , le , lle etc . in real world problems , noise from unknown sourses may largely change the dimension reduction result by changing either the principal directions , or the data distribution along those directions . aside from some unusual cases that involve nonlinear bias in the measurements , and because of the central limit theorem
, it is always assumed that the noise vector in the raw data obeys the i.i.d .
gaussian distribution , which simplifies both data processing and error analysis .
however , when doing dimension reductions , the gaussian distribution is not preserved .
this is because the low dimensional data is contained in the first few singular vectors of some predefined kernel , and elements of any singular vector are bounded by one , which prevent them from being gaussian variables . in @xcite , it is pointed out in a statistical and asymptotic way that for a fixed data matrix , as the noise level goes to 0 , the distribution of singular vectors would eventually become very close to a multivariate normal distribution . in this paper , we go further in this direction and provide non - asymptotic bound on the distance between the perturbed singular vectors and a multi - variate normal random variable . in applications , our result could be used in two ways : * to predict for a fixed data matrix and noise level , whether or not can one reliably ( within a given confidence interval ) assume that the noise after dimension reduction is gaussian .
* for a given noise level , to determine how many samples are necessary for the noise in the embedded space to be close ( e.g. , the distance is less than some given threshold ) to gaussian .
we state the mathematical description of our problem as follows .
+ + * problem * : _ suppose @xmath0 is an @xmath1 noiseless data matrix , and @xmath2 is the observed noisy data .
their associated svds are : @xmath3 and @xmath4 where the entries of @xmath5 are i.i.d . @xmath6 and @xmath7 is an absolute constant . due to the randomness of @xmath5 ,
all the quantities in ( 2 ) are random variables whose measures are induced by those of @xmath5 . under this setting
, what is the distribution of @xmath8 ? _ + in this paper , without loss of generality , we focus only on the low rank and square @xmath0s ( i.e. , @xmath9 ) .
all our techniques can be carried through in high rank and rectangular cases .
the perturbation problem of singular vectors has brought great attention in the field of statistics and numerical analysis .
davis and kahan @xcite studied the deterministic perturbation on hermitain matrices and provides an upper bound for the rotation of singular vector subspaces caused by the perturbation .
they showed that the span of a group of singular vectors as a subspace is changed by an amount which is propotional to the noise power and the reciprocal of eigenvalue gap .
( the change is charaterized in canonical angle between subspaces , denoted by @xmath10 ) .
wedin @xcite extended this result to non - hermitain cases .
dopico @xcite proved that the left and right singular vectors is perturbed towards the same direction .
van vu @xcite considered random perturbations ( i.i.d .
bernoulli distribution ) and provided a tighter upper bound for the canonical angles @xmath10 , which hold with large probability .
the necessity of using the canonical angles to characterize the change is well illustrated in the following example in @xcite . consider the following two matrices , @xmath11 both matrices are diagonalizable , the eigenvectors of @xmath0 are @xmath12 and @xmath13 , while those of @xmath14 are @xmath15 and @xmath16 , for any @xmath17 .
hence , the difference of these two bases does not go to zero with @xmath17 .
+ + beside the canonical angle , some authors used another quantity : @xmath18 to describe the changes of singular subspaces .
the latter has been proved to be an upper bound of the former .
+ in this paper , we use the pointwise matrix norm @xmath19 to bound the difference between two matrices .
our technique is valid and simpler for the frobenius norm case so we omit it .
however , if one tries to use the bound we provide on pointwise norm to derive a bound on frobenius norm using the equivalence of norms , he will get a looser bound than directly running over our technique on frobenius norm , and vice versa .
the rest of the paper is organized as follows . in section 2
, we introduce notations and some existing theorems to be used . in section 3 , we state several lemmas and our main theorem . section 4
contains the proof of the main theorem . in section 5
, we apply our result to an audio signal classification problem .
startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ previous results]previous results throughout this paper , we consider only square data matrices with nontrivial dimensionality @xmath20 and rank @xmath21 .
the variables @xmath0 , @xmath14 , @xmath22 , @xmath23 , @xmath24 , @xmath25 , @xmath26 , @xmath27 , @xmath28 , remain the same as defined in ( 1 ) and ( 2 ) .
we assume @xmath9 .
in addition , we assume that all the diagonal elements of the @xmath29 matrix @xmath30 are bounded away from zero .
+ for a given matrix @xmath31 , we use @xmath32 and @xmath33 to denote the qr decomposition of @xmath31 .
the set of singular values of @xmath31 is denoted by @xmath34 ; the @xmath35th largest one is denoted by @xmath36 ; and the minimum singular value is by @xmath37 .
we use @xmath5 to denote the normalized gaussian matrix whose entries are @xmath38 . for any matrix , @xmath39 denotes the frobenius norm , and @xmath40 the spectral norm .
in addition , we use @xmath19 to denote the component - wise maximum norm of a matrix , i.e. , @xmath41 .
the following theorem is due to dopico @xcite , who provided the worst - case bound on the frobenius norm of the deviation of singular vector subspace under perturbation .
he proved that this bound is proportional to the frobenius norm of the perturbation as well as the reciprocal of the eigenvalue gap .
let @xmath0 and @xmath14 and their svds be as defined in ( 1 ) and ( 2 ) .
define @xmath42 if @xmath43 , then @xmath44 where @xmath45 , @xmath46 .
moreover , the left hand side of ( 3 ) is minimized for @xmath47 where @xmath48 is any svd of @xmath49 , and in this case , the equality can be attained . in the proof of our main theorem ( theorem 4 )
, we will frequently encounter the spectral norm of gaussian matrices , which is known to be bounded in the following theorem .
let w be an @xmath50 matrix with i.i.d .
normal entries with mean 0 and variance @xmath51 .
then , its largest and smallest singular values obey : @xmath52 in order to estimate the gap of eigenvalues in ( 3 ) , we will make use of the following result .
( @xcite ) let @xmath31 be an @xmath50 matrix and @xmath53 be a perturbation of @xmath31 .
let @xmath54 and @xmath55 be the @xmath56th largest eigenvalue of @xmath31 and @xmath53 , respectively .
then , for @xmath57 , @xmath58 startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ main theorem]main theorem we now ready to state our main theorem .
let @xmath0 and @xmath14 and their svds be defined as in ( 1 ) and ( 2 ) .
assume @xmath59 are small enough such that the following defined quantities satisfies @xmath60 and @xmath61 ( see below ) .
then with probability ( with respect to the random gaussian noise w ) exceeding @xmath62 , there exists an unitary @xmath63 matrix @xmath64 , such that for @xmath65 , we have @xmath66 where @xmath67 is a gaussian matrix defined by @xmath68 , and where @xmath69 are defined later in formulas ( 19 ) , ( 20 ) , ( 25 ) , ( 29 ) , ( 17 ) , and where @xmath70 , @xmath71 , and @xmath72 .
note that the order of @xmath73 in terms of @xmath17 and @xmath74 are : @xmath75 , @xmath76 , @xmath77 , @xmath78 and @xmath79 .
if the eigenvalues of @xmath0 are all different , then ( 4 ) becomes @xmath80 where @xmath81 .
* remark 1 * : we observe that the perturbation @xmath8 can be approximated by a gaussian term @xmath82 only when it is the leading term in the error .
the average magnitude of this term has asymptotic order of @xmath83 as @xmath17 and @xmath84 going to 0 , while the order of the leading terms on the right hand side is either @xmath85 or @xmath86 .
therefore , to ensure the gaussian term is higher in order , we need the following condition on the @xmath87 pair : @xmath88 before going into the proof , we first establish three useful lemmas .
the first one is an elementary observation in linear algebra , and the last one is a direct application of theorem 2 , so we omit their proofs . if @xmath89 , then @xmath90 let @xmath91 be the distribution of the product of two independent @xmath6 normal random variables .
let @xmath92 , @xmath93 , ... , @xmath94 be i.i.d .
random variables drawn from @xmath91 . if @xmath20 , we have @xmath95 where @xmath96 .
if @xmath97 , one can verify that for all @xmath98 , @xmath99 we apply markov s inequality ( see eg .
@xcite ) to have : @xmath100 letting @xmath101 in the above formula , we get @xmath102 let @xmath103 be an @xmath1 gaussian matrix whose elements are i.i.d .
@xmath104 .
let @xmath103 be written as @xmath105 with @xmath106 a @xmath29 matrix . then , with probability exceeding @xmath107 , @xmath108 startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ proof of the main theorem]proof of the main theorem the idea of the proof is the following : our goal is to charaterize the perturbed singular vector space @xmath109 , which is a left orthogonal matrix and satisfies @xmath110 , meaning that it , together with @xmath111 , diagonalizes @xmath14 .
thus , if we can construct two other orthogonal matrices which also diagonalize @xmath14 with small error , then , by theorem 1 , they are expected to be good approximations to @xmath109 and @xmath111 .
+ + we start the proof similarly to that in @xcite .
define the random matrix @xmath112 as follows , @xmath113 where @xmath114 and @xmath115 due to the invariant property of gaussian matrix , @xmath112 has the same distribution as @xmath5 .
the asymtotic order of @xmath17 in the error term @xmath116 is not satisfactory , we thus want to further diagonalize @xmath112 by using the following two matrices , @xmath117 with @xmath118 , and @xmath119 with @xmath120
. note that in doing so , we start to differ from @xcite by defining @xmath121 and @xmath122 explicitly .
the second order terms in @xmath121 and @xmath122 are crucial for obtaining ( 4 ) .
+ keeping in mind that @xmath121 and @xmath122 are not yet unitary , we multiply the left hand side of ( 6 ) by @xmath123 on the left , and by @xmath122 on the right , to obtain , @xmath124 where the matrix @xmath125 includes all the terms whose order on @xmath17 is greater than or equal to 3 , so @xmath125 is @xmath126 .
compare ( 7 ) with ( 6 ) , we see the above operation has changed the order of error term from @xmath127 to @xmath126 .
if we denote the eigenvector subspaces corresponding to the two diagonal blocks by @xmath128 , @xmath129 , then they have the following expressions , @xmath130 recall that @xmath131 and @xmath132 denote the qr decomposition of a matrix @xmath31 .
now , we want to orthogonalize @xmath133 and @xmath134 . for that purpose , for both sides of ( 7 ) , we multiply them by @xmath135 on the left , and @xmath136 on the right . this way we obtain : @xmath137 where @xmath138 with a little abuse of notation , we continue to use @xmath125 to denote the error term in ( 10 ) , though it was already changed by the left and right multiplication .
we denote by @xmath139 the svd of the @xmath140 , with @xmath28 . in this case ,
( 10 ) becomes @xmath141 on the other hand , the left hand side of ( 11 ) also satisfies @xmath142 combining ( 11 ) with ( 12 ) , we obtain : @xmath143 moving everything but @xmath0 on the left hand side to the right by multiplying the inverse of each matrix and utilizing the fact that @xmath144 and @xmath145 are orthogonal to each other , we derive : @xmath146 ( 13 ) combined with theorem 1 imply that @xmath109 and @xmath147 are the first @xmath35 left singular vectors of two very similar matrices ( different by the err term ) and that they are close .
keeping this useful result in mind , we first turn to look at the big picture .
our final goal is to approximate by a gaussian variable the difference between @xmath148 and @xmath109 up to a roation : @xmath149 ( we willl define the unitary matrix m explicitly later ) , which can be decomposed as : @xmath150 we insert ( 8) into ( 14 ) to get @xmath151 here , the gaussian term is the second to last term on the right , so we want to prove all other terms are small .
for that purpose , we move the gaussian term to the left and take the component - wise matrix norm on both sides , to have : @xmath152 observe that the left hand side of ( 15 ) is exactly what we want to bound in this theorem . the rest of the proof is divided into three parts to bound each term @xmath153 , @xmath154 , @xmath155 on the right hand side .
+ from ( 2 ) and ( 13 ) , we obtain that @xmath156 and @xmath157 are the left eigenspaces of @xmath14 and @xmath158 , respectively .
thus , we want to use theorem 1 to bound @xmath153 . for this purpose , we fisrt calculate the key parameter @xmath159 in that theorem .
+ + recall @xmath2 . by theorem 2 and theorem 3
, the @xmath56th largest sigular value of @xmath0 and that of @xmath14 obey , with probability over @xmath160 , that @xmath161 equation ( 16 ) implies the following lower bound on @xmath159 : @xmath162 whenever @xmath61 , we can apply theorem 1 to the two svds in ( 2 ) and ( 13 ) , to obtain @xmath163 the matrix @xmath125 , defined in ( 7 ) and modified in ( 10 ) and ( 13 ) , is essentially a sum of several products of gaussian matrices .
applying lemma 1 on @xmath125 and utilizing lemma 3 , we obtain the following bound : @xmath164 holds with probability over @xmath107 whenever @xmath17 is small enough such that @xmath165 . here , @xmath166 and @xmath167 where @xmath70 , @xmath71 , and @xmath72 .
+ now , the right hand side of ( 18 ) has the following bound : @xmath168 we are now ready to define the rotation @xmath64 which first appears in ( 14 ) . comparing ( 18 ) with the first term of ( 14 ) , we have @xmath169 where the explicit form of @xmath170 is given in theorem 1
. + we plug ( 21 ) into ( 18 ) , to obtain : @xmath171 we start with breaking @xmath154 into two parts : @xmath172 we estimate iv and v separately .
@xmath173 observe that the entries of @xmath174 are i.i.d .
@xmath6 and are independent of those in @xmath175
. therefore we can apply lemma 2 to each entry @xmath176 and those of @xmath177 to get , with probability at least @xmath178 , with @xmath179 , @xmath180 for @xmath181 , we first observe the following random upper bound , @xmath182 using ( 24 ) , the union bound , as well as the following inequality , @xmath183 we can estimate the probability that @xmath181 exceeds the value @xmath184 , @xmath185 where @xmath186 is the first element of @xmath187 . therefore , with probability exceeding @xmath188 , we have @xmath189 we combine the estimates of @xmath190 and @xmath181 to get , with probability greater than @xmath191 , @xmath192 the following calculation shows how far away is @xmath133 from unitary .
@xmath193 hence , @xmath194 following a procedure similar to the one we used to obtain the bound in ( 21 ) , we derive the distance between this covariance matrix and the identity matrix in frobenius norm : @xmath195 here , @xmath196 was defined in ( 19 ) .
when @xmath17 is small enough such that @xmath197 , theorem 2.2 in @xcite ] shows that the distance @xmath198 can be bounded by a function of @xmath199 : @xmath200 thus , @xmath201 to estimate ( 27 ) , we first note that from ( 25 ) and the assumption that @xmath202 , we obtain @xmath203 therefore , @xmath204 we insert ( 26 ) and ( 28 ) into ( 27 ) , to arrive at the bound : @xmath205 furthermore , from ( 8) and lemma 3 , it is straightforward to estimate : @xmath206 we combine the above two inequalities to get : @xmath207 we now aggregate the estimates of @xmath153 , @xmath154 and @xmath155 to get ( 4 ) and ( 5 ) .
startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ application]application we use the m - psk ( phase shift keying ) modulation classification problem as an example to show how our result is used to make a sampling strategy and a new classification method
. in the so - called adaptive modulation system , the modulation type is varying with time . when the condition of channels ( such as fading or interference ) changes , the transmitter seeks to find the modulation type which best adapts to the new environment .
meanwhile , if the receiver is not informed of these changes , a modulation classification procedure needs to be carried out as soon as the signal is received . here , we consider the classification problem for the widely used mpsk type of modulations which conveys the data by changing the phase of a carrier wave . here
@xmath64 stands for the number of available phases and usually takes value 2 , 4 , 8 , 16 or 32 .
an mpsk signal @xmath208 has the following mathematical representation : @xmath209 where @xmath103 is called the _ symbol period / duration _ ; @xmath210 is a function supported on @xmath211 $ ] and is called _ baseband waveform _ ; the frequency @xmath212 is called _ carrier frequency _ , and @xmath213 is the _ carrier phase_. all these parameters , whether assumed known or not , are fixed for the duration of the signal .
@xmath214 denotes an additive white gaussian noise with two sided power spectral density @xmath215 .
the information is encoded in the phase parameter @xmath216 .
@xmath64 denotes the possible choice of phase shifts .
if @xmath217 , then each symbol period can encode @xmath218 binary bits .
the most popular cases are @xmath219 and @xmath220 , where the modulation is called 2-psk and 4-psk or more often bpsk and qpsk , respectively .
the classification among m - psk modulations is defined to be the process of detecting the right value of @xmath64 from a received noisy signal @xmath208 . among the literature about mpsk classifications
( see e.g. @xcite ,
@xcite , @xcite ) , there exist many different types of model settings , ranging from the fully blind classifications , i.e. , where all the parameters in ( 30 ) are unknown , to the constrained classfication problems where @xmath64 is the only unknown parameter . for illustration purposes ,
we consider a simple , yet nontrivial , partially blind model assuming that @xmath103 and @xmath221 are known , @xmath222}$ ] , @xmath223 for some unknown integer @xmath35 , @xmath213 is unknown , and @xmath224 is chosen unifromly from @xmath225 .
suppose we digitize the data in the following way .
we take @xmath226 uniform samples from @xmath67 periods of @xmath208 and store them in an @xmath227 matrix @xmath0 , where @xmath228 denote the @xmath229th sample of the @xmath74th period , which has the expression : @xmath230 where @xmath5 is a guassian noise matrix with i.i.d .
entries . when noise is absent ( i.e. , @xmath231 ) , each column of @xmath0 is a function of @xmath224 . since @xmath224 has at most @xmath64 choices , the columns of @xmath0 also have only @xmath64 patterns .
if we deem each column as an @xmath170 dimensional data point , then it merely is a high dimensional embedding of a zero dimensional parameter space . in other words ,
the @xmath170 dimensional graph of @xmath0 consists only of @xmath64 points . when noise is added , these @xmath64 points becomes @xmath64 clusters .
hence , the classification problem of finding the correct @xmath64 is reduced to a clustering problem of finding the total number of clusters .
the complexity of all well known clustering methods , such as k - means and mean shift grows exponentially with dimension .
therefore , for large data sets , a preprocessing step of dimension reduction is necessary .
the dimension reduction procedure has another two advantages over other well known methods ( e.g. , @xcite , @xcite , @xcite ) : * no carrier removal procedure is needed , relaxing the digitization rate to below the nyquist rate of the carrier frequency . *
classification and detection are completed simultaneously . for our problem setting ,
pca is the most suitable technique for the following reasons : * the signal has gaussian noise and pca is just a linear - gaussian latent variable model .
* this is a linear model when we deem @xmath232 as parameters of @xmath0 ( we will provide more discussion on this observation later ) . in telecommunications ,
the 2-d graph of all evaluations of @xmath232 with @xmath233 is formally called the constellation diagram of the mpsk modulation ( see figure 1 ) , and can be used as an identifier of the modulation type . therefore , the idea of our method is using pca to map the high dimensional data to the two - dimensional constellation diagram on which clustering algorithm is applied to find the true cluster quantity .
now we provide more detail to legistate the linear model observation . by definition , the matrix @xmath0 has the following decomposition : @xmath234 where for @xmath235 , @xmath236 , and @xmath237 , @xmath238 it is immediate to verify that when @xmath239 , we have @xmath240 , and @xmath181 is nearly orthogonal since we have assumed that @xmath224 are choosen unifromly at random from the parameter space ( rigorous calculation of the distance between @xmath181 and an orthogonal matrix can be found in [ 8 ] ) .
this decomposition clearly shows how the noiseless part of @xmath0 linearly depends on @xmath232 .
when noise is absent , pca does the job of seperating @xmath241 from @xmath242 , where the latter is just a scaled version of the constellation . with noise
is added , as explained earlier , each individual point in the graph turns into a cluster ( figure 2 ) .
even though it is quite obvious to a human observer how many clusters there are in figure 2 without any prior knowledge , the exisiting clustering algorithms are surprisingly inferior by requiring either the number of clusters or the cluster radius as input . when the number of clusters is unknown as in our setting , many previous work suggest to do brute force on all the possible numbers ( which are 2 , 4 , 8 , 16 , 32 , maybe also 64 , 128 ) , and compare the clustering results in some ad hoc way to decide which is more likely . a more pleasant way to do this is by finding the cluster radius , which is exactly the place to use our results in this paper . before applying our results , we first normalizate the singular values of @xmath0 by setting @xmath243 , so that the singular values of @xmath14 do not change with the matrix size , @xmath244 in hardware implementations , a larger value of @xmath170 is usually more difficult to realize than that of @xmath67 , because @xmath170 corresponds to the sampling rate and @xmath67 is the sample duration . hence , we assume @xmath245 . a generalized version of theorem 4 for rectangular matrix can be similarly built , by padding zeros to form an @xmath246 matrix . since now
the factor @xmath247 in ( 34 ) is a normalized gaussian matrix , the other factor , @xmath248 , denotes the energy of noise , corresponding to @xmath17 in theorem 4 . as @xmath249 , @xmath250 .
theorem 4 implies that for a given matrix , and a given confidence level , there exists a threshold @xmath251 such that , as long as @xmath252 , we can assume the first two eigenvectors of @xmath14 have multivariate normal distribution . in general , @xmath251 is a function of both @xmath170 and @xmath67 , which makes the inquality @xmath253 difficult to solve .
fortunately , in our model , where @xmath254 is uniformly bounded for all @xmath67 , it can be verified from ( 5 ) that @xmath251 is universal for all sizes of @xmath0 . thus solving the simple inequality @xmath255 gives us the feasible region of @xmath256 . with any feasible @xmath170
, we can apply the mean shift clustering method with gaussian kernel . by some elementary calculations
, one can derive that the @xmath257 percentile of the gaussian noise on each data point in the embedded space equals @xmath258 .. we set this number to be the radius .
since the rank of @xmath0 for bpsk signals is one , then trying to invert the second singular value in theorem 4 makes the problem near singular and cause large error in the second dimension as shown in figure 2 .
fortunately , this case is easy to be recogonized by simply examing whether the second singular value of @xmath0 is much smaller than the first one . in our first experiment
, we generate a qpsk signal with carrier frequence of 1ghz , symbol rate 10mhz , and damped by awgn with @xmath259 .
we use the sampling rate 21 samples per symbol ( much lower than the nyquist rate of the carrier frequence , and satisfies @xmath260 ) and sample 200 symbols . in figure 3 , the result of pca is plotted , together with a circle whose radius is the @xmath257 percentile predicted by the theorem , and whose centers are those found by the meanshift algorithm .
we can see that the prediction of radius is quite accurate . in our second experiment
, we let the snr decrease and examine the performance of the above algorithm .
an classification is deemed as successful only when the number of clusters returned by the mean shift algorithm @xmath261 strictly equal to the true type @xmath64 .
the result is plotted in figure 4 .
as expected , when noise grows , the eigenvector distribution deviates from the gaussian distribution and the predicted radius becomes too small for the algorithm to find the correct @xmath64 .
startsectionsection1@ -3.5ex plus -1ex minus -.2ex 2.3ex plus.2ex * * [ conclusion]conclusion in this paper , we provided a condition under which the perturbation of the principal eigenvectors of a matrix under gaussian noise has a near - gaussian distribution .
the condition is nonasymptotic and is useful in application .
we provided an simple example of audio signal classification problem to illustrate how our theorem can be used to make sampling strategy and to form new classification technique .
more details about this new classification scheme is discussed in @xcite . |
it is well known that in classical statistical mechanics the evolution of the probability densities in phase space @xmath0 is given by the liouville equation @xmath1 where @xmath2 is the liouville operator @xmath3 and @xmath4 is the hamiltonian in the phase space @xmath5 of the system . in @xcite and @xcite
kvn formulated classical mechanics in a hilbert space made up of complex square integrable functions over the phase space variables @xmath6 .
in particular they postulated , as equation of evolution for @xmath6 , the liouville equation itself : @xmath7 starting from ( [ liopsi ] ) it is easy to prove that , since the liouvillian @xmath2 contains only first order derivatives , the liouville equation ( [ liorho ] ) for the probability densities @xmath0 can be derived via the postulate @xmath8
. finally kvn imposed on the states of their hilbert space the following scalar product : @xmath9 with this choice the liouvillian @xmath2 is a hermitian operator . therefore @xmath10 is a conserved quantity and @xmath11 can be consistently interpreted as the probability density of finding a particle in a point of the phase space .
now every theory formulated via states and operators in a hilbert space can be formulated also via path integrals .
it was proved in ref .
@xcite and @xcite that it is possible to describe the evolution ( [ liopsi ] ) in the following manner : @xmath12 where we have indicated with @xmath13 all the phase space variables .
the kernel of propagation @xmath14 has the following path integral expression : @xmath15 \label{path}\ ] ] where the double prime in @xmath16 indicates that the integration is over paths with fixed end points in @xmath17 .
the @xmath18 in the weight of the path integral ( [ path ] ) is given by : @xmath19 having a path integral we can introduce the concept of a commutator as feynman did in the quantum case : given two functions @xmath20 and @xmath21 , we can evaluate the following quantity under the path integral ( [ path ] ) : @xmath22 \equiv\lim_{\epsilon\to 0}\langle o_{\scriptscriptstyle 1}(t+\epsilon)o_{\scriptscriptstyle 2}(t)-o_{\scriptscriptstyle 2 } ( t+\epsilon)o_{\scriptscriptstyle 1}(t)\rangle$ ] .
what we get is @xcite : @xmath23=0,\qquad\quad [ \hat{\lambda}_a,\hat{\lambda}_b]=0 , \qquad\quad [ \hat{\varphi}^a,\hat{\lambda}_b]=i\delta_b^a . \label{comm}\ ] ] the first commutator of ( [ comm ] ) tells us that the positions @xmath24 commute with the momenta @xmath25 , i.e. that we are doing classical and not quantum mechanics .
the last commutator instead tells us the @xmath26 are something like the momenta conjugate to @xmath27 . in order to satisfy ( [ comm ] )
we can use the representation in which @xmath27 is a multiplicative operator and @xmath26 a derivative one : @xmath28 via the previous operatorial realization , the @xmath18 of ( [ liouvilliano ] ) can also be turned into an operator : @xmath29 therefore @xmath2 is just the liouville operator of eq .
( [ liouv ] ) .
this confirms that the operatorial formalism lying behind the path integral ( [ path ] ) is nothing more than the kvn one .
the main problem we are interested in this paper is the quantization of classical mechanics , once it is formulated via operatorial or path integral techniques .
we know that usually the quantization of a system is performed via the dirac s correspondence rules , i.e. by replacing the classical poisson brackets @xmath30 by commutators according to the following relation : @xmath31_{\scriptscriptstyle qm}}{i\hbar}.\ ] ] now , in classical mechanics formulated la kvn , the poisson brackets are already replaced by the kvn commutators ( [ comm ] ) .
so the quantization of the system can be performed : either finding suitable rules to go from the kvn commutators to the quantum ones ; or finding a way to reproduce the algebra of quantum observables using kvn commutators and operators.in this paper we will find a compact way to implement the second alternative , by using the fact that the set of hermitian operators in the kvn theory includes also operators depending on @xmath32 and not only the functions @xmath33 .
in particular in * sec .
2 * we will introduce the following map : @xmath34 where the domain of @xmath35 is made up of all the standard classical observables @xmath33 while the image of @xmath35 is made up of suitable hermitian operators living in the kvn hilbert space and depending also on @xmath32 . using the kvn commutators ( [ comm ] ) all the functions of the form @xmath36 which appear on the rhs of ( [ mapq ] ) satisfy exactly the algebra of the observables of quantum mechanics .
so we can say that , among all the operators of the kvn hilbert space , there is a set whose algebra is isomorphic to the algebra of quantum observables .
the most interesting feature of our approach is the split between the _ quantum _ observables of the rhs of ( [ mapq ] ) and the _ classical _ ones @xmath33 . in particular in * sec .
3 * we will show that , while the classical energy @xmath37 commutes with the liouvillian @xmath2 , the associated quantum energy is not conserved under the evolution generated by @xmath2 . therefore , in order to preserve the conservation of energy , we must replace also the liouvillian @xmath2 with a more complicated object @xmath38 which is well known in the literature because it is the moyal operator which appears in the equation of evolution of the wigner functions @xcite .
this replacement is the first step necessary in order to go from the classical path integral ( [ path ] ) to the quantum one .
the second step is a sort of polarization prescription which reduces the kvn hilbert space to the standard hilbert space of quantum mechanics . in * sec .
4 * we will see how this polarization procedure can be justified by requiring the irreducibility of the hilbert space of quantum mechanics under the action of the _ quantum _ observables . in the * conclusions * we will outline some of the possible future applications of the content of this paper .
usually physicists identify the observables of classical mechanics with the functions of the phase space variables @xmath39 . in the operatorial approach to classical mechanics
such observables become the functions of the commuting operators @xmath27 .
therefore the algebra of classical observables is abelian and there is no uncertainty principle involving the operators @xmath27 which can be diagonalized simultaneously .
every simultaneous eigenstate of @xmath27 determines uniquely a point in the phase space @xmath40 which represents the state of the system . on such a state
all the observables of classical mechanics assume a well defined value .
in fact if @xmath41 then @xmath42 , i.e. the eigenstate @xmath40 of @xmath43 is an eigenstate also for all the observables @xmath44 and the associated eigenvalue is just the function @xmath45 evaluated on the point of the phase space @xmath46 . in this approach
the mean values and the probability distributions of the classical observables are completely independent of the phases of the wave functions @xmath47 and only the modulus @xmath48 is significant from a physical point of view @xcite . besides the operators @xmath27 in the operatorial formulation of classical mechanics we have also the operators @xmath26 which allow us to generate a non trivial evolution via the liouvillian @xmath18 of eq .
( [ liouvilliano ] ) .
such operators @xmath26 do not commute with @xmath27 and this immediately suggests that it might be possible to construct a non abelian algebra of operators by considering suitable combinations of @xmath43 and @xmath32
. in particular it would be very interesting to discover whether it is possible to reproduce , using only the kvn commutators ( [ comm ] ) , the non abelian algebra of quantum observables . in order to do this
we can define the following operators , known in the literature on wigner functions as bopp operators , @xcite : @xmath49 via the kvn commutators ( [ comm ] ) it is very easy to prove that they satisfy the usual heisenberg algebra : @xmath50=i\hbar\delta_{jk}$ ] .
since both @xmath43 and @xmath32 are hermitian under the kvn scalar product ( [ scalar ] ) we have that also @xmath51 and @xmath52 are hermitian operators with the same scalar product . not only , but the _ classical _ operators @xmath53 and @xmath54 can be obtained from the associated _ quantum _ ones in the limit @xmath55 .
we would like to stress that the _ quantum _ positions and momenta are now different entities " than the _ classical _ ones .
this may be difficult to accept but at the same time it may be the crucial element which makes quantum mechanics so counter - intuitive and different from classical mechanics .
the rules ( [ rules ] ) can be easily generalized and the quantum operator @xmath56 associated to the classical observable @xmath33 will be given by the following function : @xmath57 since @xmath43 and @xmath32 do not commute it is clear that in ( [ capeffe ] ) we must specify also the ordering used .
the one we choose , once we expand ( [ capeffe ] ) in @xmath58 , is the following : @xmath59 the operators @xmath56 are hermitian under the kvn scalar product ( [ scalar ] ) and , since the basic commutators @xmath50=i\hbar\delta_{jk}$ ] are satisfied by the choice ( [ rules ] ) , the hermitian functions of @xmath51 and @xmath52 given by ( [ capeffe ] ) or ( [ expansion ] ) will satisfy the correct quantum commutators . for example the three components of the angular momentum turn out to be : @xmath60 -\frac{1}{4}\hbar^2\bigl[\hat{\lambda}_{p_y}\hat{\lambda}_z-\hat{\lambda}_y\hat{\lambda}_{p_z}\bigr]\medskip \\ \displaystyle \hat{m}_y=\hat{z}\hat{p}_x-\hat{x}\hat{p}_z-\frac{1}{2}\hbar\bigl[\hat{\lambda}_z\hat{x}-\hat{\lambda}_{x } \hat{z}+\hat{\lambda}_{p_z}\hat{p}_x-\hat{\lambda}_{p_x}\hat{p}_z\bigr ] -\frac{1}{4}\hbar^2\bigl[\hat{\lambda}_{p_z}\hat{\lambda}_x-\hat{\lambda}_z\hat{\lambda}_{p_x}\bigr]\medskip \\ \displaystyle \hat{m}_z=\hat{x}\hat{p}_y-\hat{y}\hat{p}_x-\frac{1}{2}\hbar\bigl[\hat{\lambda}_x\hat{y}-\hat{\lambda}_{y } \hat{x}+\hat{\lambda}_{p_x}\hat{p}_y-\hat{\lambda}_{p_y}\hat{p}_x\bigr ] -\frac{1}{4}\hbar^2\bigl[\hat{\lambda}_{p_x}\hat{\lambda}_y-\hat{\lambda}_x\hat{\lambda}_{p_y}\bigr ] \end{array}\ ] ] and they obey the usual algebra @xmath61=i\hbar \epsilon_{ijk}\hat{m}_k$ ] under the kvn commutators ( [ comm ] ) .
all this confirms that it is possible to build the quantum observables @xmath56 by adding to the associated classical ones @xmath33 a suitable combination of kvn operators which vanishes in the classical limit : @xmath62 .
as we have seen in the previous section , in classical mechanics the observable energy is given by the operator @xmath37 .
according to the map @xmath35 of ( [ mapq ] ) at the quantum level the energy has instead to be identified with the operator @xmath63 where the classical position @xmath24 and momentum @xmath25 are replaced by the associated bopp operators @xmath64 and @xmath65 .
now at the classical level the energy @xmath37 is a conserved quantity since it commutes with the liouvillian @xmath2 which is the generator of the time evolution : @xmath66=-i\partial_ah\omega^{ab } \partial_bh=0.$ ] what happens at the quantum level ? does the operator @xmath63 still commute with the liouvillian @xmath2 ?
the answer is no .
in fact , using ( [ expansion ] ) , we have that the commutator of the liouvillian with the quantum energy @xmath63 is : @xmath67&=&-\frac{\hbar^2}{8}\omega^{\alpha\beta } \omega^{a_{\scriptscriptstyle 1}b_{\scriptscriptstyle 1}}\omega^{a_{\scriptscriptstyle 2 } b_{\scriptscriptstyle 2}}(i\hat{\lambda}_{a_{\scriptscriptstyle 1 } } \hat{\lambda}_{a_{\scriptscriptstyle 2 } } \partial_{\alpha}\partial_{b_{\scriptscriptstyle 1 } } \partial_{b_{\scriptscriptstyle 2}}h\partial_{\beta}h+\nonumber\\ & & + \hat{\lambda}_{\alpha}\partial_{\beta } \partial_{a_{\scriptscriptstyle 1 } } \partial_{a_{\scriptscriptstyle 2}}h\partial_{b_{\scriptscriptstyle 1 } } \partial_{b_{\scriptscriptstyle 2}}h)+o(\hbar^3 ) . \label{notcons}\end{aligned}\ ] ] for hamiltonians more than quadratic in @xmath68 and @xmath69 the previous commutator is different from zero and the energy is not conserved under the evolution generated by the liouvillian .
therefore if we identify the quantum energy with @xmath63 and we require that this quantity has to be conserved in time , then we must consistently change also the operator which generates the time evolution .
in particular in order to compensate the rhs of ( [ notcons ] ) we must add to the liouvillian a term cubic in @xmath70 and in the symplectic matrix : @xmath71 . in fact the commutator of such a term with @xmath63 : @xmath72&= & \displaystyle \frac{\hbar^2}{8}\omega^{\alpha\beta } \omega^{a_{\scriptscriptstyle 1}b_{\scriptscriptstyle 1 } } \omega^{a_{\scriptscriptstyle 2}b_{\scriptscriptstyle 2 } } ( i\hat{\lambda}_{a_{\scriptscriptstyle 1}}\hat{\lambda}_{a_{\scriptscriptstyle 2 } } \partial_{\alpha}\partial_{b_{\scriptscriptstyle 1 } } \partial_{b_{\scriptscriptstyle 2}}h\partial_{\beta}h+\nonumber\\ & & + \hat{\lambda}_{\alpha}\partial_{\beta } \partial_{a_{\scriptscriptstyle 1 } } \partial_{a_{\scriptscriptstyle 2}}h\partial_{b_{\scriptscriptstyle 1 } } \partial_{b_{\scriptscriptstyle 2}}h)+o(\hbar^3 ) \nonumber\end{aligned}\ ] ] is just the opposite of the rhs of ( [ notcons ] ) .
if we consider also the other terms of the expansion in @xmath58 we have that @xmath63 is conserved if we replace the liouvillian @xmath73 with the following operator : @xmath74 from ( [ arr ] ) we see that @xmath38 is given by the liouvillian ( @xmath75 ) plus a sum of terms which are proportional to the even powers of @xmath58 and in the limit @xmath55 we have that @xmath38 reduces to the liouvillian @xmath2 itself . at the path integral
level this first step of replacing @xmath2 with @xmath38 implies that the kernel of propagation ( [ path ] ) must be replaced by @xcite : @xmath76 .
\label{kernelg}\ ] ] if we include , besides @xmath70 and @xmath17 , also the forms @xmath77 and the vector fields then the associated path integral has been worked out in ref . @xcite .
since the kinetic term of ( [ kernelg ] ) is the usual one , @xmath78 , the kvn commutators ( [ comm ] ) and their operatorial realization ( [ realiz ] ) are unchanged .
if we limit ourselves to the case of a 2d phase space with a hamiltonian quadratic in the momenta , @xmath79 , then the operatorial realization of @xmath80 becomes : @xmath81 before going on it is interesting to note that @xmath38 is just the operator which appears in the equation of evolution of the wigner functions @xmath82 , see ref .
@xcite : @xmath83 . in any case , in order to avoid possible misunderstandings , we want to stress that in our approach the operator @xmath38 replaces the liouvillian @xmath2 in making the evolution of the elements of the kvn hilbert space , which are the probability amplitudes @xmath6 and not the quasi - probability distributions @xmath82 of the wigner approach .
let us now define , besides the bopp operators of eq .
( [ rules ] ) , the following operators : @xmath84 it is then possible to prove @xcite that the @xmath38 of ( [ arr ] ) can be written in terms of the standard hamiltonian @xmath85 as : @xmath86 where the expansion in @xmath58 of @xmath63 is given by the rhs of ( [ expansion ] ) with @xmath45 replaced by @xmath85 while the expansion in @xmath58 of @xmath87 is : @xmath88 from ( [ difference ] ) it is also easy to understand the reason why the energy @xmath63 commutes with the operator @xmath89 . in fact
@xmath38 is proportional to the difference of the quantum energy itself @xmath63 and an operator , @xmath87 , which commutes with the quantum energy because the unbarred operators @xmath90 of ( [ rules ] ) always commute with the barred ones @xmath91 of ( [ delta ] ) . since the operator of evolution @xmath38 is now expressed via the operators ( [ rules])-([delta ] ) , it is quite natural to write down also the path integral ( [ kernelg ] ) in terms of @xmath92 and @xmath93 . since
@xmath24 and @xmath94 are coupled together in the definition of @xmath64 and @xmath95 , it is more convenient to use the representation in which the multiplicative operators are @xmath24 and @xmath94 instead of @xmath24 and @xmath25 @xcite . in this representation
the kvn hilbert space is made up of functions of @xmath96 and their kernel of propagation becomes : @xmath97 .
\nonumber\end{aligned}\ ] ] the inverse of ( [ rules])-([delta ] ) are given by : @xmath98 using them we can rewrite the kernel of propagation ( [ beta ] ) as : @xmath99\biggr]\cdot\nonumber\\ & \hspace{-0.2cm}&\hspace{-0.2 cm } \int { \cal d}^{\prime\prime}\bar{q}\frac{{\cal d}\bar{p}}{\hbar}\ ; \textrm{exp}\ , \biggl[-\frac{i}{\hbar}\int dt \bigl[\bar{p}\dot{\bar{q}}-h(\bar{q},\bar{p})\bigr]\biggr ] . \label{trentuno}\end{aligned}\ ] ] this path integral generates the evolution of the square integrable functions @xmath100 . from ( [ trentuno ] ) it is easy to note that the evolution of the variables @xmath101 is completely decoupled from the evolution of the variables @xmath102 .
so if we consider as initial wave functions those @xmath100 which can be factorized as @xmath103 , then they will remain factorized during the time evolution .
in fact the variables @xmath101 and @xmath102 can not become entangled because in ( [ trentuno ] ) there is no interaction between them .
furthermore if we limit the kvn hilbert space to the one whose basis is given by all the eigenstates @xmath104 of the operator @xmath64 , then from ( [ trentuno ] ) we have that the wave functions @xmath105 evolve just with the usual quantum kernel of evolution for a system described by a hamiltonian @xmath106 : @xmath107\biggr].\ ] ] in the next section we will analyze the reasons why we have to restrict the kvn hilbert space to the usual one in the quantum case and we will show that also this prescription is just a consequence of the choice of the observables ( [ mapq ] ) .
in this section we will show how the usual hilbert space of quantum mechanics can be embedded in the kvn hilbert space .
let us suppose we use in quantum mechanics the coordinate representation : the positions @xmath64 are operators of multiplication by @xmath108 , the momenta @xmath65 are given by @xmath109 , while the hilbert space is made up of the complex and square integrable functions of @xmath108 .
how can we obtain all this starting from the kvn states and operators ? as we have already seen in the previous sections , since @xmath64 is a linear combination of @xmath24 and @xmath110 , it is more convenient to use , at the kvn hilbert space level , the representation in which @xmath24 and @xmath110 are multiplicative operators while , to satisfy ( [ comm ] ) , @xmath111 and @xmath25 must be given by @xmath112 and @xmath113 respectively . in this representation the hilbert space of the theory
is made up of the complex wave functions @xmath114 which are square integrable under the kvn scalar product @xmath115 .
equivalently , if we use the variables @xmath108 and @xmath116 , the kvn hilbert space will be given by all the wave functions @xmath100 which are square integrable under the scalar product : @xmath117 .
the quantum position @xmath118 multiplies a wave function @xmath100 by @xmath119 while @xmath120 becomes the following operator : @xmath121 .
it is easy to realize that @xmath64 and @xmath65 act on the part in @xmath108 of the wave functions @xmath100 just as the quantum observables act on the standard wave functions of quantum mechanics .
in fact : @xmath122 in general the action of the quantum observable @xmath56 on @xmath100 is given by : @xmath123 clearly the space of the complex and square integrable wave functions @xmath100 is too big in order to be an irreducible representation of the algebra of quantum observables .
in fact , from ( [ effe ] ) we see that every quantum observable @xmath124 modifies only the part in @xmath108 of the wave function @xmath100 while it acts as the identity operator on the part in @xmath116 . therefore if we require that the quantum operators @xmath64 and @xmath65 act _
( see page 434 of @xcite ) on a space of states @xmath125 then we can not identify @xmath125 with the full kvn hilbert space of the square integrable wave functions @xmath100 .
in fact there exists at least one non - trivial subspace of the kvn hilbert space which is invariant under the action of the quantum positions @xmath64 , the quantum momenta @xmath65 and of every other quantum observable @xmath126 . this non - trivial subspace
is given by the set of _ all _ the possible square integrable functions in @xmath108 times one
_ particular _ square integrable function in @xmath116 , let us call it @xmath127 : @xmath128 do not modify the function @xmath127 while they map a wave function of the form @xmath129 in another of the same type @xmath130 .
in other words the quantum observables map every vector of @xmath131 in another vector of @xmath131 .
since the function @xmath127 is fixed the space @xmath131 is isomorphic to the usual hilbert space of quantum mechanics : @xmath132 also the kvn scalar product between two different states of @xmath131 , let us say @xmath133 and @xmath134 : @xmath135 induces the standard scalar product of quantum mechanics @xmath136 if the fixed state @xmath137 is normalized according to : @xmath138 . before going on let us notice that the request of irreducibility is crucial in order to guarantee a one to one correspondence between the states of the theory and the vectors of the space @xmath131 @xcite . in fact let us suppose we consider an arbitrary state belonging to @xmath131 and we replace in it the function @xmath127 with a _ different _ square integrable function @xmath139 .
what we obtain is a _ different _ state of the kvn hilbert space which corresponds to the _ same _ expectation values for _ all _ the quantum observables @xmath140 , i.e. which corresponds to the _ same _ physical situation . to avoid this redundancy we have to restrict the kvn hilbert space to _ only _ one of the non - trivial subspaces which are invariant under the action of @xmath64 and @xmath65 .
it is clear that the choice of this invariant subspace is not unique .
for example we could choose the @xmath131 of ( [ aro ] ) in which the wave function @xmath127 is fixed or the subspace latexmath:[\[{\bf h}_{\sigma}=\biggl\ { \psi(q)\sigma(\bar{q})\ ; \textrm{with } \
, \int dqd\bar{q}\ , @xmath139 is fixed or every other subspace of the same form .
anyway it is easy to realize that all these subspaces are isomorphic to the standard quantum hilbert space of ( [ quantumhil ] ) and that they are completely equivalent from a physical point of view .
these considerations do not depend on the particular representation that we use .
for example , in order to obtain the momentum representation of quantum mechanics , we have first of all to consider the @xmath142 representation of the kvn theory . in this representation @xmath111 and @xmath25 are multiplicative operators , @xmath143 and the wave functions are of the form @xmath144 or , using the bopp variables , @xmath145 . on these wave functions the positions @xmath64 act as derivative operators : @xmath146 while the momenta @xmath65 act as operators of multiplication by @xmath147 .
also in this representation if the operators @xmath64 and @xmath65 must act irreducibly on @xmath125 , then we have to identify @xmath125 with the space of all the square integrable functions which depend only on the quantum momenta @xmath147 times a particular @xmath148 . such a space is again isomorphic to the usual quantum hilbert space of the square integrable wave functions in the momentum representation @xmath149 . before concluding this section we want to stress that it is the quantization map " itself @xmath150 of eq .
( [ mapq ] ) which implies that we must change the operator of evolution by replacing the liouvillian @xmath2 with the operator @xmath38 in order to have a conserved quantum energy .
moreover we must reduce the kvn hilbert space to the usual hilbert space of quantum mechanics in order that the positions @xmath64 and the momenta @xmath65 constructed via our map ( [ mapq ] ) form an irreducible set of operators .
so the role of the map ( [ mapq ] ) in this new approach to quantization of classical mechanics is crucial .
in this paper we have shown how to embed quantum mechanics in the kvn hilbert space , which was first introduced to give an operatorial formulation of classical mechanics .
in particular we have proved that the kvn hilbert space turns out to be a suitable mathematical framework for a unifying picture of classical and quantum mechanics : in fact among all the kvn hermitian operators there are both the classical observables @xmath33 and the quantum ones @xmath151 , see the figure below .
[ kkk ] as a consequence of the choice of the observables in classical mechanics only the moduli of the kvn wave functions @xmath152 are physically significant @xcite while in quantum mechanics , if we impose the condition that @xmath64 and @xmath65 are an irreducible set of operators , the subspace of the physically significant kvn wave functions is isomorphic to the space of the square integrable wave functions @xmath129 or @xmath149 , according to the representation we use .
some ideas of the approach contained in this paper are similar to those of geometric quantization , see for example @xcite . in both cases
the starting point is the kvn hilbert space of square integrable functions on the phase space , in both cases a map is introduced to build the quantum observables and in both cases a sort of polarization is used to reduce the dimension of the quantum hilbert space to the usual one .
however in our approach the recipe to construct the observables is very simple and far from the complicated tools used in geometric quantization . the simple map ( [ mapq ] ) itself suggests in fact the two other ingredients needed to quantize a classical system .
the first one is a well - motivated physical prescription , the conservation of the quantum energy , which implies that the dynamics of the theory has to be modified .
the second ingredient is the mathematical prescription which says that the operators @xmath64 and @xmath65 must act irreducibly and which is necessary to eliminate the redundancy in the physical description of the system . anyway , like in every other approach to quantization also in our case we can not bypass the grenewald - van hove theorem @xcite and an obstruction to a full quantization @xcite of the system still remains . in geometric quantization people construct a very complicated @xmath153-map just to satisfy the relation @xmath154\ ] ] for every function @xmath45 and @xmath155 but , when they impose the request of irreducibility , the number of observables which can be quantized gets limited . in our approach
the irreducibility prescription reduces the kvn hilbert space to the usual quantum one in a more natural way than in geometric quantization and again this is motivated by the a priori choice of the observables ( [ mapq ] ) that we made .
nevertheless our map ( [ mapq ] ) can not satisfy eq .
( [ kit ] ) for every function @xmath45 and @xmath155 as a simple calculation can show .
therefore the grenewald - van hove theorem is not bypassed by our quantization method .
the last point we want to mention is that having a unique big hilbert space for both classical and quantum mechanics can help in studying and appreciating the similarities and differences between the two theories . in particular , for what concerns the observables we must note , see * fig . 1 *
, that within the kvn theory there is enough space to describe both classical and quantum mechanics but there is further space available for further observables which could describe some other possible regimes , like the ones at the border between classical and quantum mechanics @xcite or the models of dynamical reduction @xcite .
we think that it would be interesting to investigate these issues in the future .
i would like to thank e. gozzi for his precious help and suggestions .
this research has been supported by grants from infn , miur and the university of trieste .
99 b. o. koopman , proc .
17 ( 1931 ) 315 .
j. von neumann , ann .
33 ( 1932 ) 587 ; ibid .
33 ( 1932 ) 789 .
e. gozzi , m. reuter , w. d. thacker , phys .
d 40 ( 1989 ) 3363 .
d. mauro , int .
phys . a 17 ( 2002 ) 1301 .
j. e. moyal , proc .
( 1949 ) 99 .
+ for a review of the wigner function formalism , see : + w. p. schleich , quantum optics in phase space , wiley - vch , berlin , 2001 .
e. gozzi , d. mauro , to appear .
m. s. marinov , phys .
lett . a 153 ( 1991 ) 5 .
e. gozzi , m. reuter , int .
phys . a 9 ( 1994 ) 2191 .
r. kubo , jour .
jap . 19 ( 1964 ) 2127 .
e. gozzi , d. mauro , ann .
phys . 296 ( 2002 ) 152 .
r. abraham , j. e. marsden , foundations of mechanics , benjamin , new york , 1978 .
n. m. j. woodhouse , geometric quantization , oxford university press , oxford , 1980 .
l. fonda , g. c. ghirardi , symmetry principles in quantum physics , dekker , new york , 1970 .
h. j. grenewald , physica 12 ( 1946 ) 405 + l. van hove , mem . de lacad .
. de belgique ( classe des sci . ) 26 ( 1951 ) 61 .
l. djosi , n. gisin , w. t. strunz , phys rev . a 61 ( 2000 ) 022108 + o. v. prezhdo , v. v. kisil , phys .
rev a 56 ( 1997 ) 162 + a. peres , d. terno , phys rev .
a ( 2001 ) 022101 + j. caro , l. l. salcedo , phys .
a ( 1999 ) 842 .
g. c. ghirardi , a. rimini , t. weber , phys .
d 34 ( 1986 ) , 470 . | in this paper we find a simple rule to reproduce the algebra of quantum observables using only the commutators and operators which appear in the koopman - von neumann ( kvn ) formulation of classical mechanics .
the usual hilbert space of quantum mechanics becomes embedded in the kvn hilbert space : in particular it turns out to be the subspace on which the quantum positions q and momenta p act irreducibly .
d. mauro dipartimento di fisica teorica , universit
di trieste , + strada costiera 11 , p.o.box 586 , trieste , italy , + and infn , sezione di trieste , italy + e - mail : [email protected]_ |
in many multiuser wireless communications scenarios , having sufficient csit is a crucial ingredient that facilitates improved performance . while being useful , perfect csit is also hard and time - consuming to obtain , hence the need for communication schemes that can utilize partial or delayed csit knowledge ( see @xcite ) . in this context of multiuser communications ,
we here consider the broadcast channel ( bc ) , and specifically focus on the two - user multiple - input single - output ( miso ) bc , where a two - antenna transmitter communicates to two single - antenna receivers . in this
setting , the channel model takes the form @xmath0 where for any time instant @xmath1 , @xmath2 represent the channel vectors for user 1 and 2 respectively , where @xmath3 represent unit power awgn noise , where @xmath4 is the input signal with power constraint @xmath5 , and where in this case , @xmath6 also takes the role of the signal - to - noise ratio ( snr ) .
it is well known that in this setting , the presence of full csit allows for the optimal @xmath7 degree - of - freedom ( dof ) per user , whereas the complete absence of csit causes a substantial degradation to just @xmath8 dof per user , the corresponding dof pair @xmath9 is given by @xmath10 the corresponding dof region is then the set of all achievable dof pairs . ] .
an interesting scheme that bridges this performance gap by utilizing partial csit knowledge , was recently presented in @xcite which showed that delayed csit knowledge can still be useful in improving the dof region of the broadcast channel . in the above described two - user miso bc setting , and under the assumption that at time @xmath1 , the transmitter knows the delayed channel states ( @xmath11 ) up to time @xmath12 , the work in @xcite showed that each user can achieve @xmath13 dof , providing a clear improvement over the case of no csit .
this result was later generalized in @xcite which considered the natural extension where , in addition to the aforementioned perfect knowledge of prior csit , the transmitter also had imperfect knowledge of current csit ; at time @xmath1 the transmitter had estimates @xmath14 of @xmath15 and @xmath16 , with estimation errors @xmath17 having i.i.d .
gaussian entries with power @xmath18 for some non - negative parameter @xmath19 that described the quality of the estimate of the current csit . in this setting of
` mixed ' csit ( perfect prior csit and imperfect current csit ) , and for @xmath20 denoting the dof for the first and second user over the aforementioned two - user bc , the work in @xcite showed the optimal dof region to take the form , @xmath21 corresponding to a polygon with corner points @xmath22 , nicely bridging the gap between the case of @xmath23 explored in @xcite , and the case of @xmath24 ( and naturally @xmath25 ) corresponding to perfect csit . throughout this paper , @xmath26 , @xmath27 , @xmath28 ,
respectively denote the inverse , transpose , and conjugate transpose of a matrix , while @xmath29 denotes the complex conjugate , and @xmath30 denotes the euclidean norm .
@xmath31 denotes the magnitude of a scalar , and @xmath32 denotes a diagonal matrix .
logarithms are of base 2 .
@xmath33 comes from the standard landau notation , where @xmath34 implies @xmath35 .
we also use @xmath36 to denote _ exponential equality _ , i.e. , we write @xmath37 to denote @xmath38 .
finally , in the spirit of @xcite we consider a unit coherence period , as well as perfect knowledge of channel state information at the receivers ( perfect csir ) .
motivated by the fact that in multiuser settings , the quality of csit feedback may vary across different links , we extend the approach in @xcite to consider unequal quality of current csit knowledge for @xmath15 and @xmath16 . specifically under the same set of assumptions mentioned above , and in the presence of perfect prior csit , we now consider the case where at time @xmath1 , the transmitter has estimates @xmath14 of the current @xmath15 and @xmath16 , with estimation errors @xmath39 having i.i.d .
gaussian entries with power @xmath40 for some non - negative parameters @xmath41 that describe the generally unequal quality of the estimates of the current csit for the two users links .
we proceed to describe the optimal dof region of the general mixed - csit two - user miso bc ( two - antenna transmitter ) .
the optimal schemes are presented in section [ sec : achievability ] , parts of the proof of the schemes performance are presented in appendix [ sec : achievable ] , while the outer bound proof is placed in appendix [ sec : outerb ] . without loss of generality
, the rest of this work assumes that @xmath42 [ theorem : bc - outerb - gen ] the dof region of the two - user miso bc with general mixed - csit , is given by @xmath43 where the region is a polygon which , for @xmath44 has corner points @xmath45 and otherwise has corner points @xmath46 the above corner points , and consequently the entire dof inner bound , will be attained by the schemes to be described later on .
the result generalizes the results in @xcite as well as the result in @xcite which considered the case of ( @xmath47 ) , where one user had perfect csit and the other only prior csit .
( case 1 ) and when @xmath48 ( case 2 ) . the corner points take the following values : @xmath49 , @xmath50 , @xmath51 and @xmath52 .
, width=340 ] figure [ fig : dofcsitouterb ] depicts the general dof region for the case where @xmath44 ( case 1 ) and the case where @xmath48 ( case 2 ) .
we proceed to describe the communication schemes .
as stated , without loss of generality , we assume that @xmath53 .
we describe the three schemes @xmath54 , @xmath55 and @xmath56 that achieve the optimal dof region ( in conjunction with time - division between these same schemes ) .
specifically scheme @xmath54 achieves @xmath51 ( case 1 ) , scheme @xmath55 achieves dof points @xmath52 ( case 1 ) and @xmath49 ( case 2 ) , and scheme @xmath56 achieves @xmath50 ( case 1 and case 2 ) .
the scheme description is done for @xmath57 , and for rational @xmath41 .
the cases where @xmath58 , or @xmath59 , or where @xmath41 are not rational , can be readily handled with minor modifications .
we proceed to describe the basic notation and conventions used in our schemes .
the schemes are designed with @xmath60 phases ( @xmath60 varies from scheme to scheme ) , where the @xmath61th phase consists of @xmath62 channel uses , @xmath63 . the vectors @xmath64 and @xmath65 will denote the channel vectors seen by the first and second user respectively during timeslot @xmath1 of phase @xmath61 , while @xmath66 and @xmath67 will denote the estimates of these channels at the transmitter during the same time , and @xmath68 , @xmath69 will denote the estimation errors . furthermore @xmath70 and @xmath71 will denote the independent information symbols that may be sent during phase-@xmath61 , timeslot-@xmath1 , and which are meant for user 1 , while symbols @xmath72 and @xmath73 are meant for user 2 .
vectors @xmath74 and @xmath75 are the unit - norm beamformers for @xmath70 and @xmath72 respectively , chosen so that @xmath74 is orthogonal to @xmath67 , and so that @xmath75 is orthogonal to @xmath66 .
furthermore @xmath76 are the randomly chosen unit - norm beamformers for @xmath71 and @xmath73 respectively .
another notation that will be shared between schemes includes @xmath77 that denotes the interference seen by user 1 and user 2 respectively , during timeslot @xmath1 of phase @xmath61 . for @xmath78
being the accumulated interference to both users during phase @xmath61 , we will let @xmath79 be a quantized version of @xmath78 , and we will consider the mapping where the total information in @xmath79 is split evenly across symbols @xmath80 transmitted during the next phase .
in addition we use @xmath81 to denote the randomly chosen unit - norm beamformer of @xmath82 .
furthermore , unless stated otherwise , @xmath83 will be the general form of the transmitted vector at timeslot @xmath1 of phase @xmath61 .
as noted above under each summand , the average power that is assigned to each symbol , throughout a specific phase , will be denoted as follows : @xmath84 furthermore each of the above symbols carries a certain amount of information , per timeslot , where this amount may vary across different phases .
specifically we use @xmath85 to mean that , during phase @xmath61 , each symbol @xmath86 carries @xmath87 bits .
similarly we use @xmath88 to describe the prelog factor of the number of bits in @xmath89 respectively , again for phase @xmath61 . finally the received signals during phase @xmath61 for the first and second user ,
are respectively denoted as @xmath90 and @xmath91 , where generally the signals take the following form @xmath92 as stated , scheme @xmath54 has @xmath60 phases , where the phase durations @xmath93 are chosen to be integers such that @xmath94 where @xmath95 , @xmath96 , @xmath97 , and where @xmath98 is any constant such that @xmath99 . during phase 1 ( @xmath100 channel uses ) ,
the transmit signal is @xmath101 while the power and rate are set as @xmath102 the received signals at the two users then take the form @xmath103 where under each term we noted the order of the summand s average power . at this point , and after the end of the first phase , the transmitter can use its knowledge of delayed csit to reconstruct @xmath104 ( cf . ) , and quantize each term as @xmath105 where @xmath106 are the quantized values , and where @xmath107 are the quantization errors . noting that @xmath108
, we choose a quantization rate that assigns each @xmath109 a total of @xmath110 bits , and each @xmath111 a total of @xmath112 bits , thus allowing for @xmath113 ( @xcite ) . at this point the @xmath114 bits representing @xmath115 , are distributed evenly across the set @xmath116 which will be sequentially transmitted during the next phase .
this transmission of @xmath116 will help each of the users cancel the interference from the other user , and it will also serve as an extra observation that allows for decoding of all private information of that same user . during phase 2 ( @xmath117 channel uses ) , the transmit signal takes the exact form in @xmath118 where we set power and rate as @xmath119 and where we note that @xmath120 satisfies @xmath121 .
the received signals during this phase are given as @xmath122 for @xmath123 , where under each term we noted the order of the summand s average power .
at this point , based on , , each user decodes @xmath124 by treating the other signals as noise . after decoding @xmath116 and fully reconstructing @xmath125
, user 1 goes back one phase and subtracts @xmath111 from @xmath126 to remove ( up to bounded noise ) the interference corresponding to @xmath127 .
the same user will also use the estimate @xmath109 of @xmath128 as an extra observation which , together with the observation @xmath126 , present the user with a @xmath129 mimo channel that allows for decoding of both @xmath130 and @xmath131 similarly user 2 , after fully reconstructing @xmath125 , subtracts @xmath109 from @xmath132 , to remove ( up to bounded noise ) the interference corresponding to @xmath128 , and also uses the estimate @xmath111 of @xmath127 as an extra observation which , together with the observation @xmath132 , allow for decoding of both @xmath133 and @xmath134 further exposition to the details regarding the achievability of the mentioned rates , can be found in appendix [ sec : achievable ] .
consequently after the end of the second phase , the transmitter can use its knowledge of delayed csit to reconstruct @xmath135 , and quantize each term to @xmath136 . with @xmath137 ,
we choose a quantization rate that assigns each @xmath138 a total of @xmath139 bits , and each @xmath140 a total of @xmath141 bits , thus allowing for @xmath142 .
then the @xmath143 bits representing @xmath144 , are split evenly across the set @xmath145 which will be sequentially transmitted in the next phase so that user 1 can eventually decode @xmath146 , and user 2 can decode @xmath147 .
we now proceed with the general description of phase @xmath61 .
phase @xmath61 ( @xmath149 channel uses ) is almost identical to phase 2 , with one difference being the different relationship between @xmath62 and @xmath150 .
the transmit signal takes the same form as in phase 2 ( cf .
, ) , the rates and powers of the symbols are the same ( cf . ) and the received signals @xmath151
( @xmath152 ) take the same form as in , .
most of the actions are also the same , where based on , ( corresponding now to phase @xmath61 ) , each user decodes @xmath153 by treating the other signals as noise , and then goes back one phase and reconstructs @xmath154 .
as before , user 1 then subtracts @xmath155 from @xmath156 to remove , up to bounded noise , the interference corresponding to @xmath157 .
the same user also employs the estimate @xmath158 of @xmath159 as an extra observation which , together with the observation @xmath160 obtained after decoding @xmath161 , allow for decoding of both @xmath162 and @xmath163 .
similar actions are performed by user 2 . as before ,
after the end of phase @xmath61 , the transmitter can use its knowledge of delayed csit to reconstruct @xmath164 , and quantize each term to @xmath165 with the same rate as in phase 2 ( @xmath139 bits for each @xmath166 , and @xmath141 bits for each @xmath167 ) .
finally the accumulated @xmath168 bits representing all the quantized values @xmath169 , are distributed evenly across the set @xmath80 which will be sequentially transmitted in the next phase .
more details can be found in appendix [ sec : achievable ] . during the last phase ( @xmath170 channel uses ) , the transmit signal is @xmath171 where we set power and rate as @xmath172 the received signals are @xmath173 for @xmath174 . at this point , as before , the power and rate allocation of the different symbols allow both users to decode @xmath175 by treating the other signals as noise .
consequently user 1 can remove @xmath176 from @xmath177 and decode @xmath178 , and similarly user 2 can remove @xmath179 from @xmath180 and decode @xmath181 .
finally each user goes back one phase and reconstructs @xmath182 , which allows for decoding of @xmath183 and @xmath184 at user 1 and of @xmath185 and @xmath186 at user 2 , all as described for the previous phases ( see appendix [ sec : achievable ] for more details ) .
table [ tab : x1summary ] summarizes the parameters of scheme @xmath54 .
the use of symbol @xmath187 is meant to indicate precoding that is orthogonal to the channel estimate ( rather than random ) .
the table s last row indicates the prelog factor of the quantization rate .
.summary of scheme @xmath54 .
[ cols="^,^,^,^,^",options="header " , ] [ tab : x2summary ] [ [ dof - calculation - for - scheme - mathcal - x_2 ] ] dof calculation for scheme @xmath55 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we proceed to add up the total amount of information transmitted during this scheme . in accordance to the declared pre - log factors
@xmath188 and phase durations ( see table [ tab : x2summary ] ) , and irrespective of whether @xmath41 fall under case 1 or case 2 , we have that @xmath189 where is due to . regarding the second user and the declared @xmath190 , for case 1 ( @xmath191 ) we see that @xmath192 where we have used to get , where we have used that @xmath191 implies @xmath193 , and where we have considered an asymptotically large @xmath60 .
when @xmath194 ( @xmath195 ) , then gives that @xmath196 which , in the high @xmath60 regime , gives @xmath197 when @xmath198 ( @xmath199 ) , then gives that @xmath200 which , for large @xmath60 , gives @xmath201 in conclusion , scheme @xmath55 achieves dof pair @xmath52 ( case 1 ) , else it achieves @xmath49 .
this is the simplest of all three schemes , and it consists of a single channel use ( @xmath202 ) during which the transmitter sends @xmath203 where @xmath204 is orthogonal to @xmath205 , @xmath206 is orthogonal to @xmath207 , and where the power and rates are set as @xmath208 resulting in received signals of the form @xmath209 after transmission , both receivers first decode @xmath210 by treating the other signals as noise , and then user 1 utilizes its knowledge of @xmath211 to reconstruct @xmath212 and remove it from @xmath213 , thus being able to decode @xmath214 , while after decoding @xmath210 , user 2 removes @xmath215 from @xmath216 , and decodes @xmath217 .
the details for the achievability of @xmath218 follow closely the exposition in appendix [ sec : achievable ] .
consequently the dof point @xmath219 can be achieved by associating @xmath210 to information intended entirely for the second user .
the work provided analysis and communication schemes for the setting of the two - user miso bc with general mixed csit . the work can be seen as a natural extension of the result in @xcite and of the recent results in @xcite , to the case where the csit feedback quality varies across different links .
we will here focus on achievability details for scheme @xmath54 . the clarifications of the details carry over easily to the other two schemes .
regarding @xmath220 ( @xmath221 - see ) , we recall that during phase @xmath61 , both users decode @xmath153 ( from @xmath222 - see , ) by treating all other signals as noise . consequently for @xmath223
, we note that @xmath224 to get @xmath225 similarly for the last phase @xmath60 ( see , , ) , we note that @xmath226 to get @xmath227 regarding achievability for @xmath228 , @xmath229 , @xmath230 and @xmath231 ( see , , ) , we note that each element in @xmath116 has enough bits ( recall that @xmath232 ) , to match the quantization rate of @xmath233 that is necessary in order to have a bounded quantization noise . consequently going back to phase 1 ,
user 1 is presented with @xmath100 linearly independent @xmath129 equivalent mimo channels of the form @xmath234 ( @xmath235 ) , where again we note that the described quantization rate results in a bounded equivalent noise , which then immediately gives that @xmath228 and @xmath229 are achievable .
similarly for user 2 , the presented @xmath100 linearly independent @xmath129 equivalent mimo channels @xmath236 ( @xmath235 ) , allow for decoding at a rate corresponding to @xmath230 and @xmath231 .
regarding achievability for @xmath237 , @xmath238 , @xmath239 and @xmath240 , ( @xmath241 - see , , ) , we note that during phase @xmath61 , both users can decode @xmath153 , and as a result user 1 can remove @xmath242 from @xmath90 , and user 2 can remove @xmath243 from @xmath91 ( @xmath244 ) . as a result user 1
is presented with @xmath62 linearly independent @xmath129 equivalent mimo channels of the form @xmath245 ( @xmath244 ) .
given that the rate associated to @xmath80 , matches the quantization rate for @xmath246 , allows for a bounded variance of the equivalent noise , and in turn for decoding of @xmath247 at a rate corresponding to @xmath237 and @xmath238 .
similarly user 2 is presented with @xmath62 independent @xmath129 mimo channels of the form @xmath248 allowing for decoding of @xmath249 ( @xmath244 ) at rates corresponding to @xmath239 and @xmath240 .
regarding achievability for @xmath250 and @xmath251 ( see , , ) , we note that , after decoding @xmath175 , user 1 can remove @xmath176 from @xmath252 , and user 2 can remove @xmath179 from @xmath180 , ( @xmath253 ) . consequently during this phase , user 1 sees @xmath254 linearly independent siso channels of the form @xmath255 ( @xmath253 ) which can be readily shown to support @xmath250 .
a similar argument gives achievability for @xmath251 .
@xmath256 we here adopt the outer bound approach in @xcite to the asymmetric case of @xmath257 .
as in @xcite , we first linearly convert the original bc in , to an equivalent bc ( see , ) having the same dof region as the original bc ( cf.@xcite ) , and we then consider the degraded version of the equivalent bc in the absence of delayed feedback , which matches in capacity the degraded bc with feedback ( for the memoryless case ) , and which exceeds the capacity of the equivalent bc .
the final step considers the compound and degraded version of the equivalent bc without delayed feedback , whose dof region will serve as an outer bound on the dof region of the original bc . where @xmath259^{t } { \triangleq}\frac{1}{\sqrt{p } } { { { \uppercase{{\bm{q}}}}}}^{-1}_{t } { { { \lowercase{{\bm{x}}}}}}_{t},\ ] ] where @xmath260 \in \mathbb{c}^{2\times 2}$ ] is , with probability 1 , an invertible matrix , where @xmath261 is chosen to be of unit norm and orthogonal to @xmath262 , and where @xmath263 is chosen to be of unit norm and orthogonal to @xmath264 .
furthermore each receiver normalizes to get where @xmath266 , @xmath267 , @xmath268 , @xmath269 .
consequently @xmath270 and @xmath271 have identity covariance matrices , and the average power of @xmath272 , @xmath273 , @xmath274 and @xmath275 does not scale with @xmath6 , i.e. , in the high - snr region this power is of order @xmath276 . with the same csit knowledge mapped from the original bc ,
it can be shown ( see @xcite ) that the dof region of the equivalent bc in matches the dof region of the original bc in . towards designing the degraded version of the above equivalent bc , we supply the second user with knowledge of @xmath277 , and towards designing the compound version of the above degraded equivalent bc , we add two extra users ( user 3 and 4 ) . in this compound version , the received signals for the first two users are as in , while the received signals of the added ( virtual ) users are given by we here note that by definition , @xmath279 and @xmath280 are statistically equivalent to the original @xmath272 and @xmath273 respectively , and that @xmath281 and @xmath282 are statistically equivalent to the original @xmath274 and @xmath275 .
furthermore we note that user 3 is interested in the same message as user 1 , while user 4 is interested in the same message as user 2 .
also we recall that in the specific degraded compound bc , user 1 knows @xmath277 , user 2 knows @xmath283 and @xmath277 , user 3 knows @xmath284 , and user 4 knows @xmath285 and @xmath284 .
finally we remove delayed feedback - a removal known to not affect the capacity of the degraded bc without memory @xcite .
we now proceed to calculate an outer bound on the dof region of this degraded compound bc which at least matches the dof of the previous degraded bc and which serves as an outer bound on the dof region of the original bc .
we consider communication over the described equivalent degraded compound bc , letting @xmath286 be the large number of fading realizations over which communication takes place , and letting @xmath287 be the rates of the first and second user .
we also let @xmath288}{\triangleq}\{{{{\lowercase{{\bm{h}}}}}}_{t } , { { { \lowercase{{\bm{g}}}}}}_{t},\hat{{{{\lowercase{{\bm{h}}}}}}}_{t } , \hat{{{{\lowercase{{\bm{g}}}}}}}_{t}\}^{n}_{t=1}$ ] , @xmath289}{\triangleq}\{y^{'(i)}_{t}\}_{t=1}^{n}$ ] and @xmath290 } { \triangleq}\{y^{''(i)}_{t}\}_{t=1}^{n}$ ] for @xmath291 . using fano s inequality
, we have @xmath292}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) + n o(n ) \nonumber \\&\leq\ ! n \log\ ! p \!+\ !
n o(\log\ !
h(y^{'(1)}_{[n]}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \!+\ !
n o(n ) , \label{eq : r1-bound-2}\end{aligned}\ ] ] as well as @xmath293}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) + n o(n ) \nonumber \\&\leq\ !
n o(\log\ !
! h(y^{''(1)}_{[n]}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \!+\ !
n o(n ) , \label{eq : r1-bound-12}\end{aligned}\ ] ] which is added to to give @xmath294}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \nonumber \\ & \quad -h(y^{''(1)}_{[n]}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n]})+
2n o(n ) \nonumber \\ & \leq 2n\log p+ 2n o(\log p ) \nonumber \\ & \quad -h(y^{'(1)}_{[n]},y^{''(1)}_{[n]}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) + 2n o(n ) .
\label{eq:2r1-bound}\end{aligned}\ ] ] let @xmath295 where @xmath296 , @xmath297 , and let @xmath298}{\triangleq}\{z_{t}\}_{t=1}^{n}$ ] . consequently @xmath299},y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \nonumber \\ & \quad + h(w_1,w_2|y^{'(1)}_{[n]},y^{''(1)}_{[n]},z_{[n]},{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \nonumber \\ & = i(w_1,w_2;y^{'(1)}_{[n]},y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \nonumber \\ & \quad + n o(\log p)+
n o(n ) \label{eq : r1r2-ob-1 } \\ & = i(w_1;y^{'(1)}_{[n]},y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) \nonumber \\ & \quad + i(w_2;y^{'(1)}_{[n]},y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n ] } , w_1 ) \nonumber \\ & \quad + n o(\log p)+
n o(n ) , \label{eq : r1-r2 } \end{aligned}\ ] ] where the transition to uses the fact that the high snr variance of @xmath300 and @xmath301 scales as @xmath276 and @xmath302 respectively , which in turn means that knowledge of @xmath303}\}_{t=1}^{n}$ ] , implies knowledge of @xmath304 and of @xmath305 , up to bounded noise level .
furthermore @xmath306}\!,y^{''(1)}_{[n]}\!,\ !
! z_{[n]},{{{\uppercase{{\bm{h}}}}}}_{[n]}\ ! )
\nonumber \\ & = \!\!i(\
! w_1;y^{'(1)}_{[n]}\!,y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]}\ !
n o(n ) , \label{eq : r1}\end{aligned}\ ] ] since again knowledge of @xmath303}\}_{t=1}^{n}$ ] provides for @xmath307 up to bounded noise level . now combining and , gives @xmath308},y^{''(1)}_{[n]},z_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1 ) \!+\ !
n o(\log p)\!+\ !
n o(n ) \\ & = i(w_2;y^{'(1)}_{[n]},y^{''(1)}_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1 ) \\
& \quad \!+\ !
i(w_2;z_{[n]}|y^{'(1)}_{[n]},y^{''(1)}_{[n]},{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1 ) \!+\ ! n o(\log p)\!+\ !
n o(n ) \nonumber\\ & = \
! h(y^{'(1)}_{[n]},y^{''(1)}_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1)\!-\!\underbrace{h(y^{'(1)}_{[n]},y^{''(1)}_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1,w_2)}_{n o(\log p ) } \\ & \quad -\underbrace{h(z_{[n]}|y^{'(1)}_{[n]},y^{''(1)}_{[n]},{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1,w_2)}_{n o(\log p ) } \\ & \quad + \underbrace{h(z_{[n]}|y^{'(1)}_{[n]},y^{''(1)}_{[n]},{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1)}_{\le h(z_{[n ] } ) } + n o(\log p)+
n o(n ) \\ & \leq \ !
h(y^{'(1)}_{[n]},y^{''(1)}_{[n]}|{{{\uppercase{{\bm{h}}}}}}_{[n]},w_1 ) \!+\!h(z_{[n ] } ) \!+\ !
n o(\log p)\!+\ !
n o(n ) \\ & \leq \
! h(y^{'(1)}_{[n]},y^{''(1)}_{[n]}|w_1,{{{\uppercase{{\bm{h}}}}}}_{[n ] } ) + n \alpha_1\log p \\ & \quad + n o(\log p)+
n o(n ) , \label{eq : r2-bound - final}\end{aligned}\ ] ] which is combined with to give @xmath309 which in turn proves the outer bound @xmath310 as described in .
finally interchanging the roles of the two users and of @xmath41 , gives @xmath311 naturally the single antenna constraint gives that @xmath312 .
@xmath256 c. s. vaze and m. k. varanasi , `` the degrees of freedom region of two - user and certain three - user mimo broadcast channel with delayed csi , '' dec .
2010 , submitted to _
ieee trans .
inform . theory _
, available on arxiv:1101.0306v2 [ cs.it ] .
j. xu , j. g. andrews , and s. a. jafar , `` broadcast channels with delayed finite - rate feedback : predict or observe ? '' may 2011 , submitted to _
ieee trans . on wireless communications _
, available on arxiv:1105.3686v1 [ cs.it ] .
m. a. maddah - ali and d. n. c. tse , `` completely stale transmitter channel state information is still very useful , '' sep .
2011 , submitted to _
ieee trans .
inform . theory _
, available on arxiv:1010.1499v2 [ cs.it ] .
m. kobayashi , s. yang , d. gesbert , and x. yi , `` on the degrees of freedom of time correlated miso broadcast channel with delayed csit , '' feb .
2012 , submitted to _ proc .
information theory ( isit ) 2012 _ , available on arxiv:1202.1909v1 [ cs.it ] .
s. yang , m. kobayashi , d. gesbert , and x. yi , `` degrees of freedom of time correlated miso broadcast channel with delayed csit , '' mar .
2012 , submitted to _
ieee trans .
inform . theory _ , available on arxiv:1203.2550v1 [ cs.it ] .
t. gou and s. jafar , `` optimal use of current and outdated channel state information - degrees of freedom of the miso bc with mixed csit , '' mar .
2012 , submitted to _ ieee communications letters _ , available on arxiv:1203.1301v1 [ cs.it ] . | in the setting of the two - user broadcast channel , recent work by maddah - ali and tse has shown that knowledge of prior channel state information at the transmitter ( csit ) can be useful , even in the absence of any knowledge of current csit .
very recent work by kobayashi et al .
, yang et al . , and gou and jafar , extended this to the case where , instead of no current csit knowledge , the transmitter has partial knowledge , and where under a symmetry assumption , the quality of this knowledge is identical for the different users channels . motivated by the fact that in multiuser settings , the quality of csit feedback may vary across different links ,
we here generalize the above results to the natural setting where the current csit quality varies for different users channels . for this setting
we derive the optimal degrees - of - freedom ( dof ) region , and provide novel multi - phase broadcast schemes that achieve this optimal region .
finally this generalization incorporates and generalizes the corresponding result in maleki et al . which considered the broadcast channel with one user having perfect csit and the other only having prior csit . |
oriented matroids are a combinatorial model , introduced independently by bland , folkman , las vergnas , and lawrence ( published in @xcite and @xcite )
. oriented matroids abstract various objects such as point configurations , vector configurations , polytopes , hyperplane arrangements , and digraphs , and provide a unified framework for discussing various combinatorial aspects of these objects .
they have a rich structure , which can describe , for example , face structures of polytopes and hyperplane arrangements , partitions of point sets by hyperplanes , linear programming duality , and gale duality in polytope theory . nowadays ,
oriented matroid theory is a fairly rich subject of research , with connections to various research fields such as discrete and computational geometry , graph theory , operations research , topology , and algebraic geometry ( see @xcite ) .
one of the outstanding results in oriented matroid theory is the topological representation theorem , introduced by folkman and lawrence @xcite ( see @xcite for simplified proofs ) , which says that there is a one - to - one correspondence between oriented matroids and equivalence classes of _ pseudosphere arrangements_. this theorem bridges topology and combinatorics , and shows that oriented matroids constitute a natural combinatorial model . motivated by oriented matroid theory , it is natural to consider whether a useful theory can be developed by abstracting combinatorial aspects of other related objects .
because one of the common ways of understanding oriented matroids is the combinatorial abstraction of partitions of point sets in the euclidean space by families of hyperplanes , we consider combinatorial abstraction of partitions of point sets by other geometric objects .
recently , eli and matouek @xcite proposed a new interesting generalization of the erds - szekeres theorem . to prove the theorem , they investigated the combinatorial properties of partitions of point sets by graphs of certain families of polynomial functions of degree @xmath0 .
in this paper , we discuss more closely what kind of combinatorial properties are exhibited by such partitions .
in fact , we show that the combinatorial properties observed by eli and matouek represent in some sense almost _ all _ of the combinatorial properties that can be proved using some natural geometric properties . to do so , we first observe a slightly stronger combinatorial property of partitions than those observed by eli and matouek , and then we introduce a combinatorial model called _ degree-@xmath0 oriented matroids _ by axiomatizing the observed combinatorial properties .
we prove that degree-@xmath0 oriented matroids completely characterize partitions arising from _ @xmath0-intersecting pseudoconfigurations of points _ , which are a natural geometric generalization of configurations of points and graphs of polynomial functions of degree @xmath0 .
this provides an analogue of the topological representation theorem for oriented matroids , and shows that degree-@xmath0 oriented matroids capture the essential combinatorial properties of partitions of point sets by graphs of polynomial functions of degree @xmath0 .
the partitions of a point set in the euclidean space by a certain family of spheres determine an oriented matroid , which is called a _
delaunay oriented matroid _ ( see ( * ? ? ?
* section 1.9 ) ) .
santos @xcite proved that partitions of a point set in the 2-dimensional euclidean space by spheres defined by a smooth , strictly convex distance function also fulfill the oriented matroid axioms .
miyata @xcite proved that the class of partitions of point sets in the 2-dimensional euclidean space by pseudocircles coincides with the rank @xmath2 matroid polytopes .
ardila and develin @xcite introduced tropical oriented matroids as a combinatorial abstraction of tropical hyperplane arrangements .
horn @xcite proved that every tropical oriented matroid can be represented as a tropical pseudohyperplane arrangement . here , we summarize the notation that will be employed in this paper . in the following , we assume that @xmath3 is a finite ordered set , @xmath4 is a sign vector on @xmath5 , @xmath6 is a point in the @xmath1-dimensional euclidean space , and
@xmath7 is a positive integer . 2 * @xmath8 . * @xmath9 for @xmath10 . * @xmath11 * for @xmath12 .
* @xmath13 : = ( \lambda_1,\dots,\lambda_{i-1},k,\lambda_{i+1}\dots,\lambda_r)$ ] * for @xmath10 .
* @xmath14 . * @xmath15 ( resp .
@xmath16 ) * @xmath17 ( resp .
* @xmath20 : the @xmath21-coordinate of @xmath6 . * @xmath22 : the @xmath23-coordinate of @xmath6 .
in this section , we summarize some basic facts regarding oriented matroids .
see @xcite for further details .
let @xmath24}$ ] be a point configuration in general position ( i.e. , no @xmath25 points of @xmath6 lie on the same hyperplane ) in @xmath26 , and let @xmath27}$ ] be the vector configuration in @xmath28 with @xmath29 . to see how @xmath6 is separated by hyperplanes ,
let us consider the map @xmath30^{d+1 } \rightarrow \ { + 1,-1,0 \}$ ] defined by @xmath31$,}\ ] ] where @xmath32 ( resp .
@xmath18 ) if @xmath33 ( resp .
@xmath34 ) .
then , we have @xmath35 the map @xmath36 contains rich combinatorial information regarding @xmath6 , such as convexity , the face lattice of the convex hull , and possible combinatorial types of triangulations ( see @xcite ) .
the chirotope axioms of oriented matroids are obtained by abstracting the properties of @xmath36 .
( chirotope axioms for oriented matroids ) + for @xmath37 and a finite set @xmath5 , a map @xmath38 is called a _ chirotope _ if it satisfies the following axioms .
the pair @xmath39 is called an _ oriented matroid of rank @xmath7_. * @xmath40 is not identically zero .
* @xmath41 , for any @xmath42 and any permutation @xmath43 on @xmath5 . * for any @xmath44 , we have @xmath45 ) \chi ( \mu [ \mu_s | \lambda_1 ] ) \mid s = 1,\dots , r \ } \supset \ { + 1,-1\ } \text { or } = \ { 0 \ } .\ ] ] we remark that ( b3 ) is combinatorial abstraction of the grassmann - plcker relations : @xmath46})\det(v_{\mu[\mu_s|\lambda_1]})},\ ] ] where @xmath47 for @xmath48,d+1)$ ] . the set @xmath49 } \mid
\lambda \in [ n]^d \}$ ] also contains equivalent information to @xmath36 .
the cocircuit axioms of oriented matroids are introduced by abstracting the properties of the set @xmath50 .
( cocircuit axioms for oriented matroids ) + a collection @xmath51 satisfying axioms ( c0)(c3 ) is called the set of _ cocircuits _ of an oriented matroid . * @xmath52 .
* @xmath53 .
* for all @xmath54 , if @xmath55 , then @xmath56 or @xmath57 .
* for any @xmath54 and @xmath58 , there exists @xmath59 with * @xmath60 , @xmath61 , and @xmath62 . from a chirotope @xmath40
, we can construct the cocircuits @xmath63 .
it is also possible to reconstruct the chirotope @xmath40 ( up to a sign reversal ) from the cocircuits @xmath64 .
a rank @xmath7 oriented matroid @xmath65 is said to be _ uniform _ if @xmath66 for any @xmath67 ( equivalently if @xmath68 for any cocircuit @xmath4 ) . if the underlying strucure of the set @xmath69 is known to be uniform , i.e. , if @xmath70 for any @xmath71 , then axiom ( c3 ) can be replaced by the following axiom : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ( c3 ) for any @xmath54 with @xmath72 and @xmath58 , there exists @xmath59 with @xmath60 , @xmath61 , and @xmath62 .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ more generally , axiom ( c3 ) can be replaced by the axiom of _ modular elimination_. for further details , see ( * ? ? ?
* section 3.6 ) .
an oriented matroid @xmath73 is _ acyclic _ if for any @xmath74 there exists @xmath75 with @xmath76 and @xmath77 .
it can easily be seen that oriented matroids arising from point configurations are acyclic .
+ + one of the outstanding facts in oriented matriod theory is that oriented matroids always admit topological representations , as established by folkman and lawrence @xcite . here
, we explain a variant of this fact , which was originally formulated in terms of _ allowable sequences _ by goodman and pollack @xcite .
first , we observe that oriented matroids also arise from generalization of point configurations , called _ pseudoconfigurations of points _
( also called _ generalized configurations of points _ ) .
( pseudoconfigurations of points ) + a pair @xmath78 of a point configuration @xmath79}$ ] in @xmath80 and a collection @xmath81 of unbounded jordan curves is called a _ pseudoconfiguration of points _ ( or a _ generalized configuration of points _ ) if the following hold . * for any @xmath82 , there exist at least two points of @xmath6 lying on @xmath83 . *
for any two points of @xmath6 , there exists a unique curve in @xmath81 that contains both points . *
any pair of ( distinct ) curves @xmath84 intersects at most once . for each @xmath82
, we label the two connected components of @xmath85 arbitrarily as @xmath86 and @xmath87 . then
, we assign the sign vector @xmath88 such that @xmath89 \mid p_e \in l\}$ ] , @xmath90 \mid p_e \in l^+\}$ ] and @xmath91 \mid p_e \in l^-\}$ ] , and we let @xmath92 .
then , @xmath93 , { \cal c}^*_{pp})$ ] turns out to be an acyclic oriented matroid of rank @xmath94 .
goodman and pollack @xcite proved that in fact the converse also holds .
( topological representation theorem for acyclic oriented matroids of rank @xmath94 @xcite ) + for any acyclic oriented matroid @xmath95 of rank @xmath94 , there exists a pseudoconfiguration of points @xmath96 with @xmath97 . ] here , the assumption that @xmath95 is acyclic is not important , because non - acyclic oriented matroids can be represented as _ signed pseudoconfigurations of points _ , where each point has a sign . the notion of pseudoconfigurations of points in the @xmath98-dimensional euclidean space can be introduced analogously , but not every acyclic oriented matroid of rank @xmath25 can be represented as a pseudoconfiguration of points in the @xmath99-dimensional euclidean space . for further details ,
see ( * ? ? ?
* section 5.3 ) .
however , oriented matroids of general rank can be represented as pseudosphere arrangements @xcite , with further details presented in ( * ? ? ?
* section 5.2 ) .
in this section , we introduce degree-@xmath0 oriented matroids as a combinatorial model , which captures the essential combinatorial properties of configurations of points and graphs of polynomial functions . to do so , we first review some results that were introduced by eli and matouek @xcite .
eli and matouek @xcite introduced the notion of @xmath0th - order monotonity , which is a generalization of the usual notion of monotonity , described as follows .
( because it is more convenient in our context to reinterpret the @xmath100st - order monotonity of eli and matouek as the @xmath0th - order mononity , the following definitions are slightly different from the originals . )
let @xmath101 be a point configuration in the @xmath1-dimensional euclidean space with @xmath102 .
we assume that @xmath6 is in _ @xmath0-general position _ ,
i.e. , no @xmath103 points of @xmath6 lie on the graph of a polynomial function of degree at most @xmath0 .
furthermore , we define a @xmath104-tuple of points in @xmath6 to be _ positive _ ( resp . _ negative _ ) if they lie on the graph of a function whose ( @xmath105)st order derivative is everywhere nonnegative ( resp .
nonpositive ) .
a subset @xmath106 $ ] is said to be _
@xmath0th - order monotone _ if its @xmath104-tuples are either all positive or all negative .
this notion can be stated in an alternative manner using the map @xmath107^{k+2 } \rightarrow \ { + 1,-1,0\}$ ] , defined as follows .
@xmath108 where @xmath109 is the newton interpolation polynomial of the points @xmath110 , i.e. , @xmath111 is the unique polynomial function of degree @xmath0 whose graph passes through the points @xmath110 . under this definition ,
a subset @xmath3 is @xmath0th - order monotone if and only if @xmath112 for all @xmath113 ( see ( * ? ? ?
* lemma 2.4 ) ) .
this map contains information on which side of the graph of the polynomial function of degree @xmath0 determined by @xmath110 the point @xmath114 lies . in other words , the map @xmath115 contains information regarding the partitions of @xmath6 by graphs of polynomial functions of degree @xmath0 .
+ + it is shown in @xcite that the map @xmath115 can be computed in terms of determinants of a higher - dimensional space .
( @xmath104-dimensional linear representability ( * ? ? ? * lemma 5.1 ) ) + let @xmath116 be the map that sends each point @xmath117 to @xmath118 .
then , we have @xmath119^{k+2}$},\end{aligned}\ ] ] i.e. , the map @xmath115 is a chirotope of an oriented matroid of rank @xmath103 .
the proof is not difficult , and we refer the reader to @xcite .
the latter statement will be proved in proposition [ prop : pp_def_om ] in a more general context .
+ + eli and matouek @xcite additionally observed the following useful property .
( transitivity ( * ? ? ?
* lemma 2.5 ) ) + if @xmath120 for @xmath121,k+3)$ ] , then we have @xmath122 for all @xmath123 .
[ prop : tans ] this proposition is proved in @xcite , using newton interpolation polynomials . here
, we provide a geometric proof for @xmath124 .
the generalization of this is straightforward . without loss of generality , we assume that @xmath125 , which means that the point @xmath126 is above the graph @xmath127 .
note that the graphs @xmath127 and @xmath128 intersect at @xmath129 and @xmath130 , and that they do not intersect elsewhere .
this indicates that the graph @xmath128 lies above the graph @xmath127 for @xmath131 .
the point @xmath132 is above the graph @xmath128 by the assumption @xmath133 , and it follows that @xmath132 lies above the graph @xmath127 , which implies that @xmath134 .
the same argument can be applied when either of the graphs @xmath135 or @xmath136 are considered instead of @xmath127 . + + in @xcite , this property is used to prove a new generalization of the erds - szekeres theorem , which states that there is always a @xmath0-monotone subset of size @xmath137 in any point set @xmath6 of size @xmath138 in @xmath0-general position , where @xmath139 is the @xmath0th tower function .
our first observation is that @xmath140 actually admits the following stronger property .
( ( @xmath141)-locally unimodal property ) + for any @xmath142,k+3)$ ] and @xmath143 with @xmath144 ( lexicographic order , i.e. , there exist @xmath145 ( @xmath146 ) with @xmath147 ) , it holds that if @xmath148 , then @xmath149 . the case with @xmath124 can easily be proved by careful examination of the proof of proposition [ prop : tans ] .
we will prove this proposition later in a more general context ( proposition [ prop : pp_def_om ] ) , and so we omit the full proof here . + + in the next section , we will prove that the above - mentioned properties are in some sense _ all _ combinatorial properties that can be proved using some natural geometric properties .
this motivates us to consider combinatorial structures characterized by those properties .
( degree-@xmath0 oriented matroids ) + let @xmath5 be a finite ordered set .
we say that a rank @xmath103 uniform oriented matroid @xmath150 is called a _
degree-@xmath0 uniform oriented matroid _ if it satisfies the following condition . for any @xmath151 and @xmath143 with @xmath144 ( lexicographic order )
, it holds that if @xmath152 , then @xmath153 .
in this section , we prove that degree-@xmath0 oriented matroids can always be represented by the following generalization of configurations formed by points and graphs of polynomial functions of degree @xmath0 .
( @xmath0-intersecting pseudoconfiguration of points ) + [ def : pp_configuration ] a pair @xmath78 of a point configuration @xmath154 ( @xmath155 ) in the 2-dimensional euclidean space and a collection @xmath81 of @xmath21-monotone jordan curves is called a _ @xmath0-intersecting pseudoconfiguration of points _
if the following conditions hold : for any @xmath82 , there exist at least @xmath105 points of @xmath6 lying on @xmath83 . for any @xmath105 points of @xmath6
, there exists a unique curve @xmath82 passing though each point . for any @xmath84 ( @xmath156 ) , @xmath157 and @xmath158 intersect ( transversally ) at most @xmath0 times . here , a jordan curve is called _ @xmath21-monotone _ if it intersects with any vertical line at most once . for @xmath82 , we denote @xmath159 . if @xmath160 for all @xmath82 , then the configuration @xmath96 is said to be _
simple_. when @xmath96 is a simple configuration , we denote the curve determined by points @xmath110 by @xmath161 .
an @xmath21-monotone jordan curve @xmath83 can be written as @xmath162 , for some continuous function @xmath163 .
we define @xmath164 , @xmath165 , and @xmath166 , @xmath167 .
we call @xmath86 ( resp .
@xmath87 ) the _ @xmath168-side _ ( resp .
_ @xmath169-side _ ) of @xmath83 . for @xmath170 ,
the _ subconfiguration _ induced by @xmath171 is a @xmath0-intersecting pseudoconfiguration @xmath172 , where @xmath173 . + -intersecting pseudoconfiguraiton of points ]
+ for each @xmath0-intersecting pseudoconfigurations of points , a partition function is defined in a similar manner as in the case of configurations formed by points and graphs of polynomial functions of degree @xmath0 . to define and analyze this
, we require the following notion .
( lenses ) + let @xmath174 be a @xmath0-intersecting pseudoconfiguration of points . a _ lens _ ( or _ full lens _ ) of @xmath174 is a region that can be represented as @xmath175 \times \mathbb{r})$ ] , where @xmath176 , and @xmath177 and @xmath178 are the @xmath21-coordinates of consecutive intersection points of @xmath157 and @xmath158 .
note that @xmath179 \times \mathbb{r})$ ] consists of two points .
we call these the _ end points _ of @xmath180 .
the end point with the greater ( resp .
smaller ) @xmath21-coordinate is called the right ( resp .
left ) end point , and is denoted by @xmath181 ( resp .
@xmath182 ) . a _ half lens _ of @xmath174 is a region represented as @xmath183 or @xmath184 \times \mathbb{r})$ ] , where @xmath176 , and @xmath177 and @xmath178 are the @xmath21-coordinates of the leftmost and rightmost intersection points of @xmath157 and @xmath158 , respectively .
we call a full or half lens @xmath180 an _ empty lens _ if @xmath185 .
given a simple @xmath0-intersecting pseudoconfiguration of points @xmath78 ( or more generally a configuration satisfying only ( pp1 ) and ( pp2 ) ) , we define a map @xmath186^{k+2 } \rightarrow \ { + , -,0\}$ ] as follows . * for @xmath187
$ ] with @xmath188 , we have @xmath189 * @xmath190 if @xmath191 for some @xmath192 $ ] ( @xmath193 ) . *
@xmath194 for any @xmath187 $ ] and any permutation @xmath43 on @xmath195 $ ] .
the map @xmath196 is well - defined .
it suffices to show that for any permutation @xmath43 on @xmath195 $ ] with @xmath197 and any sign @xmath198 , we have @xmath199 .
first , we pick such a @xmath43 arbitrarily . then
, we have @xmath200 for some @xmath201 $ ] and @xmath202 . if @xmath203 , then the proposition is trivial , and thus we assume that @xmath204 .
then , the curves @xmath205 and @xmath206 intersect at the points @xmath207 . by the condition of @xmath0-intersecting pseudoconfigurations of points , these two curves must not intersect elsewhere , and thus the curve @xmath206 must lie above @xmath205 over the interval @xmath208 if @xmath209 is above @xmath205 . because the above - below relationship of @xmath205 and @xmath206 is reversed at each end point of each lens formed by these two curves , the curve @xmath206 must lie above ( resp .
below ) @xmath205 over the interval @xmath210 if @xmath211 is even , i.e. , if @xmath212 ( resp . if @xmath211 is odd , i.e. , if @xmath213 ) .
therefore , we have @xmath214 .
the same discussion applies in the case that @xmath209 is below @xmath205 . for every simple @xmath0-intersecting pseudoconfiguration of points @xmath215},l)$ ] ,
the map @xmath196 defines a degree-@xmath0 uniform oriented matroid .
[ prop : pp_def_om ] first , we prove that the map @xmath196 is a chirotope of an oriented matroid of rank @xmath103 . for this , it suffices to prove that the set @xmath216^{k+1 } \ }
\setminus \ { 0 \}$ ] fulfills the cocircuit axioms of oriented matroids .
clearly , axioms ( c0)(c2 ) are satisfied and we need only verify axiom ( c3 ) . because we have @xmath217 for all @xmath218 , it suffices to verify axiom ( c3 ) .
take @xmath219,k+1)$ ] with @xmath220 .
let @xmath221 be sign vectors that correspond to @xmath222 and @xmath223 , respectively ( @xmath4 and @xmath224 are determined uniquely up to a sign reversal ) .
take an @xmath58 and any @xmath225 , and let @xmath226 be the sign vector with @xmath227 that corresponds to @xmath228 .
let us verify that @xmath229 is a required cocircuit in ( c3 ) .
note that @xmath222 and @xmath223 form @xmath105 lenses ( see figure [ fig : pp_def_om ] ) . because @xmath228 already intersects with @xmath222 and @xmath223 @xmath0 times at the points with indices in @xmath230 , it can not intersect with @xmath222 or @xmath223 elsewhere .
therefore , if @xmath231 is contained inside of one of the lenses , then the whole of @xmath228 must lie in the lenses .
take any @xmath232 $ ] with @xmath233 .
then , @xmath234 lies outside of the lenses formed by @xmath222 and @xmath223 , and thus @xmath234 and @xmath235 lie on the same side of @xmath228 . if @xmath231 is outside of the lenses , then the whole of @xmath228 must lie outside of the lenses ( except for the end points ) .
points @xmath234 with @xmath233 corresponds to points inside of the lenses and a similar discussion shows that @xmath236 .
therefore , we have @xmath61 and @xmath62 .
next , we confirm the ( @xmath141)-locally unimodal property .
suppose that there exist @xmath142,k+3)$ ] and @xmath237 such that @xmath238 and @xmath239 , @xmath240 .
let @xmath241 and @xmath242 ( @xmath146 ) be the integers such that @xmath243 , @xmath244 , and @xmath245 .
let @xmath246 be the integer such that @xmath247 , where we assume that @xmath248 and @xmath249 .
let the integers @xmath250 and @xmath251 be defined similarly . since @xmath252 ,
the point @xmath253 is on the @xmath254-side of @xmath255 and on the @xmath256-side of @xmath257 . because @xmath255 and @xmath257 must not intersect more than @xmath0 times , the curves @xmath255 and @xmath257 form lenses with end points @xmath258 , and @xmath259 must lie inside of these lenses .
now , we remark that the point @xmath260 is on the @xmath261-side of @xmath255 and the @xmath262-side of @xmath259 , because @xmath263 and @xmath264 .
therefore , the curve @xmath257 must intersect with either @xmath255 or @xmath259 in @xmath265 .
this means that @xmath257 must intersect with either @xmath255 or @xmath259 at least @xmath105 times , which is a contradiction . here , we prove that every degree-@xmath0 oriented matroid admits a geometric representation as a @xmath0-intersecting pseudoconfiguration of points . to this end , we introduce two operations .
( empty lens elimination i ) + let @xmath180 be an empty lens represented as @xmath266 \times \mathbb{r})$ ] , where @xmath84 , and @xmath177 ( resp .
@xmath178 ) can be @xmath267 ( resp .
@xmath268 ) . transform @xmath157 and @xmath158 by connecting @xmath269 $ ] , @xmath270 $ ] , and @xmath271 ( when @xmath272 ) , and by connecting @xmath273 $ ] , @xmath274 $ ] , and @xmath275 ( when @xmath276 ) , for sufficiently small @xmath277 , so that the new curves do not touch around the vertical lines @xmath278 and @xmath279 ( see figure [ fig : transform1 ] ) .
( empty lens elimination ii ) + let @xmath280 be lenses represented as @xmath281 \times \mathbb{r})$ ] and @xmath282 \times \mathbb{r})$ ] ,
, @xmath283 \times \mathbb{r})$ ] where @xmath84 , and @xmath284 , @xmath285 .
suppose that @xmath286 , @xmath287 , , @xmath288 and @xmath289 are empty for some @xmath290 .
transform @xmath157 and @xmath158 by connecting @xmath291 $ ] , @xmath292 $ ] , and @xmath293 , and by connecting @xmath294 $ ] , @xmath295 $ ] , and @xmath296 , for sufficiently small @xmath277 , so that the new curves do not touch around the vertical lines @xmath297 and @xmath298 ( see figure [ fig : transform2 ] ) .
the two operations decribed above decrease the total number of intersection points of the curves in @xmath81 , without altering the above - below relationships between the points in @xmath6 and the curves in @xmath81 .
however , they may invalidate the condition that each pair of curves in @xmath81 intersect at most @xmath0 times .
however , we prove that a @xmath0-intersecting pseudoconfiguration of points is always obtained after the operations are applied as far as possible .
let @xmath299 be a simple configuration with a finite point set @xmath6 in @xmath80 and a collection @xmath81 of @xmath21-monotone jordan curves satisfying ( pp1 ) and ( pp2 ) in definition [ def : pp_configuration ] ( with some pair of curves possibly intersecting more than @xmath0 times ) .
we assume that the map @xmath196 fulfills the axioms of degree-@xmath0 oriented matroids .
if it is impossible to apply the empty lens eliminations to @xmath96 , then @xmath96 is a @xmath0-intersecting pseudoconfiguration of points .
assume that there is a pair of curves in @xmath81 that intersect more than @xmath0 times .
let @xmath300 be a minimal subset of @xmath6 such that there are two curves in @xmath301 intersecting more than @xmath0 times after the empty lens eliminations are applied as far as possible .
let @xmath157 and @xmath158 be curves in @xmath301 that have the smallest @xmath21-coordinate for the @xmath100st intersection point .
the curves @xmath157 and @xmath158 form @xmath302 full lenses and two half lenses .
let us label these @xmath303 full and half lenses by @xmath304 in increasing order for the @xmath21-coordinates .
because the empty lens eliminations i and ii can not be applied , there must exist @xmath303 distinct points @xmath305 with @xmath306 .
note that there are no points outside of the lenses @xmath307 , by the minimality of @xmath300 .
let @xmath308 and @xmath309 .
we now consider several possible cases separately .
+ + * ( case i ) * @xmath310 . the curve @xmath311 must intersect with @xmath157 and @xmath158 at least twice in total in each of the full lenses @xmath312 ( to enter and to leave , where passing through an end point is counted twice ) and at least once in each of the half or full lenses @xmath313 and @xmath314 . if @xmath311 intersects with @xmath157 and @xmath158 more than twice in total in some lens @xmath315 ( @xmath316 ) , this means that they intersect at least four times in @xmath315 , and that either of the @xmath100st intersection point of @xmath157 and @xmath311 or that of @xmath158 and @xmath311 belongs to the halfspace @xmath317 ) .
this contradicts the minimality assumption for the @xmath21-coordinate of the @xmath100st intersection point of @xmath157 and @xmath158 .
therefore , the curve @xmath311 actually intersects with @xmath157 and @xmath158 exactly twice in total in each full lens and once or twice in each half lens . if @xmath318 , then the curve @xmath319 intersects with @xmath157 and @xmath158 a total of @xmath320 or @xmath321 times in the halfspace @xmath322 , and thus @xmath105 times with @xmath157 or @xmath158 . without loss of generality , we assume that @xmath311 and @xmath157 intersect @xmath105 times in the halfspace @xmath323 . because of the minimality assumption
, the @xmath100st intersection point of @xmath311 and @xmath157 must coincide with @xmath324 .
let @xmath325 be the lenses formed by @xmath157 and @xmath311 ( ordered by the @xmath21-coordinates ) .
we have @xmath326 , , @xmath327 .
on the other hand , if @xmath328 , then it must hold that @xmath329 .
this mirrors the case with @xmath318 and the remainder of the discussion proceeds in the same manner .
therefore , we consider only the case with @xmath318 in what follows .
we can classify the possible situations into the following two cases .
+ + * ( case i - i ) * @xmath330 .
let @xmath331 \mid p_u \in ( \widetilde{p } \setminus \widetilde{p}(l_{i_{k+1 } } ) ) \cap \bigcup_{i=1}^{k+1}{(r'_i \setminus \
{ e_l(r'_i ) , e_r(r'_i ) \ } ) } \}$ ] , and let @xmath332 $ ] be such that @xmath333 .
the curves @xmath334}$ ] and @xmath311 intersect @xmath105 times in the halfspace @xmath335 ( cf .
figures [ fig : discussion2 ] and [ fig : discussion2_2 ] ) . let @xmath336 be the @xmath103 leftmost lenses formed by @xmath334}$ ] and @xmath311 , which are labeled in increasing order of the @xmath21-coordinates .
then , we have @xmath337 , @xmath338 , and @xmath339 .
assume that @xmath340 and @xmath341 .
then , by the above - below relationship of @xmath334}$ ] and @xmath311 , it must hold that if @xmath342 is on the @xmath43-side ( @xmath343 ) of @xmath344 , then @xmath345 must lie on the @xmath43-side of @xmath346}$ ] ( cf .
figures [ fig : discussion2 ] and [ fig : discussion2_2 ] ) .
this implies that @xmath347,i_{j^*})$ ] , which contradicts the chirotope axioms .
a similar discussion also leads to a contradiction if @xmath348 and @xmath349 .
therefore , we have @xmath350 or @xmath351 .
first , we consider the case that @xmath350 .
let us assume that @xmath352 .
remark that there must exist a point @xmath353 for some @xmath354 $ ] because otherwise empty lens elimination ii can be applied to @xmath355 . without loss of generality , we assume that @xmath356 and @xmath357,a ) = -1 $ ]
. then , we have @xmath358 \setminus \ { j \ } ) = ( -1)^{k - a^*}$ ] and @xmath358 \setminus \ { i_{j^ * } \ } ) = ( -1)^{k - a^*+1}$ ] , where @xmath359 : = ( i_1,\dots , i_{j^*},j , i_{j^*+1},\dots , i_{a^*-1},a , i_{a^*},\dots , i_{k+1})$ ] . by the @xmath360-locally unimodal property , we have @xmath361 \setminus \ { a \ } ) = ( -1)^{k - a^*}$ ] , and thus @xmath362 .
this means that the point @xmath342 is on the @xmath363-side of @xmath311 , and thus that @xmath334}$ ] is on the @xmath363-side of @xmath311 at @xmath364 .
on the other hand , by the assumption @xmath356 and @xmath357,a ) = -1 $ ] , the curve @xmath334}$ ] must be above @xmath311 at @xmath365 . because the above - below relationship of @xmath334}$ ] and @xmath311 is reversed at each end point of each lens , the curve @xmath334}$ ]
must lie on the @xmath366-side of @xmath311 at @xmath364 .
this is a contradiction .
we also obtain a contradiction in the case that @xmath367 . the case that @xmath351 can be treated similarly .
+ * ( case i - ii ) * @xmath368 .
let @xmath369 be the points of @xmath370 ordered in increasing order of the @xmath21-coordinates .
then , for some @xmath371 $ ] it holds that @xmath372 and @xmath373 .
first , we assume that @xmath374 and derive a contradiction .
take @xmath375 , and note that @xmath376 .
then , the curve @xmath377}$ ] intersects with @xmath157 and @xmath344 twice in total in each of the lenses @xmath378 , @xmath379 , and once or twice in @xmath380 ( see figure [ fig : discussion3 ] ) .
now , we remark that @xmath381}$ ] does not intersect twice with @xmath157 or @xmath311 inside each of the lenses listed above because otherwise an empty lens is formed . therefore ,
@xmath381}$ ] intersects exactly once with @xmath157 and @xmath311 in each of the full lenses listed above .
let us consider the case that @xmath382 .
if @xmath381}$ ] intersects a total of two times with @xmath157 and @xmath311 in @xmath383 , then it intersects a total of more than @xmath384 times with @xmath157 and @xmath311 , and thus more than @xmath0 times with @xmath157 or @xmath311 . since @xmath385 } ) \cup \widetilde{p}(l_1 ) \subset \widetilde{p } \setminus \ { p_{i_c } \}$ ] and @xmath385 } ) \cup \widetilde{p}(l_{i_{k+1 } } ) \subset \widetilde{p } \setminus \ { p_{j_{b+2 } } \}$ ] , this is a contradiction .
if @xmath381}$ ] intersects just once in total with @xmath157 and @xmath311 in @xmath383 , then @xmath311 lies between @xmath157 and @xmath381}$ ] around the vertical line @xmath386 , and the situations are the same for all of the end points of the lenses @xmath387 .
this leads to the conclusion that @xmath381}$ ] intersects with both @xmath157 and @xmath311 in @xmath388 , and thus that @xmath381}$ ] intersects more than @xmath0 times with @xmath157 or @xmath311 .
this contradicts the minimality assumption of @xmath300 .
we also obtain a contradiction in the case that @xmath389 .
finally , we see that a contradiction also occurs in the case that @xmath390 .
let @xmath375 .
assume that @xmath157 and @xmath311 intersect in @xmath391 , and form a full lens @xmath392 .
then , we can apply empty lens elimination ii to @xmath393 and @xmath392 , which is a contradiction .
therefore , the curves @xmath157 and @xmath311 do not intersect in @xmath391 . without loss of generality ,
we assume that @xmath157 is above @xmath311 at @xmath394 , i.e. , @xmath395 .
then , by considering the above - below relationship of @xmath157 and @xmath311 in each lens , we see that the point @xmath396 is on the @xmath397-side of @xmath157 , i.e. , @xmath398 .
this contradicts the chirotope axioms .
+ + * ( case ii ) * @xmath399 or @xmath400 .
we can apply the same discussion as above , by considering @xmath157 and @xmath158 instead of @xmath157 and @xmath311 .
let @xmath95 be a degree-@xmath0 uniform oriented matroid on the ground set @xmath401 $ ] .
then , there exists a @xmath0-intersecting pseudoconfiguration of points @xmath402},l)$ ] such that @xmath97 .
let @xmath403 .
for each @xmath404,k+1)$ ] , we construct @xmath161 so that @xmath342 lies above ( resp .
below ) it if @xmath405 ( resp .
@xmath406 ) and @xmath110 lie on @xmath161 , and any three curves intersect at the same point @xmath407 . then ,
apply empty lens eliminations i and ii as far as possible .
in this paper , we have introduced a new class of oriented matroids , called degree-@xmath0 oriented matroids , which abstracts the combinatorial behavior of partitions of point configurations in the 2-dimensional euclidean space by graphs of polynomial functions of degree @xmath0 . as a first step
, we proved that there exits a natural class of geometric objects , called @xmath0-intersecting pseudoconfigurations of points , that corresponds to degree-@xmath0 oriented matroids .
this provides evidence that the defintion of degree-@xmath0 oriented matroids is natural one . from another point of view , degree-@xmath0 oriented matroids provide a class of oriented matroids of rank @xmath103 that can be represented by @xmath1-dimensional geometric objects
. it would be interesting in future work to verify open conjectures , such as the las vergnas simplicial conjecture @xcite , which are known to be true for rank 3 oriented matroids @xcite but not for those of higher rank .
the author is grateful to yuya hodokuma for the opportunity to read the paper @xcite together , which motivated the author to introduce degree-@xmath0 oriented matroids .
this research is partially supported by jsps grant - in - aid for young scientists ( b ) 26730002 .
99 f. ardila and m. develin .
tropical hyperplane arrangements and oriented matroids , mathematische zeitschrift , volume 262 , issue 4 , pp.795816 , 2009 .
a. bjrner , m. las vergnas , b. sturmfels , n. white and g.m .
oriented matroids ( 2nd edition ) , cambridge university press , 1999 .
bland and m. las vergnas .
orientability of matroids , journal of combinatorial theory , series b , volume 24 , issue 1 , pp . 94123 , 1978 . j. bokowski , s. mock and i. streinu .
on the folkman-lawrence topological representation theorem for oriented matroids of rank 3 , european journal of combinatorics , volume 22 , issue 5 , pp . 601615 , 2001 . j. bokowski , s. king , s. mock and i. streinu .
the topological representation of oriented matroids , discrete and computational geometry , volume 33 , issue 4 , pp . 645668 , 2005 . m. eli and j. matouek .
higher - order erds - szekeres theorems , advances in mathematics , volume 244 , issue 10 , pp . 115 , 2013 . j. folkman and j. lawrence .
oriented matroids , journal of combinatorial theory , ser .
b 25 , pp . 199236 , 1978 . j.e .
goodman and r. pollack .
semispaces of configurations , cell complexes of arrangements , journal of combinatorial theory , series a , volume 37 , issue 3 , pp . 257293 , 1984 . s. horn .
topological representation theorem for tropical oriented matroids , journal of combinatorial theory , series a , volume 142 , pp . 77112 , 2016 .
m. las vergnas .
convexity in oriented matroids , journal of combinatorial theory , series b , volume 29 , pp . 231234 , 1980 .
die teilung der projektiven ebene durch geradeoder pseudogerade , ber .
math - phys .
256267 , 1926 . h. miyata .
a 2-dimensional topological representation theorem for rank @xmath2 matroid polytopes , ieice technical report , volume 116 , issue 211 , pp . 4552 , 2016 . f. santos .
on delaunay oriented matroids for convex distance functions , discrete and computational geometry , volume 16 , issue 2 , pp.197210 , 1996 . | oriented matroids are a combinatorial model , which can be viewed as a combinatorial abstraction of partitions of point sets in the euclidean space by families of hyperplanes .
they capture essential combinatorial properties of point configurations , hyperplane arrangements , and polytopes , and oriented matroid theory has been developed in the context of various research fields . in this paper
, we introduce a new class of oriented matroids , called degree-@xmath0 oriented matroids , which captures the essential combinatorial properties of partitions of point sets in the @xmath1-dimensional euclidean space by graphs of polynomial functions of degree @xmath0 .
we prove that the notion of degree-@xmath0 oriented matroids completely characterizes combinatorial structures arising from a natural geometric generalization of configurations formed by points and graphs of polynomial functions degree @xmath0 .
this may be viewed as an analogue of the folkman - lawrence topological representation theorem for oriented matroids . |
the nature of the central engine of gamma - ray bursts ( grbs ) remains unclear .
currently favored models invoke a binary merger or a collapse of compact objects .
these models lead to the formation of a transient hot and dense accretion torus / disk around a black hole ( bh ) of a few solar masses .
the typical mass accretion rates in grb models are extremely high , of the order of a fraction of solar mass up to a few solar masses per second . under such conditions ,
the disk becomes dense and hot enough in the inner regions to cool via neutrino losses .
for this reason , popham , woosley & fryer ( 1999 , hereafter pwf99 ) named these disk neutrino - dominated accretion flows ( ndaf ) .
energy extraction from the bh - accretion disk for powering grb is also possible , such as by blandford - znajek ( bz ) process ( see lee et al .
2000 for a review of this model ) , and blandford - payne ( bp ) process and parker instabilities in the disk ( narayan et al .
1992 ; meszaros & rees 1997 ) .
ndaf has been extensively discussed by many authors , e.g. , pwf99 , narayan , piran & kumar ( 2001 , hereafter npk01 ) , kohri & mineshige ( 2002 ) , di matteo , perna & narayan ( 2002 , hereafter dpn ) , chen & beloborodov ( 2006 ) , janiuk et al .
( 2004 , 2007 ) and gu , liu & lu ( 2006 , hereafter gll06 ) .
however , due to its low energy conversion efficiencies and the effects of neutrino opacity , the power produced by neutrino - anineutrino annihilation can hardly to match those of the energetic short - hard grbs ( e.g. grb080913 ) and x - ray flares .
recently , some authors ( e.g. , dpn02 , fan et al . 2005 , shibata et al .
2006 and 2007 and perez - ramirez et al .
2008 ) suggested that the mhd process should be considered in the disk model .
moreover , as shown in shibata et al .
( 2006 , 2007 ) , the magnetic braking and magneto - rotational instability ( mri , balbus & hawley 1991 ) in the disk play a role in angular momentum transporting , which causes turbulent motion , resultant shock heating , and mass accretion onto the bh . on the other hand
, researches showed that the magnetic fields can be magnified up to @xmath0 g by virtue of mri or dynamo process ( pudritz & fahlman 1982 and references therein ) in hyperaccretion disk .
these considerations stimulate us to discuss the magnetized ndaf .
based on the work of npk01 and dpn02 , xie et al .
( 2007 ) discussed the bz and bp processes in ndaf .
they found that the jet of grb may be magnetically - dominated , which is also obtained by mhd simulations of mizuno et al.(2004 ) .
recently , the magnetic coupling ( mc ) between the central spinning bh and their surrounding accretion disk has been paid much attention ( e.g. blandford 1999 ; van putten 1999 ; li & paczynski 2000 ; li 2002 ; wang et al .
2002 ) . as a variant of the bz process
, the mc process exerts a torque on the disk , and transports the rotational energy from the bh to the disk .
the effects of mc torque has been discussed in some disk models , for example , lai ( 1998 ) and lee ( 1999 ) in a neutron star with slim disk , li ( 2002 ) , wang et al .
( 2002 , 2003 ) , kluzniak and rappaport ( 2007 ) in a compact object with thin disk , ye et al .
( 2007 ) and ma et al .
( 2007 ) in a bh with advection - dominated accretion flow ( adaf ) .
it is found , the disk properties are greatly changed and its luminosity is augmented significantly due to the rotational energy of bh extracted in the mc process .
therefore , it is attractive for us to investigate the effects of mc torque on ndaf . to highlight the effects of mc torque , we ignore other mhd process , such as bz and bp mechanism , and
we refer to this model as mcndaf .
this paper is organized as follows . in sect .
2 we describe the mcndaf model , which is a relativistic steady state thin disk .
the effects of mhd stress are described by the dimensionless parameter @xmath1 .
the main equations are based on dpn02 and npk01 .
recently , gll06 , chen & beloborodov ( 2006 ) and shibata et al .
( 2006 , 2007 ) argued that the general relativistic ( gr ) effects are important for ndaf , so we introduce gr correction factors to the equations .
the mc torque appears in the angular momentum equation .
we solve the set of equations for the solutions in mcndaf in sect . 3 , and compute the neutrino and neutrino annihilation luminosities .
following gll06 , we include the neutrino radiation from the optically thick region in the computation for the neutrino luminosity .
to show the effects of mc torque , we compare it with previous results . in sect .
4 we discuss the stability of the accretion flow .
we also discuss the physical origin of the instabilities in mcndaf .
finally , we summarize our results and discuss some related issues in sect .
considering that central bhs are rapidly rotating in most candidate grb engines , we discuss a model of a steady state disk around a kerr bh , in which neutrino loss and transfer are taken into account .
our model is presented in the context given by dpn02 , and the gr corrections are adopted from riffert & herold ( 1995 ) . as mentioned by dpn02 and pwf99 ,
although in grb central engines the accretion rate may vary , it is expected to vary significantly only in the outer disk .
hence it seems reasonable to study the main properties in the inner neutrino - cooled disk by assuming a constant accretion rate . because the gas cools efficiently
, we are entitled to discuss the mcndaf model in the context of a thin disk ( shakura & sunyaev 1973 ) .
the accuracy of the thin - disk approximation is not perfect at large radii , where the disk is thick .
fortunately , the details of the outer region have little effect on the solution for the neutrino - cooled disk ( chen & beloborodov 2006 ) .
the mcndaf model is a relativistic steady thin disk , and the large - scale magnetic field contributing to the mc process and the small - scale tangled magnetic field related to the viscosity are included .
we assume that these two kinds of fields work independently , and the large - scale magnetic field remains constant at the bh horizon .
following blandford ( 1976 ) we assume that the magnetic field @xmath2 on the disk varies as @xmath3 , where @xmath4 is the disk radius in terms of the marginally stable orbit @xmath5 ( novikov & thorne 1973 ) , and @xmath6 is the power law index indicating the degree of concentration of the magnetic field in the central region of the disk .
based on equipartition relation the magnetic field at the horizon is related to the mass density at the inner disk as follows ( mckinney 2005 ) , @xmath7 where @xmath8 , @xmath9 and @xmath10 .
the mc torque is derived in wang et al .
( 2002 ) based on an equivalent circuit given by macdonald & thorne ( 1982 ) as follows , @xmath11 where @xmath12 , @xmath13 is the dimensionless bh spin parameter defined by the bh mass @xmath14 and angular momentum @xmath15 , @xmath16 and @xmath17 is the ratio of the angular velocity of the disk @xmath18 to that of the horizon @xmath19 $ ] .
the mapping relation between the angular coordinate @xmath20 on the horizon and the radial coordinate @xmath21 on the disk is derived based on the conservation of magnetic flux as follows ( wang et al . 2003 ) , @xmath22 where @xmath23 , @xmath24 where @xmath25 . a number of works ( pwf99 ; chen & beloborodov , 2006 ) used accurate equations of relativistic hydrodynamics in kerr spacetime to study ndaf .
gll06 found that the gr effect must be taken into account for the power of grb .
the relativistic correction factors for a thin accretion disk around a kerr bh have been given by riffert & herold ( 1995 ) , @xmath26 @xmath27 @xmath28 @xmath29 the equation of the conservation of mass remains valid , while hydrostatic equilibrium in the vertical direction leads to a corrected expression for the half thickness of the disk ( riffert & herold 1995 ; reynoso , romero & sampayo 2006 ) : @xmath30 the viscous shear @xmath31 is corrected by @xmath32 the basic equations of mcndaf are given as follows .
the continuity equation : @xmath33 \2 . the total pressure consists of five terms , radiation pressure , gas pressure , degeneracy pressure , neutrino pressure and magnetic pressure : @xmath34 where @xmath35 is the magnetic pressure contributed by the tangled magnetic field in the disk , and @xmath36 is the ratio of the magnetic pressure to the total pressure .
@xmath37 is the neutrino energy density defined as ( popham & narayan 1995 ) @xmath38 in equation ( [ eq13 ] ) @xmath39 is the sum of absorptive and scattering optical depths calculated for each neutrino flavor @xmath40 .
the absorptive optical depths for the three neutrino flavors are ( kohri et al . 2005 ) @xmath41 @xmath42 where @xmath43 is the mass fraction of free nucleons approximately given by ( e.g. , pwf99 ; qian & woosley 1996 ) , @xmath44 the total scattering optical depth is given by dpn02 as @xmath45 \3 . combining the conservation of the angular momentum with equation ( [ eq11 ] )
, we have @xmath46 where @xmath47 is the specific angular momentum of the accreting gas .
the flux of angular momentum transferred magnetically from the bh to the disk , @xmath48 , is related to the mc torque @xmath49 by @xmath50 vanishing of @xmath31 ( or @xmath51 at @xmath5 leads to @xmath52 \4 .
the equation for the energy balance is @xmath53 where @xmath54 represents the viscous dissipation , and @xmath55 is the total cooling rate due to neutrino losses @xmath56 , photodisintegration @xmath57 and advection @xmath58 . we employ a bridging formula for calculating @xmath56 , which is valid in both the optically thin and thick cases .
the expressions for @xmath56 , @xmath57 and @xmath58 are ( dpn02 ; gll06 ) @xmath59 @xmath60 @xmath61 where @xmath62 is the entropy density of neutrinos .
note that the cooling function given by the bridging formula reduces to the optically thin expression for small optical depths ( as adopted in pwf99 ) but differs significantly from the latter at optical depths @xmath63 1 . by considering the mc effects ,
the heating rate @xmath64 is expressed as @xmath65 where the second term is the mc contribution . as we can see from equation ( [ eq20 ] ) ,
the magnetic torque may deposit angular momentum in the inner disk , and this extra angular momentum must be transported outwards by the viscous torque in the disk , resulting in energy dissipation and increasing the disk luminosity based on equation ( [ eq25 ] ) . defining @xmath66 and
@xmath67 as the contributions due to the gravitational release and the mc process , respectively , we have the ratio @xmath68 versus the disk radius @xmath69 as shown in figure 1 . from figure 1
we find that @xmath70 is much greater than @xmath71 in the inner disk , where the neutrino cooling dominates .
the ratio @xmath72 is very sensitive to the value of @xmath73 and @xmath6 , it increases monotonically with the increasing @xmath73 and @xmath6 .
this implies that the mc effects are more important for the greater @xmath73 and @xmath6 . for simplicity
, we choose @xmath74 and @xmath6=3 in the calculations , and discuss the influence of their values in sect .
we solve numerically equations ( [ eq12 ] ) , ( [ eq20 ] ) and ( [ eq21 ] ) to find the disk temperature @xmath75 and density @xmath76 versus the disk radius with the given @xmath73 , @xmath6 and @xmath77 ( where @xmath77 is the accretion rate in units of @xmath78 .
we take @xmath79 for the fully photodisintegrated nuclear , which is appropriate in the inner disk . in the calculation
, we do not include the cooling term arising from the photodisintegration , because it is much less than the neutrino cooling rate in the inner disk ( janiuk et al .
furthermore , @xmath80 , @xmath81 and @xmath82 are adopted in calculations .
as discussed in sect .
2 , the mc process applies a strong torque on the disk , resulting in huge viscous dissipation .
this would lead to a more powerful neutrino radiation , and neutrino annihilation luminosity . here
, we will show the effects of mc torque on the neutrino annihilation luminosity . to show this
, we compare the results of mcndaf with the ndaf model without mc ( hereafter ndaf refers to the model without mc ) .
gll06 pointed out that the gr effects and the neutrino radiation from the optically thick region are important for the ndaf luminosity .
therefore we include these two effects in our calculations for both mcndaf and ndaf .
the neutrino luminosity from the accretion flow is @xmath83 we are interested primarily in the properties of the inner accretion flow , where neutrino processes are important . as argued in pwf99 , npk01 and dpn02 , the flows are fully advection - dominated for @xmath84 , since neutrino cooling is not important and photons are completely trapped .
thus we concentrate the discussion in the region from @xmath5 to @xmath85 .
our method for calculating neutrino annihilation is similar to pwf99 and rosswog et al .
the disk is modeled as a grid of cells in the equatorial plane .
a cell @xmath86 has its neutrino mean energy @xmath87 and luminosity @xmath88 , and the height above ( or below ) the disk is @xmath89 .
the angle at which neutrinos from cell @xmath86 encounter antineutrinos from another cell @xmath90 at that point is denoted as @xmath91 .
then the neutrino annihilation luminosity at that point is given by the summation over all pairs of cells , @xmath92 where @xmath93 and @xmath94 .
the total neutrino annihilation luminosity is obtained by integrating over the whole space outside the bh and the disk , @xmath95 as shown in figure 2 , the variations of @xmath96 and @xmath97 versus @xmath77 for mcndaf are indicated by the thin and thick solid lines , respectively , while those for ndaf are marked by the dotted and dashed lines , respectively .
it is found that @xmath96 and @xmath98 are greatly strengthened in mcndaf .
this means the spin energy acts as a powerful source for ndaf .
according to our calculations for mcndaf @xmath98 varies from @xmath99 to @xmath100 for @xmath101 .
we find that @xmath98 nearly stays constant around @xmath102 for the accretion rate above @xmath103 .
this implies that the effect of neutrino optical depth becomes important .
our results for ndaf are in good agreement with those given by pwf99 and gll06 , but larger than those in dpn02 .
this is because the gr effects are taken into account in this paper as well as in pwf99 and gll06 .
dpn02 discussed the thermal , viscous and gravitational stability properties of ndaf solutions .
they found that ndaf is stable in most cases . but this result is not consistent with the variability in grb lightcurve .
to explain the x - ray flares , it is need that after the prompt gamma - ray emission has ceased , the central engine can be restarted ( fan & wei 2005 ; zhang et al .
based on this scenario , perna et al .
( 2006 ) suggested that the x - ray flares could be produced by accretion of matter after the breaking of the disk due to the setting up of various instabilities either gravitational or viscous .
therefore , it is attractive for us to examine whether mcndaf solution is stable .
the condition for viscous stability is given by @xmath104 the stability curves for several radii in the disk are shown in figure 3 . from figure 3 , we find the @xmath105 curves show s - shape , in which the branch of solutions with negative slop is viscously unstable . it is shown that the viscous instability occurs at larger radius for larger accretion rate .
however , we find this unstable can only occur at @xmath106 , see figure 5 .
it is clearly shown in figure 3 that the disk is unstable at @xmath107 for @xmath108 , @xmath109 for @xmath110 , and @xmath111 for @xmath112 .
to understood this s - shape we draw figure 4 .
inspecting figures 3 and 4 , we find the viscous instability occurs when the disk is neutrino cooing and radiation pressure dominated .
for @xmath113 , the mc torque will become very small , and the disk is optically thin to neutrinos . as discussed in npk01 , an optically thin ndaf is viscously stable for all pressure cases
. therefore , we will find no viscous instability when @xmath114 . if the accretion rate beyond @xmath115 , the mc torque transport enormous energy into the inner disk , and make the inner disk optically thick to neutrinos . in this time , we have @xmath116 for neutrino cooling and radiation pressure dominated case , and the disk will be viscously unstable . if gas pressure dominates , we have @xmath117 and @xmath118 for optically thin and thick ndaf , respectively . in the region where advection dominated ,
the disk is viscous stable , which is the well - known property of slim disk . from the above analysis
, we conclude that the appearance of the viscous instability is due to the mc torque .
first , the mc torque results in an optically thick neutrino - cooling dominated flow in the inner disk part for high accretion rates .
second , the high opacity will drop the neutrino - cooling rate , and leave a hot and thick disk . in this region ,
the gas pressure drops and radiation pressure becomes important .
we checked that the unstable solutions appear for black hole spin @xmath119 and magnetic pressure @xmath120 .
the disk is thermally unstable if @xmath121 .
then any small increase ( decrease ) in temperature leads to heating rate which is more ( less ) than the cooling rate , and as a consequence a further increase ( decrease ) of the temperate .
for an optically thick ndaf in which neutrino cooling and radiation pressure dominates , we have @xmath122 and @xmath123 .
therefore , the disk is also thermally unstable at the negative slope of the s - shape @xmath105 curves shown in figure 3 .
it is noticed that janiuk et al .
( 2007 ) obtain the viscous instability occurring at @xmath124 without the s - shape curves , which is caused by the behavior distribution for @xmath43 and photodisintegration term .
however , in mcndaf , the photodisintegration term is not included and @xmath79 is assumed for simplicity .
therefore , we infer that the s - shape curves in mcndaf arise from the mc effects . finally , we check the gravitational stability condition , for which the toomre parameter @xmath125 should be larger than unity . for keplerian disk ,
@xmath125 is given by @xmath126 .
@xmath125 decrease with increasing @xmath127 so that the flow is most unstable on the outside .
in this paper we investigate some properties in mcndaf . the angular momentum deposited in the disk by the magnetic torque exerted by the bh leads to a substantial additional dissipation of energy in the disk , which is greater than that expected from gravitational release alone .
we obtain a series of mcndaf solutions being different significantly from ndaf . \1 .
the neutrino annihilation luminosity in mcndaf varies from @xmath99 to @xmath128 for @xmath101 , while for ndaf the value range is from @xmath129 to @xmath130 , i.e. , it is greatly improved by the mc torque .
recently , observations show that half of the swift bursts exhibit x - ray flares .
fan et al .
( 2005 ) pointed out that the energy from ndaf can not match the x - ray flares detected in grb 050724 of @xmath131100 s , which is also the time scale of the central engine .
the time averaged isotropic luminosity of the x - ray flare component is @xmath132 . if we assume that the total mass available for accretion is @xmath133 ( a typical value for the compact object merger scenarios and massive star collapse scenario ) , and that most of the mass is accreted during the x - ray flare phase , the time averaged accretion rate is about @xmath134 . at this accretion rate ,
the jet luminosity powered by neutrino annihilation is @xmath135 for ndaf without mc torque , and it is insufficient to power the x - ray flares .
but the power produce by mcndaf can satisfy this requirement .
very recently , the new observation of the highest redshift ( z=6.7 ) swift source grb080913 puts a very strong constraint on the central engine ( perez - ramirez et al .
the duration of this short burst is @xmath136 , and the isotropic energy required is @xmath137 .
if the central engine is ndaf without mc torque , the jet collimation factor obeys @xmath138 , i.e. , the jet should be strongly collimated .
if invoke the mc torque in ndaf , the observed energy can be easily satisfied .
the disk becomes thermally and viscously unstable in its inner region for @xmath106 .
it is very interesting for us to obtain the @xmath105 curves behaving s - shape , which may produce a limit - cycle activity .
the disk is rather thick in the inner region , and therefore the thermal and viscous timescales are close to each other .
the timescale for the unstable can be estimated by the viscous timescale,@xmath139(r / h)^2 $ ] , which is about 10ms at the inner disk . following the discussions in janiuk et al .
( 2007 ) , these instabilities will lead to a variable energy output on millisecond timescales , which may correspond to the variability in the gamma - ray luminosity .
the irregularity in the overall outflow can also help produce internal shocks .
on the other hand , the thermal - viscous instability may be accompanied by the disk breaking , which can lead to the several episodic accretion events and explain the long - time activity accounting for the x - ray flares .
therefore , the mcndaf can easily power both grbs and its x - ray flares , naturally interpret the variability in the gamma - ray luminosity , and explain the production of x - ray flares .
however , there are several issues should be addressed .
first , in all of the mcndaf solutions , we assume @xmath6=3 and a large bh spin @xmath74 but does not give any reason . from figure 1
we find the mc contribution is sensitive to @xmath73 and @xmath6 .
thus for very small bh spin and value of @xmath6 , the mc effects may be ignored , and the solutions return to ndaf solutions . considering that the bh is spun - down in the mc process , while it is spun - up in the accretion process .
the two processes with opposite effects result in a state with an equilibrium spin @xmath140 .
calculation shows that @xmath140 is greater than 0.85 for @xmath141 .
this result implies that the mc effects are dominant in the whole duration of grb , for which a fast - spinning bh is the central engine .
secondly , we made many simplifications in the mcndaf model , such as we omit the photodisintegration term in the cooling rates , we assume the disk is thin , and a very simple magnetic configuration , and so on .
recently , liu et al . ( 2007 ) took into account more realistic microphysics .
chen & beloborodov ( 2006 ) worked out the ndaf solution under full kerr metric .
it is necessary to combine these effects with the mc process and work out a more detailed model in the future .
recently , zhang & dai ( 2007 ) proposed a hyperaccretion disk around a neutron star .
they found , compared with a bh disk , the hyperaccretion disk around a neutron star can be cooled more efficiently and produce a much higher neutrino luminosity .
as discussed by kluzniak & rappaport ( 2007 ) , the magnetic dipole of the neutron star can also torque the disk .
therefore , it is also attractive to consider the effects of the magnetic torque in the context of hyperaccretion disk around a neutron star .
we thank t. liu for helpful discussions , and also thank the anonymous referee for his / her valuable comments and constructive suggestions " .
this work is supported by national natural science foundation of china under grants 10873005 , 10847127 and 10703002 , the research fund for the doctoral program of higher education under grant 200804870050 and national basic research program of china under grant no .
2009cb824800 .
balbus s. a. , & hawley j. f. , 1991 , apj , 376 , 214 blandford , r. d. 1976 , mnras , 176 , 465 blandford , r. d. 1999 , in asp conf . ser .
160 , astrophysical discs : an ec summer school , ed .
j. a. sellwood & j. goodman ( san francisco : asp ) , 265 blandford , r. d. , & payne , d. g. 1982 , mnras , 199 , 883 blandford , r. d. , & znajek , r. l. 1977 , mnras , 179 , 433 chen , w. x. , & beloborodov , a. m. 2007 , apj , 657 , 383 di matteo , t. , perna , r. , & narayan , r. 2002 , apj , 579 , 706 ( dpn02 ) fan y. z. , & wei d. m. , 2005 , mnras , 364 , l42 fan , y. z. , zhang , b. & proga , d. 2005 , apj , 635 , l129 gu , w. m. , liu , t. , & lu , j. f. , 2006 , apj , 643 , l87 ( gll06 ) janiuk , a. , perna , r. , di matteo , t. , & czerny , b. 2004 , mnras , 355 , 950 januik , a. , yuan , y. , perna , r. , & di matteo , t. 2007 , apj , 664 , 1011 kluzniak , w. , & rappaport , s. 2007 , apj , 671 , 1990 kohri , k. , & mineshige , s. 2002 , apj , 577 , 311 kohri , k. , narayan , r. , & piran , t. 2005 , apj , 629 , 341 lai , d. 1998 , apj , 502 , 721 lee , u. 1999 , apj , 511 , 359 lee , h. k. , wijers , r. a. m. j. , & brown , g. e. 2000 , phys . rep . , 325 , 83 li , l. x. , & paczynski , b. 2000 , apj , 534 , l197 li , l. x. 2002 , apj , 567 , 463 liu , t. , gu , w. m. , xue , l. , & lu , j. f. 2007 , apj , 661 , 1025 ma , r. y. , yuan , f. , & wang , d. x. 2007 , apj , 671 , 1981 macdonald , d. , & thorne , k. s. 1982 , mnras , 198 , 345 mckinney , j. c. 2005 , apj , 630 , l5 mizuno y. et al .
2004 , apj , 606 , 395 narayan , r. , piran , t. , & kumar , p. 2001
, apj , 557 , 949 ( npk01 ) novikov , i. d. , & thorne , k. s. 1973 , in black holes , les astres occlus , ed . b. & c. dewitt ( new york : gordon & breach ) , 343 perez - ramirez , d. , de ugarte postigo , a. , & gorosabel , j. , et al .
2008 , arxiv:0810.2107 popham , r. , & narayan , r. 1995 , apj , 442 , 337 popham , r. , woosley , s. e. , & fryer , c. 1999 , apj , 518 , 356 ( pwf99 ) pudritz , r. e. , & fahlman , g. g. 1982 , mnras , 198 , 689 qian , y. z. , & woosley , s. e. 1996 , apj , 471 , 331 reynoso , m. m. , romero , g. e. , & sampayo , o. a. 2006 , a&a , 454 , 11 riffert , h. , & herold , h. 1995 , apj , 450 , 508 rosswog , s. , ramirez - ruiz , e. , & davies , m. 2003 , mnras , 345 , 1077 sharkura , n. i. , & sunyaev , r. a. 1973 , a&a , 24 , 337 shibata m. et al .
, 2006 , prl , 96 , 031102 shibata , m. , sekiguchi , y. , & takahashi , r. 2007 , progress of theoretical physics , 118 , 257 van putten , m. h. p. m. 1999 , science , 284 , 115 wang , d. x. , xiao , k. , & lei , w. h. 2002 , mnras , 335 , 655 wang , d. x. , lei , w. h. , & ma , r. y. 2003 , mnras , 342 , 851 xie , y. , huang , c. y. , & lei , w. h. 2007 , chjaa , 7 , 685 ye , y. c. , wang , d. x. , & ma , r. y. 2007 , new a , 12 , 471 zhang b. , fan y. z. , dyks j. et al .
, 2006 , apj , 642 , 354 zhang , d. , & dai , z. g. 2007 , apj , 683 , 329 | recent observations and theoretical work on gamma - ray bursts ( grbs ) favor the central engine model of a kerr black hole ( bh ) surrounded by a magnetized neutrino - dominated accretion flow ( ndaf ) .
the magnetic coupling between the bh and disk through a large - scale closed magnetic field exerts a torque on the disk , and transports the rotational energy from the bh to the disk .
we investigate the properties of the ndaf with this magnetic torque . for a rapid spinning bh , the magnetic torque transfers enormous rotational energy from bh into the inner disk .
there are two consequences : ( i ) the luminosity of neutrino annihilation is greatly augmented ; ( ii ) the disk becomes thermally and viscously unstable in the inner region , and behaves s - shape of the surface density versus accretion rate .
it turns out that magnetically torqued ndaf can be invoked to interpret the variability of gamma - ray luminosity .
in addition , we discuss the possibility of restarting the central engine to produce the x - ray flares with required energy . |
the temperature and density effects to the properties of hadrons are interesting and important for the early universe and for high energy physics .
so far the temperature effect has been extensively studied @xcite .
but there are not so many results concerning the density effect .
as is well known , there are difficulties with the simulation of a finite density system by the lattice qcd approach .
the reason is that the fermionic determinant becomes complex and can not be considered as a part of the probability .
if we factor out the phase of the determinant and put it into observables , it leads to oscillating contributions in quantum averages .
this makes it hard to obtain reliable results .
it has been also known that the naive quenched approximation leads to an essentially different world @xcite . in spite of this difficult situation ,
density effects to hadrons such as a mass - shift are very interesting and important subjects both theoretically and experimentally @xcite .
there are several approaches to circumvent this difficulty and they seem successful to a limited extent @xcite . another way is to study two - color qcd @xcite .
also , a non - trivial quenched approximation can be defined @xcite by taking the simultaneous limit where the quark mass and the logarithm of the chemical potential both become infinite @xcite . as for the response to the chemical potential
, only the baryon number susceptibility at zero baryon density has been studied , and an abrupt jump at the transition point has been reported @xcite . in this paper , we examine the response to the chemical potential of the mass of hadrons at finite temperature and at zero baryon density in full su(3 ) qcd simulations@xcite .
as stated in the introduction , the lattice study at finite density is still difficult . in this situation ,
an interesting possibility is to examine the response of physical quantities to the chemical potential at zero chemical potential .
here we consider a hadronic correlator @xmath3 , @xmath4 where @xmath5 denotes the hadron operator .
suppose that this hadronic correlator is dominated by a single pole , then @xmath6 we take derivative with respect to chemical potential @xmath7 , @xmath8 in eq.([deri-1 ] ) , if the left hand side is measured as a function of @xmath9 , the linear term gives the chemical potential response of the hadron mass , while the constant term gives the response of the coupling .
the next problem is to get the derivative of the correlator . for this purpose , we go back to the definition of the hadronic correlator . @xmath10 \mbox{tr}\,(m(n;0)\gamma m(0;n)\gamma^{\dagger})\nonumber \\ & & \mbox{det}(d)\exp(-s_g ) \label{deri0}\end{aligned}\ ] ] where @xmath11 \mbox{det}(d)\exp(-s_g)$ ] and @xmath12 .
then , using the formulae @xmath13 and @xmath14 we get @xmath15 where the short hand notations @xmath16 and @xmath17 \mbox{det}(d)\exp(-s_g ) y\end{aligned}\ ] ] are used . here
we calculate the first and second terms of the right hand side of eq.([deri2 ] ) and ignore the other terms .
@xmath18 is pure imaginary and its average value , i.e. @xmath19 will be zero@xcite . if @xmath20 is real , the third term does not contribute as real and we may neglect it . for degenerate quarks
, @xmath20 is real .
although we consider here non - degenerate systems for which @xmath20 is not real , we assume that the contribution from the third term is small and drop it in this study .
we introduce two independent chemical potentials , @xmath21 and @xmath22 for two different quarks ( @xmath23 ) .
we introduce convenient definitions of the chemical potential , @xmath24 and @xmath25 and define the following derivatives with respect to @xmath26 and @xmath27 .
@xmath28 @xmath29 @xmath26 is usual chemical potential corresponding to baryon number . in this study
we consider the response with respect to @xmath26 . in the case of the staggered fermion formalism , the derivative of the fermion operator is @xmath30 . \label{deri3}\end{aligned}\ ] ] at @xmath31 , we can evaluate eqs .
( [ deri0 ] ) and ( [ deri2 ] ) by usual techniques of lattice qcd simulation .
here we present our preliminary result for @xmath2 .
lattice parameters are as follows : * * lattice size * : + @xmath32 , with 16 for @xmath9 direction , 4 for the temperature direction . *
* quarks * : + @xmath33=0.25 and @xmath34=0.025 for @xmath35=5.26 , and @xmath33=0.25 and @xmath34=0.0125 for @xmath35=5.33 .
dynamical k - s fermions with @xmath1 are the light quarks ( @xmath34 ) .
the meson operator , @xmath36 , is made of the two different quarks .
* * temperature * : + @xmath37 for @xmath38 ( confinement ) , and + @xmath39 for @xmath40 ( deconfinement ) . *
* statistics * : + 40 - 90 configurations for each parameter . the results for the @xmath42 in eq.([deri-1a ] ) are shown in fig.1 and fig.2 .
fig.1 is for @xmath37 which corresponds to the confinement phase and fig.2 is for @xmath39 which corresponds to the deconfinement phase . in the confinement phase ( @xmath37 , @xmath38 ) , where @xmath3 is the hadron propagator along the @xmath9 direction . @xmath43 and @xmath44 as is stated above , the coefficient of the linear term in @xmath42 corresponds to @xmath45 .
although this is preliminary and the statistics are small , we can see that @xmath2 is negative in the confinement phase and @xmath2 is positive in the deconfinement phase . in the deconfinement phase ( @xmath39 , @xmath40 ) , where @xmath3 is the hadron propagator along the @xmath9 direction .
@xmath43 and @xmath46
one of our masses , @xmath33 , is in the region of the strange quark . if the behaviors observed in figs .
1 and 2 are also seen in the charm quark mass region , then the mass of the heavy - light meson , @xmath48 , might decrease at very small chemical potential just below the deconfinement temperature . then the mass of @xmath49 system might be lower than the mass of the @xmath50 meson , and a new decay mode @xmath51 will open .
if this happens in , for example , ultra - relativistic heavy ion collisions , @xmath50 mesons produced by this collision will decay as @xmath52 and eventually the amount of @xmath47 produced through @xmath53 channel will decrease .
this means that @xmath47 suppression might occur even in the confinement phase as suggested by hayashigaki@xcite .
the results of this paper are very preliminary and qualitative and should be checked in the near future . however
a model calculation based on njl model@xcite suggests the same tendency .
we thank i. o. stamatescu for a critical reading of the manuscript . this work is partially supported by the grant - in - aide for scientific research by jsps and monbusho , japan ( no .
10640272 , no . 11440080 , no .
11694085 , no . 11740159 and no .
12554008 ) .
99 see for example : the proceedings of `` quark matter '' and `` lattice field theory '' conferences .
stephanov , _ phys.rev.lett . _ * 76 * 4472 ( 1996 ) . t. hatsuda and s.h .
lee , _ phys .
rev . _ * c46 * 34 ( 1992 ) .
g. agakishiev et al . , ceres collaboration , nucl .
a638 * 159 ( 1998 ) .
barbour , s.e .
morrison , e.g. klepfish , j.b .
kogut and m. lombardo , _ nucl.phys . _ * b(proc.suppl.)60a * 220 ( 1998 ) .
a. nakamura , phys .
, 149b ( 1984 ) 391 .
lombardo , hep - lat/9907025 , hep - lat/9906006 .
s. hands , j. b. kogut , m - p .
lombardo and s. e. morrison , nucl .
b558 ( 1999 ) 327 .
s. morrison and s. hands , hep - lat/9902012 .
s. hands et al .
, hep - lat/0006018 . s. j. hands , j. b. kogut , s. e. morrison , d. k. sinclair , hep - lat/0010028 .
j. engels , o. kaczemarek , f. karsch , e. laermann , nucl .
b558 ( 1999 ) 307 .
i.bender , t.hashimoto , f.karsch , v.linke , a.nakamura , m.plewnia , i .- o.stamatescu and w.wetzel , _ nucl.phys . _ * b(proc.suppl.)26 * 323 ( 1992 ) ; t.c.blum , j.e.hetrick and d.toussaint , phys
* 76 * 1019 ( 1996 ) .
s. gottlieb et al . , phys . rev .
* d55 * 6852 ( 1997 ) see also ph .
de forcrand et al . , qcdtaro - collaboration , + _ nucl.phys . _
* b(proc.suppl.)73 * 477 ( 1999 ) .
a. hayashigaki , `` new directions in quantum chromodynamics '' , ed . by c - r .
ji and d - p .
min , aip conference proceedings 494 , melville , new york , 1999 , p429 .
see t. hatsuda and t. kunihiro , phys .
* c247 * 221 ( 1994 ) . | we study the response of meson masses to the chemical potential ( @xmath0 ) at high temperature and at zero chemical potential on @xmath1 lattice with staggered fermions .
preliminary results for the meson composed of different quarks show that @xmath2 is negative in the confinement phase and positive in the deconfinement phase . |
the three indications in favor of neutrino oscillations found in solar neutrino experiments @xcite , in atmospheric neutrino experiments @xcite and in the accelerator lsnd experiment @xcite require the existence of three independent scales of neutrino mass - squared differences ( see @xcite ) : @xmath0 ( msw ) or @xmath1 ( vo ) @xcite , @xmath2 @xcite , @xmath3 @xcite .
hence , at least four massive neutrinos must exist in nature .
considering the minimal possibility of four massive neutrinos @xcite , we have the mixing relation @xmath4 that connects the flavor neutrino fields @xmath5 ( @xmath6 ) with the fields @xmath7 ( @xmath8 ) of neutrinos with masses @xmath9 .
@xmath10 is the @xmath11 unitary mixing matrix .
the field @xmath12 describes sterile neutrinos , that do not participate to weak interactions and do not contribute to the invisible decay width of the @xmath13-boson , whose measurement have shown that the number of light active neutrino flavors is three ( see @xcite ) , corresponding to @xmath14 , @xmath15 and @xmath16 .
the purpose of this paper is to present the implications of the results of neutrino oscillation experiments for the mixing of four light massive neutrinos .
we will discuss also some implications for long - baseline neutrino oscillation experiments @xcite .
the plan of the paper is as follows . in section [ the mixing schemes ]
we introduce the two four - neutrino mixing schemes that are allowed by the data and we review some of their properties . in section
[ constraints from oscillation experiments ] we derive the constraints on four - neutrino mixing obtained from the results of neutrino oscillation experiments . in sections [ long - baseline disappearance experiments ] and [ long - baseline appearance experiments ]
we discuss some implications for long - baseline disappearance and appearance experiments , respectively . in section [ conclusions ]
we summarize our achievements .
in the discussion of long - baseline experiments we will neglect matter effects , that may be relevant for long - baseline accelerator experiments whose beams propagate for hundreds of kilometers in the crust of the earth , but are rather complicated to study in the framework of four - neutrino mixing ( see @xcite ) and will be treated in detail in another article @xcite .
it has been proved in @xcite that among all the possible four - neutrino mixing schemes only two can accommodate the results of all neutrino oscillation experiments : @xmath17 these two spectra are characterized by the presence of two pairs of close masses separated by a gap of about 1 ev which provides the mass - squared difference with @xmath18 . ]
@xmath19 responsible for the short - baseline ( sbl ) oscillations observed in the lsnd experiment . in the scheme
a , we have @xmath20 and @xmath21 , whereas in scheme b , @xmath22 and @xmath23 . in order to treat simultaneously the two schemes
a and b , let us rename the mass eigenstates indices as @xmath24 with this convention , in both schemes @xmath25 and @xmath26 are the two massive neutrinos whose mass - squared difference is responsible for solar neutrino oscillations and @xmath27 and @xmath28 are the two massive neutrinos whose mass - squared difference generates atmospheric neutrino oscillations : @xmath29 for @xmath30 in the following we will use the range @xmath31 that is allowed by the 99% cl maximum likelihood analysis of the lsnd data in terms of two - neutrino oscillations @xcite and by the 90% cl exclusion curves of the bugey reactor @xmath32 disappearance experiment @xcite and of the bnl e776 @xcite and karmen @xcite experiments ( see also @xcite ) . the mixing in both schemes a and b is constrained by the results of neutrino oscillation experiments @xcite . in order to quantify these constraints ,
let us define the quantities @xmath33 , with @xmath6 , as @xmath34 physically @xmath33 quantify the mixing of the flavor neutrino @xmath35 with the two massive neutrinos whose mass - squared difference is relevant for the oscillations of atmospheric neutrinos ( @xmath36 , @xmath37 in scheme a and @xmath38 , @xmath39 in scheme b ) .
let us write down the expressions , that we will need in the following , for the oscillation probabilities in short baseline experiments and in atmospheric and long - baseline experiments .
the neutrino oscillation probabilities in short - baseline ( sbl ) experiments are given by @xcite @xmath40[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } = \frac{1}{2 } \ , a_{\alpha\beta } \left ( 1 - \cos\frac { \delta{m}^{2}_{\mathrm{sbl } } l } { 2 e } \right ) \qquad ( \beta\neq\alpha ) \ , , \label{pab - sbl } \\ & & p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } = 1 - \frac{1}{2 } \
, b_{\alpha\alpha } \left ( 1 - \cos\frac { \delta{m}^{2}_{\mathrm{sbl } } l } { 2 e } \right ) \ , , \label{paa - sbl}\end{aligned}\ ] ] where @xmath41 is the source - detector distance , @xmath42 is the neutrino energy and the oscillation amplitudes are given by @xmath43 the two expressions for @xmath44 obtained by summing over @xmath45 or @xmath46 are equivalent because of the unitarity of the mixing matrix .
the oscillation probabilities of neutrinos and antineutrinos in short - baseline experiments are equal .
since the expressions ( [ pab - sbl ] ) and ( [ paa - sbl ] ) have the same form as the usual probabilities in the case of two - neutrino mixing , the oscillation amplitudes @xmath44 and @xmath47 are equivalent to the usual parameter @xmath48 used in the analysis of the experimental results ( @xmath49 is the two - neutrino mixing angle ) .
the average probabilities measured by short - baseline experiments that are sensitive to values of @xmath50 much smaller than the actual value of @xmath51 are given by @xmath52[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle = \frac{1}{2 } \ , a_{\alpha\beta } \quad ( \alpha\neq\beta ) \ , , \qquad \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \rangle = 1 - \frac{1}{2 } \ , b_{\alpha\alpha } \ , .
\label{lbl08}\ ] ] the probability of @xmath53 transitions in atmospheric and long - baseline experiments is given by @xcite @xmath54 the average oscillation probability measured in experiments sensitive to values of @xmath55 smaller than the actual value of @xmath56 is given by @xmath57[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle = \left| \sum_{k={{a}},{{b } } } u_{{\alpha}k}^ { * } u_{{\beta}k } \right|^2 + \left ( \sum_{k={{c}},{{d } } } \right ) \
, , \label{lbl05}\ ] ] for both neutrinos and antineutrinos .
the negative results of short - baseline @xmath32 and @xmath15 disappearance experiments and the results of solar neutrino experiments imply that @xmath58 and @xmath59 are constrained by @xcite @xmath60 the quantities @xmath61 and @xmath62 , that depend on @xmath30 , are given by @xmath63 where @xmath64 is the experimental upper bound , that depends on @xmath30 , for the amplitude @xmath47 , obtained from the exclusion plots of short - baseline @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } $ ] disappearance experiments ( see @xcite ) . from the 90% cl exclusion curves of the bugey reactor @xmath32 disappearance experiment @xcite and of the cdhs @xcite accelerator @xmath15 disappearance experiment
it follows that @xmath66 for @xmath30 in the lsnd range ( [ lsnd - range ] ) and @xmath67 for @xmath68 ( see @xcite ) .
furthermore , it is clear that the observed disappearance of atmospheric @xmath15 s implies that @xmath59 , that represents the mixing of @xmath15 with the two massive neutrinos responsible for atmospheric neutrino oscillations , can not be small .
indeed , it has been shown in @xcite that the up - down asymmetry of high - energy @xmath69-like events generated by atmospheric neutrinos measured in the super - kamiokande experiment ( @xmath70 @xcite ) implies the bound @xmath71 hence , @xmath59 is large also for @xmath72 , that is below the sensitivity of the short - baseline @xmath15 disappearance experiments performed so far . the constraints in eqs .
( [ c - bounds ] ) and ( [ ask - old ] ) show that @xmath58 and @xmath73 are small ( in both schemes a and b ) .
the smallness of @xmath58 implies that the oscillations of @xmath14 s and @xmath32 s in atmospheric and long - baseline experiments are suppressed .
indeed , rather strong bounds on the transition probability @xmath74[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } $ ] of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } $ ] s in long - baseline experiments ( lbl ) have been derived in @xcite .
using the fact that the oscillations of electron neutrinos are negligible in atmospheric experiments and the fact that the super - kamiokande up - down asymmetry has been measured through an average over the neutrino energy spectrum and propagation distance , we can derive a lower bound for @xmath59 stronger than the one in eq .
( [ ask - old ] ) .
we will use the updated @xcite value of the super - kamiokande up - down asymmetry , @xmath75 where @xmath76 and @xmath10 are the numbers of down - going and up - going @xmath69-like events , respectively . summing the statistic and systematic errors in quadrature
, we obtain @xmath77 since the asymmetry is due to the disappearance of @xmath15 s into @xmath16 s or @xmath78 s , we have @xmath79[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle - \langle p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle } { \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle + \langle p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle } \ , , \label{asy01}\ ] ] where @xmath80[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle $ ] is the average survival probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s in short - baseline experiments , that is the same as the average survival probability of downward - going atmospheric @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s , and @xmath81[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle $ ] is the average survival probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s in long - baseline experiments , that is the same as the average survival probability of upward - going atmospheric @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s . from eqs .
( [ lbl08 ] ) , ( [ lbl05 ] ) and from the definition ( [ sigma ] ) , we have @xmath82[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle = 1 - 2 \sigma_\mu ( 1 - \sigma_\mu ) \ , , \label{asy02 } \\ & & \langle p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle = \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle - 2 |u_{\mu{{d}}}|^2 ( \sigma_\mu - |u_{\mu{{d}}}|^2 ) \ , .
\label{asy03}\end{aligned}\ ] ] taking into account the obvious constraint @xmath83 , the long - baseline average survival probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s is bounded by @xmath52[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle - \frac{\sigma_\mu^2}{2 } \leq \langle p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \leq \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \ , . \label{asy04}\ ] ] using the lower bound in eq .
( [ asy04 ] ) , for the asymmetry ( [ asy01 ] ) we have the upper bound @xmath84 inverting this inequality , we get @xmath85 this is the lower bound for @xmath59 given by the asymmetry ( [ asy01 ] ) .
thus , in the framework of the four - neutrino schemes under consideration the asymmetry @xmath86 can not be larger than 1/3 , a condition that is satisfied by the measured asymmetry ( [ asy - sk ] ) . from eqs .
( [ asy07 ] ) and ( [ asy06 ] ) we obtain @xmath87 this is the new upper bound for @xmath88 , that is more stringent than the one in eq .
( [ ask - old ] ) .
furthermore , using the expression ( [ asy03 ] ) for @xmath81[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle $ ] , we can derive the range of @xmath89 that is allowed by the lower bound ( [ asy07 ] ) for @xmath86 ( if @xmath89 were allowed to vary in the whole range @xmath90 $ ] , the asymmetry could vanish ) . after a straightforward calculation we obtain that @xmath89 is bounded in the range @xmath91 with @xmath92 the allowed range for @xmath89 as a function of @xmath59 obtained from the lower bound ( [ asy07 ] ) for the asymmetry is shown by the shadowed area in fig .
[ umu - lim ] . since @xmath59 can be close to one , we have the allowed range @xmath93 the same allowed range obviously applies also to @xmath94 , with the constraint ( [ sigma ] ) ( for example , from fig .
[ umu - lim ] one can see that @xmath95 is possible only if @xmath96 and the corresponding value of @xmath94 is 0.59 , well within the range ( [ asy11 ] ) ) .
up to now we have derived the upper bounds for @xmath58 and @xmath73 implied by the results of neutrino oscillation experiments .
however , @xmath58 and @xmath73 , albeit small , can not vanish , because non - zero values of both @xmath58 and @xmath73 are required in order to generate the @xmath97 and @xmath98 transitions observed in the lsnd experiment @xcite .
indeed , we are going to derive lower bounds for @xmath58 and @xmath73 from the results of the lsnd experiment .
let us consider the amplitude @xmath99 of @xmath97 and @xmath98 oscillations in the lsnd experiment .
from eq .
( [ sbl - amp ] ) , using the cauchy
schwarz inequality , we obtain two inequalities : @xmath100 it is clear that both @xmath58 and @xmath73 must be bigger than zero and smaller than one in order to have a non - vanishing amplitude @xmath99 in the lsnd experiment . from the inequalities ( [ i1 ] ) and ( [ i2 ] ) we obtain
the upper bound @xmath101 already derived in @xcite , and the lower bounds @xmath102 with @xmath103 where @xmath104 is the minimum value of the amplitude @xmath99 allowed by the results of the lsnd experiment and the approximation is valid for small @xmath104 , as follows from the lsnd data .
obviously , the minimum ( [ lambda- ] ) depends on @xmath30 trough the dependence from @xmath30 of @xmath104 . summarizing , from the results of all neutrino oscillation experiments we have the bounds @xmath105 with @xmath106 given by eq.([lambda- ] ) and @xmath107 \ , .
\label{lambda+}\ ] ] in eq .
( [ bounds ] ) we have emphasized that the bounds for @xmath58 and @xmath59 depend on the value of @xmath30 , that must lie in the lsnd range ( [ lsnd - range ] ) .
the dashed line in fig . [ ce ] represents the lower bound ( [ c - lb ] ) for @xmath58 obtained from the 99% cl maximum likelihood allowed region of the lsnd experiment @xcite . the solid line in fig .
[ ce ] is the upper bound ( [ c - bounds ] ) for @xmath58 obtained from the 90% cl exclusion curve of the bugey reactor @xmath32 disappearance experiment @xcite .
the dash - dotted line in fig . [ ce ] shows the upper bound for @xmath58 obtained from the negative result of the chooz long - baseline reactor @xmath32 disappearance experiment @xcite .
this bound has been obtained taking into account that the survival probability of @xmath32 s in long - baseline experiments is bounded by @xcite @xmath108 the chooz experiment has measured a survival probability averaged over the energy spectrum @xcite @xmath109 where we added in quadrature the independent statistical and systematic errors . assuming a gaussian error distribution and taking into account that the survival probability is physically limited in the interval @xmath110 $ ] , we obtain the 90% cl lower bound ) has been obtained using a gaussian distribution normalized in the interval @xmath110 $ ] , according to the bayesian prescription ( see @xcite ) .
the lower bound obtained with the unified approach @xcite is @xmath111 , in reasonable agreement with ( [ peechoozmin ] ) . ]
@xmath112 taking into account the fact that @xmath58 is small , from eqs .
( [ peelbl ] ) and ( [ peechoozmin ] ) we obtain @xmath113 this is the upper bound represented by the dash - dotted line in fig .
one can see that the chooz bound is compatible with the bugey bound ( solid line ) and a more sensitive experiment it is necessary in order to probe the range of @xmath58 allowed by the results of the lsnd experiment ( shadowed area ) .
the bounds ( [ c - bounds ] ) , ( [ ask ] ) , ( [ omcmu - ub ] ) and ( [ c - lb ] ) for @xmath73 are illustrated in fig .
[ oms ] . the shadowed area enclosed by the solid line
is excluded by the 90% cl exclusion curve of the cdhs @xcite short - baseline @xmath15 disappearance experiment using eq .
( [ c - bounds ] ) ( @xmath114 or @xmath115 @xcite ) .
the vertically hatched area limited by the dash - dotted line is excluded by the inequality ( [ ask ] ) obtained the up - down asymmetry of high - energy @xmath69-like events generated by atmospheric neutrinos measured in the super - kamiokande experiment .
the diagonally hatched region is excluded by the results of the lsnd experiment , taking into account also the bugey 90% cl exclusion curve .
the region above the dashed line is excluded by the upper bound ( [ omcmu - ub ] ) and the region below the dotted line is excluded by the lower bound ( [ c - lb ] ) .
taking into account all these constraints , the allowed region for @xmath73 is given by the white area in fig .
from the expression ( [ lbl04 ] ) with @xmath116 ( and the corresponding one for antineutrinos ) it follows that the probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } $ ] transitions in other states in long - baseline experiments is bounded by @xcite @xmath117[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \leq 2 \sigma_\alpha ( 1 - \frac{1}{2 } \ , \sigma_\alpha ) \ , .
\label{lbl01}\ ] ] it is interesting to notice that the lower bound coincides with the average probability of @xmath35 transitions in other states in short - baseline experiments . indeed , from eqs .
( [ paa - sbl ] ) and ( [ sbl - amp ] ) we have @xmath118[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \rangle = \frac{1}{2 } \ , b_{\alpha\alpha } = 2 \sigma_\alpha ( 1 - \sigma_\alpha ) \ , .
\label{lbl02}\ ] ] this is obviously due to the fact that long - baseline @xmath35 disappearance experiments are sensitive also to short - baseline oscillations due to @xmath30 , but since these oscillations are very fast , only their average is observed in long - baseline experiments .
the surprising fact is that the interference terms in the long - baseline disappearance probability are such that the disappearance in long - baseline experiments can not be smaller than the average disappearance in short - baseline experiments .
as we will see later , the analogous propriety for @xmath53 transitions is not true . if we consider @xmath119 in eq .
( [ lbl01 ] ) , taking into account the bound ( [ bounds ] ) for @xmath58 , we have @xmath120[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \leq 2 \lambda^{+}_e \left ( 1 - \frac{1}{2 } \ , \lambda^{+}_e \right ) \simeq 2 a_e^0 \ , , \label{lbl03}\ ] ] with the approximations applying , respectively , for small @xmath104 and small @xmath121 , as realized in practice .
figure [ pee ] shows the values of the bounds ( [ lbl03 ] ) obtained from the existing data . the solid line in fig .
[ pee ] shows the upper bound in ( [ lbl03 ] ) obtained from the exclusion curve of the bugey @xmath32 short - baseline disappearance experiment and the dotted line shows the lower bound in ( [ lbl03 ] ) obtained from the lower edge of the region in the @xmath99@xmath30 plane allowed at 99% cl by the results of the lsnd experiment .
the shadowed region in fig .
[ pee ] is allowed by all data and one can see that it extends below the chooz bound ( dash - dotted line obtained from eq .
( [ peechoozmin ] ) ) .
the lower bound for @xmath122[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } $ ] lies above @xmath123 , a limit that will be very difficult to reach .
however , if @xmath30 lies in the lower part of the lsnd - allowed region , the lower bound @xmath122[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } $ ] is substantially higher and may be reachable in the near future ( see @xcite ) .
considering the disappearance of @xmath15 s in long - baseline experiments , it is clear that , since @xmath59 is large , the corresponding probability can vary almost from zero to one . indeed , using the limits ( [ bounds ] ) for @xmath59 in eq.([lbl01 ] ) we obtain @xmath124[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \leq \left ( 1 - \lambda^{-}_\mu \right ) \left ( 1 + \frac{1}{2 } \ , \lambda^{-}_\mu \right ) \simeq 1 - \frac { a_{{\mu}e}^{\mathrm{min } } } { 8 } \ , .
\label{lbl051}\ ] ] the expression ( [ lbl01 ] ) gives the bounds for the transition probability @xmath122[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } $ ] , that depends on the neutrino energy .
however , if a long baseline experiment is sensitive to values of @xmath55 smaller than the actual value of @xmath56 , its transition probability is given by the average ( [ lbl05 ] ) .
for @xmath116 in eq .
( [ lbl05 ] ) , taking into account the definition of @xmath33 in eq .
( [ sigma ] ) and the obvious constraints @xmath125 , one can derive the bounds @xmath126[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \rangle \leq 2 \sigma_\alpha \left ( 1 - \frac{3}{4 } \
, \sigma_\alpha \right ) \ , .
\label{lbl06}\ ] ] therefore , only the upper bound for the average transition probability is different from the corresponding one for the transition probability , but the difference in practice is negligible for @xmath119 , because @xmath58 is very small .
hence , the bounds in fig .
[ pee ] apply also to @xmath127[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle $ ] .
for the average transition probability of @xmath15 s into other states , a lower bound that is stronger than the lower bound implied by eq .
( [ lbl06 ] ) ( that coincides with the one in eq .
( [ lbl051 ] ) ) is given by the asymmetry ( [ asy01 ] ) , from which we get @xmath57[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle = \frac{1-\mathcal{a}_\mu}{1+\mathcal{a}_\mu } \ , \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \ , .
\label{asy21}\ ] ] taking into account the limit ( [ asy07 ] ) for @xmath86 and approximating the upper bound @xmath80[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \leq 1 - 2 \lambda^{-}_\mu ( 1 - \lambda^{-}_\mu ) $ ] with one , we obtain @xmath128[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \gtrsim 0.37 \ , .
\label{asy22}\ ] ] this is the lower bound for the average transition probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } $ ] s into other states in order to accommodate the up - down asymmetry observed in the super - kamiokande experiment .
we have also an upper bound for this transition probability , that is given by eq .
( [ lbl06 ] ) : @xmath128[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle \leq \left\ { \begin{array}{ll } \left ( 1 - \lambda^{+}_\mu \right ) \left ( 1 + 3 \lambda^{+}_\mu \right ) / 2 & \quad \mbox{for } \quad \lambda^{+}_\mu < 1/3 \ , , \\ 2/3 & \quad \mbox{for } \quad \lambda^{+}_\mu \geq 1/3 \ , , \end{array } \right . \label{lbl061}\ ] ] where we have taken into account the fact that , from eqs .
( [ ask ] ) and ( [ lambda+ ] ) , @xmath129 can be as large as about 0.38 .
it is interesting to notice that there is the possibility that @xmath127[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } \rangle $ ] is larger that @xmath130 . since in the case of two - neutrino mixing
the average transition probability of any flavor neutrino into other states is smaller than 1/2 ( equal in the case of maximal mixing ) , the measurement in long - baseline experiments of an average transition probability of @xmath15 s into other states larger than 1/2 would be an evidence of the existence of a neutrino mass - squared difference larger than the atmospheric one and more than two - neutrino mixing .
equation ( [ lbl061 ] ) shows that this is possible in the framework of four - neutrino mixing , given the existing experimental data .
taking into account the definition ( [ sbl - amp ] ) for @xmath44 , from eq .
( [ lbl04 ] ) it follows that the probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } $ ] transitions with @xmath131 in long - baseline experiments is bounded by @xcite @xmath132[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \leq \frac{1}{4 } \ , a_{\alpha\beta } + \sigma_\alpha \sigma_\beta \ , , \label{lbl07}\ ] ] and by the unitarity upper limits @xmath133[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \leq 1 - p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \ , , \qquad p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \leq 1 - p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \ , .
\label{lbl07-uni}\ ] ] the corresponding allowed range for @xmath134[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } $ ] obtained from the experimental data has been presented in ref .
the other constrained channels are @xmath53 with @xmath135 and @xmath136 , for which we have only the upper limits @xmath133[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \leq 1 - p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } } \qquad ( \alpha = e,\mu \ , \quad \beta=\tau , s ) \ , , \label{lbl071}\ ] ] due to the conservation of probability , with the upper limits for @xmath122[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } $ ] and @xmath122[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } } $ ] given in eqs .
( [ lbl03 ] ) and ( [ lbl051 ] ) , respectively . confronting the interval ( [ lbl07 ] ) with the expression for the average probability of @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } }
\to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } $ ] transitions in short - baseline experiments given in eq . (
[ lbl08 ] ) , one can see that , as anticipated in the discussion after eq .
( [ lbl02 ] ) , the probability of long - baseline @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } $ ] transitions can be smaller than the average probability of short - baseline @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } $ ] transitions .
this is due to interference effects produced by @xmath56 .
hence we have @xmath133[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } < \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle \qquad \mbox{(possible ) } \ , .
\label{lbl09}\ ] ] for a fixed source - detector distance in a long - baseline experiment , the realization of the possible inequality ( [ lbl09 ] ) depends on the value of the neutrino energy .
therefore , using a detector with sufficient energy resolution ( or with a sufficiently narrow - band beam and varying the neutrino energy ) it is possible to check experimentally if the inequality ( [ lbl09 ] ) can be realized .
the inequality ( [ lbl09 ] ) is interesting because it can be realized only if there are at least two mass - squared differences , @xmath30 and @xmath56 .
indeed , it is impossible in the case of two - neutrino mixing that implies , with an obvious notation , @xmath137[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } = \langle p^{\mathrm{sbl},2}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle \qquad \mbox{only $ \delta{m}^2_{\mathrm{sbl}}$ } \
, , \label{lbl11 } \\ & & p^{\mathrm{lbl},2}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \geq \langle p^{\mathrm{sbl},2}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle = 0 \qquad \mbox{only $ \delta{m}^2_{\mathrm{atm}}$ } \ , .
\label{lbl12}\end{aligned}\ ] ] however , the possibility of the inequality ( [ lbl09 ] ) is not unique to the four - neutrino schemes , and it can be realized , for example , in three - neutrino schemes with @xmath138 and @xmath139 . the experimental check of the inequality ( [ lbl09 ] ) could provide some information on the relative phases of the elements of the mixing matrix .
indeed , confronting eqs .
( [ lbl04 ] ) and ( [ lbl08 ] ) , taking into account of the definition ( [ sbl - amp ] ) for @xmath44 , one can see that the inequality ( [ lbl09 ] ) for neutrinos is realized if @xmath140 < \mathrm{re}\!\left
[ u_{\alpha{{c } } } u_{\beta{{c}}}^ { * } u_{\alpha{{d}}}^ { * } u_{\beta{{d } } } \right ] \ , . \label{lbl13}\ ] ] the product @xmath141 is invariant under rephasing of the mixing matrix , _ i.e. _ the transformations @xmath142 and the transformations @xmath143 . using these transformations
we can consider , without loss of generality , @xmath144 , @xmath145 , @xmath146 , @xmath147 and write the inequality ( [ lbl13 ] ) as @xmath148 where @xmath149 . if we confine @xmath150 in the interval @xmath151 , the inequality ( [ lbl14 ] ) is not realized for any value of @xmath152 only
if @xmath153 and it is realized for all values of @xmath154 only if @xmath155 since the two extreme cases ( [ lbl15 ] ) and ( [ lbl16 ] ) correspond to the absence of cp - violating phases in the @xmath156-@xmath157 sector of the mixing matrix , it is clear that the observation of the inequality ( [ lbl09 ] ) for some , but not all values of @xmath152 would be an evidence in favor of cp violation in the lepton sector due to neutrino mixing .
this fact can also be seen by noting that if cp is conserved all the elements of the mixing matrix are real and there are two possibilities : @xmath158 in the first case we have @xmath159 and @xmath160[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \leq \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle \qquad \mbox{always } \ , .
\label{lbl20}\ ] ] in the second case we have @xmath161 and @xmath162[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \geq \langle p^{\mathrm{sbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle \qquad \mbox{always } \ , .
\label{lbl22}\ ] ] the inequality ( [ lbl22 ] ) has the same origin as the inequality ( [ lbl12 ] ) in the case of two - neutrino mixing . if @xmath56 is known and the the detector has a sufficient energy resolution ( or with a sufficiently narrow - band beam and varying the neutrino energy ) , it is even possible to measure the cp - violating phase @xmath150 by looking at the variation of the transition probability as a function of @xmath152 .
for example , we have @xmath163 the ratio of eqs .
( [ lbl24 ] ) and ( [ lbl23 ] ) gives @xmath164 independently from the value of @xmath165 .
let us finally consider the average transition probabilities that are measured in long - baseline experiments that are sensitive to values of @xmath55 much smaller than the actual value of @xmath56 . considering @xmath65[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } $ ] transitions , from eq .
( [ lbl05 ] ) we have @xmath57[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle = \frac{1}{4 } \ , a_{{\mu}e } + + \ , .
\label{lbl31}\ ] ] taking into account the bounds ( [ bounds ] ) for @xmath58 and @xmath59 , the allowed range ( [ asy09 ] ) for @xmath166 and the constraint @xmath167 for @xmath168 , one can find that the average transition probability ( [ lbl31 ] ) is bounded by @xmath169[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle \leq \frac{1}{4 } \
, a_{{\mu}e } + \lambda^{+}_\mu(\mathrm{max } ) \lambda^{+}_e \ , .
\label{lbl32}\ ] ] the lower bound is reached for @xmath170 and @xmath171 or @xmath172 and @xmath173 . the quantity @xmath174 is the minimum value of @xmath175 in the allowed range for @xmath59 . for the values of @xmath30 in which @xmath59 is limited by the super - kamiokande asymmetry constraint ( [ ask ] ) , the minimum of @xmath175 , shown in fig .
[ umu - lim ] , occurs for @xmath176 the upper bound in eq.([lbl32 ] ) is reached for @xmath170 and @xmath177 or @xmath172 and @xmath178 .
the quantity @xmath179 is the maximum value of @xmath180 in the allowed range for @xmath59 , that , as shown in fig .
[ umu - lim ] , occurs always for the maximum allowed value of @xmath59 , @xmath181 .
figure [ pme ] shows the allowed range ( [ lbl32 ] ) as a function of @xmath30 ( shadowed region ) .
the dashed line in fig .
[ pme ] represents the unitarity upper bound @xmath81[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\mu } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle \leq 1 - \langle p^{\mathrm{lbl}}_{\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle $ ] , with the upper bound for @xmath127[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{e } } } \rangle $ ] given by eq .
( [ lbl06 ] ) . since @xmath182 is small , the dashed line in fig .
[ pme ] practically coincides with the solid line in fig .
[ pee ] .
the other constrained channels are @xmath53 with @xmath135 and @xmath136 , for which @xmath81[l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\alpha } } \to\stackrel{\makebox[0pt][l ] { $ \hskip-3pt\scriptscriptstyle(-)$}}{\nu_{\beta } } } \rangle $ ] has only the unitarity upper limits given by the average of eq .
( [ lbl071 ] ) .
we have presented the constraints on the neutrino mixing parameters obtained from the results of neutrino oscillation experiments in the framework of the two four - neutrino schemes ( [ ab ] ) that are compatible with all neutrino oscillation data .
we have shown that the parameters @xmath58 and @xmath73 defined in eq .
( [ sigma ] ) are small , but do not vanish .
their allowed ranges are presented in figs .
[ ce ] and [ oms ] .
we have also derived the bounds for the mixing of @xmath15 with the two massive neutrinos whose mass - squared difference is responsible for atmospheric neutrino oscillations ( eq .
( [ asy09 ] ) ) . in sections [ long - baseline disappearance experiments ] and [ long - baseline appearance experiments ] we presented the bounds for the oscillation probabilities in long - baseline experiments ( some of which have been derived already in ref .
we have also shown that a comparison of the probability measured in long - baseline appearance experiments with the corresponding average probability measured in short - baseline experiments may give information on the relative phases of the relevant elements of the mixing matrix and , in particular , may provide an evidence for the existence of cp violation in the lepton sector .
this method may be more convenient than the traditional method based on the comparison of the oscillation probabilities of neutrinos and antineutrinos if only a neutrino or antineutrino beam is available .
in this paper we have considered long - baseline experiments neglecting matter effects , that are rather complicated in the framework of four - neutrino mixing and will be discussed in detail in another article @xcite .
we think that it is important to discuss long - baseline oscillations in vacuum for at least two reasons .
first , because vacuum oscillations can be treated analytically and all the effects can be understood in a clear way .
the matter effects , that often must be calculated numerically , can be appreciated as corrections ( sometimes large ) to the oscillations in vacuum .
second , in order to measure accurately the properties of neutrinos without the uncertainty and complications caused by matter effects , it may be possible , in some indeterminate future , to perform a long - baseline experiment in vacuum .
for example , it may be possible to produce a neutrino beam in a orbital station ( existing space stations orbit at an altitude between 300 and 400 km ) and shoot it to some detector(s ) on the earth .
this experiment would allow to measure accurately the properties of neutrinos when the beam does not cross the earth and , with this knowledge , to perform an accurate tomography of the earth using the matter effects when the beam crosses the earth . at present
such an experiment is fanciful , but we know that many experiments have been performed up to now that were unthinkable or believed to be impossible in the past .
( homestake coll . ) , astrophys .
j. * 496 * , 505 ( 1998 ) ; k.s .
( kamiokande coll . ) , phys .
lett . * 77 * , 1683 ( 1996 ) ; w. hampel _
( gallex coll . ) , phys .
b * 447 * , 127 ( 1999 ) ; j.n .
( sage coll . ) , astro - ph/9907113 ; y. fukuda ( super - kamiokande coll . ) , phys .
lett . * 82 * , 2430 ( 1999 ) .
y. fukuda _
et al . _ , phys .
lett . * 82 * , 2644 ( 1999 ) ; y. fukuda _ et al . _
( kamiokande coll . ) , phys .
b * 335 * , 237 ( 1994 ) ; r. becker - szendy _ et al . _
( imb coll . ) , nucl .
b ( proc . suppl . )
* 38 * , 331 ( 1995 ) ; w.w.m .
( soudan 2 coll . ) , phys .
b * 449 * , 137 ( 1999 ) ; m. ambrosio _
( macro coll . ) , phys .
b * 434 * , 451 ( 1998 ) . c. athanassopoulos _ et al . _
( lsnd coll . ) , phys .
* 75 * , 2650 ( 1995 ) ; phys .
lett . * 77 * , 3082 ( 1996 ) ; phys .
lett . * 81 * , 1774 ( 1998 ) ; g. mills ( lsnd coll . ) , talk presented at the xxxiv@xmath183 rencontres de moriond _ electroweak interactions and unified theories _ , les arcs , march 1999 ( http://moriond.in2p3.fr/ew/transparencies ) .
c. giunti , hep - ph/9909465 , talk presented at _ neutrino mixing _
, meeting in honour of samoil bilenky s 70@xmath183 birthday , torino , 2527 march 1999 ; c. giunti , hep - ph/9910336 , talk presented at the icfa / ecfa workshop _
neutrino factories based on muon storage rings _ , @xmath184-fact99 , lyon , france , 5 - 9 july 1999 .
bahcall , p.i .
krastev and a.yu .
smirnov , phys .
d * 58 * , 096016 ( 1998 ) , hep - ph/9807216 ; y. fukuda _ et al .
* 82 * , 1810 ( 1999 ) , hep - ex/9812009 ; v. barger and k. whisnant , phys .
b * 456 * , 54 ( 1999 ) , hep - ph/9903262 ; m.c .
gonzalez - garcia _ et al .
_ , hep - ph/9906469 ; j.n .
bahcall , p.i .
krastev and a.yu .
smirnov , hep - ph/9911248 .
peltoniemi , d. tommasini and j.w.f .
valle , phys .
b * 298 * , 383 ( 1993 ) ; e.j .
et al . _ ,
b * 357 * , 608 ( 1995 ) ; s.c .
et al . _ ,
b * 430 * , 296 ( 1998 ) ; b. brahmachari and r.n .
mohapatra , phys .
b * 437 * , 100 ( 1998 ) ; s. mohanty , d.p .
roy and u. sarkar , phys .
b * 445 * , 185 ( 1998 ) ; j.t .
peltoniemi and j.w.f .
valle , nucl . phys .
b * 406 * , 409 ( 1993 ) ; q.y .
liu and a.yu .
smirnov , nucl .
b * 524 * , 505 ( 1998 ) ; d.o . caldwell and r.n .
mohapatra , phys .
d * 48 * , 3259 ( 1993 ) ; e. ma and p. roy , phys .
d * 52 * , r4780 ( 1995 ) ; a.yu .
smirnov and m. tanimoto , phys .
d * 55 * , 1665 ( 1997 ) ; n. gaur _ et al .
d * 58 * , 071301 ( 1998 ) ; e.j .
chun , c.w .
kim and u.w .
lee , phys .
d * 58 * , 093003 ( 1998 ) ; k. benakli and a.yu .
smirnov , phys .
* 79 * , 4314 ( 1997 ) ; y. chikira , n. haba and y. mimura , hep - ph/9808254 ; c. liu and j. song , phys . rev .
d * 60 * , 036002 ( 1999 ) , hep - ph/9812381 ; w. grimus , r. pfeiffer and t. schwetz , hep - ph/9905320 .
gomez - cadenas and m.c .
gonzalez - garcia , z. phys .
c * 71 * , 443 ( 1996 ) , hep - ph/9504246 ; n. okada and o. yasuda , int .
a * 12 * , 3669 ( 1997 ) , hep - ph/9606411 ; s. goswami , phys .
d * 55 * , 2931 ( 1997 ) , hep - ph/9507212 ; s.m .
bilenky , c. giunti , w. grimus and t. schwetz , astropart .
phys . * 11 * , 413 ( 1999 ) , hep - ph/9804421 ; v. barger , y.b .
dai , k. whisnant and b.l .
young , phys .
d * 59 * , 113010 ( 1999 ) , hep - ph/9901388 ; v. barger , t.j . weiler and k. whisnant , phys . lett .
b * 427 * , 97 ( 1998 ) , hep - ph/9712495 ; c. giunti , hep - ph/9906275 ( phys .
d , in press ) ; hep - ph/9906456 ( phys . lett .
b , in press ) ; s.m .
et al . _ , hep - ph/9907234 ( phys .
b , in press ) . s.m .
bilenky , c. giunti and w. grimus , eur .
j. c * 1 * , 247 ( 1998 ) , hep - ph/9607372 ; proc . of _
neutrino 96 _ , helsinki , june 1996 , edited by k. enqvist _
et al . _ , p. 174
, world scientific , 1997 , hep - ph/9609343 .
v. barger , s. pakvasa , t. j. weiler , and k. whisnant , phys .
rev . * d58 * , 093016 ( 1998 ) , hep - ph/9806328 .
s. m. bilenky , c. giunti , w. grimus , and t. schwetz , phys .
* d60 * , 073007 ( 1999 ) , hep - ph/9903454 .
s. m. bilenky , c. giunti , and w. grimus , phys . rev .
* d57 * , 1920 ( 1998 ) , hep - ph/9710209 .
m. campanelli , hep - ex/9905035 ; p. picchi and f. pietropaolo , hep - ph/9812222 ; k. zuber , hep - ex/9810022 ; y. oyama ( k2k coll . ) , hep - ex/9803014 ; p. adamson _
( minos coll . ) , numi - l-476 ( march 1999 ) ; p. cennini _ et al .
_ ( icarus coll . ) , lngs-94/99-i ( 1994 ) .
s. m. bilenkii , c. giunti , and w. grimus , prog .
* 43 * , 1 ( 1999 ) , hep - ph/9812360 .
f. dydak _ et al .
_ , phys . lett . *
134b * , 281 ( 1984 ) .
y. fukuda _
( super - kamiokande coll . ) , phys .
rev . lett . * 81 * , 1562 ( 1998 ) ; k. scholberg ( super - kamiokande coll . ) , hep - ex/9905016 . | we consider the two four - neutrino schemes that are compatible with all neutrino oscillation data .
we present the range of the corresponding mixing parameters allowed by the results of neutrino oscillation experiments .
we discuss the implications for long - baseline experiments and we suggest the possibility to reveal the presence of cp - violating phases through a comparison of the oscillation probability measured in long - baseline experiments with the corresponding average oscillation probability measured in short - baseline experiments . |
the ads / cft correspondence is a duality between the large @xmath0 limit of @xmath1 superconformal field theories and string theory in a higher dimensional anti - de sitter spacetime @xcite . in this correspondence
the ads space shows up both as a near horizon geometry of a set of coincident d - branes .
precise prescriptions for the realization of the ads / cft correspondence were formulated in @xcite . in this correspondence ,
bulk fields act as classical sources for boundary correlation functions @xcite@xmath2 @xcite .
one of the striking features @xcite of this correspondence is that it is a realization of the holographic principle @xcite@xmath2 @xcite : the degrees of freedom of a quantum theory with gravity can be mapped on the corresponding boundary " .
string theory was originally proposed as a model for strong interactions with the successful results of veneziano amplitude in the regge regime @xcite@xmath2 @xcite .
however this model was unable to reproduce the experimental results for hard and deep inelastic scatterings .
these behaviours were latter explained by qcd which is today the accepted theory to describe strong interactions . despite of its success in high energies , qcd is non perturbative in the low energy regime , where one needs involved lattice calculations .
in fact qcd and strings may be viewed as complementary theories .
this was first discovered by t hooft in the work @xcite where he found a relation between large @xmath0 @xmath1 gauge theories and strings @xcite .
one of the long standing puzzles for the string description of strong interactions was the high energy scattering at fixed angles .
for string theory in flat space such a process is soft : the amplitudes decay exponentially with energy @xcite while both experimental data and qcd theoretical predictions @xcite indicate a hard behavior ( amplitudes decaying with a power of energy . a recent solution for this puzzle was proposed by polchinski and strassler @xcite , inspired by the ads / cft correspondence , considering strings in a 10 dimensional space which assimptotically tends to an ads@xmath3 times a compact space .
this space is considered as an approximation for the space dual to a confining gauge theory which can be associated with qcd .
an energy scale was introduced and identified with the lightest glueball mass .
then they found the glueball high energy scattering amplitude at fixed angles ( with the correct qcd scaling ) integrating the string amplitude over the warped ads extra dimension weighted by the dilaton wave function .
other approaches to this problem have also been discussed in @xcite@xmath2 @xcite .
the first proposal to deform the ads / cft correspondence to describe strong interactions is due to witten @xcite . in this work , inspired by hawking and page @xcite
, he suggests that the ads space can acommodate a black hole ( the ads schwarzschild space ) and the horizon radius defines an energy scale which can be associated with the qcd mass gap .
following this proposal some authors have calculated glueball masses in this space @xcite-@xcite .
other gauge / string dualities were discussed in @xcite-@xcite . using the infrared cutoff proposed by polschinski and strassler we have calculated glueball masses in an ads slice from the zeros of bessel functions @xcite .
this approach was later extended to higher spin states and applied to calculate masses of light barions and mesons @xcite .
this description of higher spin states was then applied to obtain the regge trajectories of glueballs @xcite , which showed to be consistent with the soft pomeron trajactory .
antoher successful application of the ads / cft correspondence is the calculation of wilson loops .
using this duality one can identify the worldsheet area with the wilson loop area of the correspponding gauge theory .
then , for static strings one finds the corresponding energy configuration .
this idea has been implemented independently by rey and yee @xcite , and maldacena @xcite .
it was later generalized to others spaces @xcite .
we have considered the calculation of wilson loops in a ads slice @xcite finding the well known cornell potential @xcite-@xcite for a heavy quark anti - quark pair .
this calculation was extended to finite temperature showing a deconfining phase trasition @xcite . in this article
we are going to review these recent results regarding approximate stringy descriptions of strong interactions .
let us start by briefly reviewing some early results on strong interactions which lead to the proposal of string theory .
these results date back to the decade of 1960 .
the first is the experimental observation , in hadronic scattering , of an apparently infinite tower of resonances with mass and angular momenta related by @xmath4 where @xmath5 ( gev)@xmath6 is the regge slope .
another important fact is that the properties of hadronic scattering in the so called regge regime are nicely described by the amplitude postulated by veneziano ( in terms of mandelstan variables ( @xmath7 ) ) @xmath8 where @xmath9 is the euler gamma function and @xmath10 . a strong motivation for relating hadrons to strings is the fact that these two results of eqs .
( [ 1 ] ) and ( [ 2 ] ) can be reproduced from a relativistic bosonic string @xcite .
after quantization , one finds that the spectrum of excitations shows up in representations satisfying eq .
( [ 1 ] ) .
the string scattering amplitudes also reproduce the veneziano result eq .
( [ 2 ] ) . on the other hand , some problems concerning a possible string description of hadronic physics have been a challenge for physicists for a long time .
one of them is the behavior of the scattering amplitudes at high energy .
if one considers the regge limit , corresponding to @xmath11 with fixed @xmath12 the veneziano amplitude behaves as @xmath13 in agreement with experimental results .
actually this was one of the inputs for building up the veneziano amplitude .
however considering high energy scattering at fixed angles that correspond to the limit @xmath14 with @xmath15 fixed the amplitude behaves as ( soft scattering ) @xmath16 while experimental results for hadrons show a hard scattering behavior @xmath17 that is reproduced from qcd @xcite .
recently polchinski and strassler @xcite introduced an infrared cut off in the ads space and reproduced the hard scattering behavior of strong interactions at fixed angles from string theory .
inspired by the ads / cft correspondence they assumed a duality between gauge theory glueballs and string theory dilatons in an ads slice and found the qcd scaling .
this scaling was also obtained in @xcite from a mapping between quantum states in ads space and its boundary found in @xcite .
we considered an ads slice as approximately dual to a confining gauge theory .
the slice corresponds to the metric @xmath18 with @xmath19 where @xmath20 is an energy scale chosen as the mass of the lightest glueball .
we used a mapping between fock spaces of a scalar field in ads space and operators on the four dimensional boundary , defined in @xcite .
considering a scattering of two particles into @xmath21 particles one finds a relation between bulk and boundary scattering amplitudes @xcite @xmath22 where @xmath23 is the scaling dimension of the boundary operators and @xmath24 is the boundary momentum scale .
this leads to the scaling of the amplitude @xmath25 where @xmath26 is the total dimension of scattered particles .
this reproduces the qcd scaling of eq .
( [ 6 ] ) . for some other results concerning qcd scattering properties from string theory see
also @xcite-@xcite .
using the phenomenological approach of introducing an energy scale by considering an ads slice we found estimates for scalar glueball mass ratios @xcite . in the ads@xmath3 slice we considered dilaton fields satisfying dirichlet boundary conditions at @xmath27 @xmath28 where @xmath29 ,
@xmath30 is the momentum associated with the @xmath31 direction and @xmath32 are the zeroes of the bessel functions : @xmath33 on the boundary ( @xmath34 we considered scalar glueball states @xmath35 and their excitations @xmath36 with masses @xmath37 . assuming an approximate gauge / string duality the glueball masses are taken as proportional to the dilaton discrete modes : @xmath38 so , the ratios of glueball masses are related to zeros of the bessel functions @xmath39 note that these ratios are independent of the size of the slice @xmath40 .
our estimates compared with @xmath41 lattice @xcite and ads - schwarzschild @xcite ( in gev ) are shown in table i. .four dimensional glueball masses in gev with dirichlet boundary conditions .
the value 1.61 of the third and fourth columns is an input taken from lattice results .
[ cols="<,^,^,^",options="header " , ] we found non linear relations between spin and mass squared . considering a linear fit representing regge trajectories @xmath42 we found for neumann boundary conditions and the states @xmath43 with @xmath44 @xmath45 as shown in figure 1
. for dirichlet boundary conditions , taking the states @xmath43 with @xmath44 we found @xmath46 as shown in figure 2 .
so , neumann boundary conditions give a glueball trajectory consistent with that of pomerons , eq .
( [ 11 ] ) .
these kind of boundary conditions appear in the randall sundrum model @xcite as a consequence of the orbifold condition .
wilson loops are an important tool to discuss confinement in gauge theories since they show the behaviour of the energy associated with a given field configuration . in the ads
/ cft correspondence wilson loops for a heavy quark anti - quark pair in the gauge theory can be calculated from a static string in the ads space @xcite .
the corresponding energy is a non confining coulomb potential as expected for a conformal theory . for an excellent review and extension to other metrics
see @xcite .
we calculated wilson loops for a quark anti - quark pair in d3-brane space finding different behaviors , with respect to confinement , depending on the quark position @xcite . here
, we are going to review @xcite that the wilson loop calculated in an ads slice leads to the heavy quark anti - quark cornell potential : @xmath47 where @xmath48 and @xmath49 gev@xmath50 @xcite-@xcite .
the metric ( [ adspoincare ] ) of the ads space can be rewritten as @xmath51 where @xmath52 .
we have calculated the energy of a static string in an ads slice defined by @xmath53 . the quark anti - quark pair ( string endpoints )
is located at @xmath54 , separated by a four dimensional ( @xmath55 coordinates ) distance @xmath56 and there is an infrared cut off in the space at @xmath57 . from now on we choose @xmath58 .
note that there are two kinds of geodesics , as shown in figure 3 , depending on the value of @xmath56 .
for small quark separation @xmath59 the geodesics are curves ( like curve * a * ) with one minimum value of the coordinate @xmath60 which is related to @xmath56 by @xmath61 where @xmath62 is the elliptic integral @xmath63 the critical value corresponds to @xmath64 as in curve * b * of figure 3 .
the energy for @xmath65 can be calculated as @xmath66 \,.\ ] ] where @xmath67 is the string tension and we have subtracted the constant @xmath68 in such a way that the energy is finite even in the limit @xmath69 . the integral @xmath70 is @xmath71 d\rho \,.\ ] ] for @xmath72 , the geodesics reach the infrared brane as shown in curve * c * of figure 3 .
5.5 cm ( 0,0)(18,0 ) * ( 25,-8)(0,0)38 ( 26,-0.5)@xmath73 ( 26,19.5)@xmath74 ( 26,30)@xmath55 ( 24.5,-0.5)@xmath75 ( 24.5,19.5)@xmath75 ( 24.5,14.5)@xmath75 ( 24.5,4.5)@xmath75 ( 24.5,24.5)@xmath75 ( 24.5,-5.5)@xmath75 ( 17,18)b ( 19,14)a ( 19,24)c ( 3,10)(1,0)30 ( 33,8)@xmath76 ( 22.7,-10 ) @xmath54 ( 10,-6)@xmath77 ( 25,15)(10,10)(25,5 ) ( 25,20)(0,10)(25,0 ) ( 25,25)(17,22)(12.6,18 ) ( 25,-5)(17,-2)(12.6,2 ) ( 12.5,18)(12.5,10)(12.5,2 ) ( 12.55,18)(12.55,10)(12.55,2 ) ( 12.55,-4.3)(0,1)30(0,1)0.5 * 2.5 cm the energy can be calculated again subtracting the constant @xmath78 associated with the quark masses .
we obtain @xmath79 \,+\,\frac { 1 } { 2 \pi \alpha^\prime\,}\ , ( l - l_{crit})\ , \cr\cr = \frac { r } { \pi \alpha^\prime } \big [ i_2 ( r_1/r ) - i_1 ( r_1/r ) - 1\,\big ] + \frac { l } { 2 \pi \alpha^\prime } , \label{er1}\end{aligned}\ ] ] where we have used the definition of @xmath56 , eq . ( [ lr1 ] ) , with the identification that at @xmath80 one has @xmath81 . taking the limit @xmath82
, we find a potential which is approximately coulombian for small @xmath83 and has a leading linear confining behaviour for large distances @xmath84 the identification of the static string energy with the cornell potential ( [ cornell ] ) leads to @xmath85 with @xmath86 ^ 2.$ ] so , for the particular choice @xmath87 , we find an effective ads radius @xmath88 .
then the energies take the forms:@xmath89 the shape of this potential energy is very close to the cornell potential , as shown in figure 4 . in gev@xmath90 , for ads slices with @xmath91 and @xmath92 . for @xmath93
the energy behaves as the cornell potential.,width=302 ]
a gauge / string duality involving a gauge theory at finite temperature was proposed by witten in @xcite inspired by the work of hawking and page @xcite . in this approach , for high temperatures ,
the ads space accommodates a schwarzschild black hole and the horizon radius is proportional to the temperature . for low temperatures the dual space would be an ads space with compactified time dimension , known as thermal ads .
the gauge string duality using the ads schwarzschild space has recently been applied to obtain the viscosity of a quark gluon plasma @xcite .
an ads schwarzschild black hole metric was considered in @xcite as a phenomenological model for a space dual to a theory with both mass scale and finite temperature .
the corresponding metric is @xmath94 where @xmath95 , @xmath96 and the hawking temperature @xmath97 is related to the horizon radius by @xmath98 . at zero temperature
this space becomes an anti - de sitter ( ads ) slice .
the problem of static strings in a space with metric ( [ metric ] ) , without any cut off , was discussed in detail in @xcite . calculating the energy of static strings in this space we found a deconfinement phase transition at a critical temperature @xmath99 .
this transition shows up because , depending on the horizon radius ( temperature ) relative to the cut off position @xmath100 , we have different behaviors for the energy . if @xmath101 the string will not be affected by the presence of the brane since it does not cross the horizon .
so the energy will be that described in ref .
@xcite and there will be no confinement . if @xmath102 the energy for large quark anti - quark distance @xmath56 will grow up linearly with @xmath56 with a temperature dependent coefficient and the quarks are confined .
the critical temperature @xmath103 corresponds to @xmath104 . as in the zero temperature case
we choose @xmath105 from now on .
we show in figure 5 the energies obtained in @xcite for temperatures @xmath106 and @xmath107 .
this figure illustrates the fact that in our model the energy of static strings associated with the quark anti - quark potential present a confining behavior for temperatures below @xmath108 . in this case
there is a linear term in the energy , for large quark distances @xmath109 given by @xmath110 with @xmath111 at zero temperature this coefficient is identified with the string tension of the cornell potential @xmath112 gev@xmath113 . for temperatures
@xmath114 there is no confinement since the energy is finite when @xmath115 .
choosing the brane position to have the value @xmath116 as in the zero temperature case @xcite we find a critical temperature @xmath117 .
our results agree with lattice calculations for qcd @xcite-@xcite only for high temperatures .
the results from fluctuations of strings in flat space at finite temperature @xcite and from lattice calculations @xcite imply corrections to the string tension of order @xmath118 at low temperatures , while our model predicts corrections of order @xmath119 , as can be seen from eq .
( [ tensiont ] ) .
there is also a very recent result by herzog @xcite which indicates that the dual space for temperatures below @xmath99 should be indeed the thermal ads .
if we had considered the thermal ads metric for low temperatures , instead of the ads schwarzschild black hole metric , we would get no thermal corrections to the string tension
. it would be interesting to find a holographic phenomenological model that gives the expected low temperature corrections .
let us mention the recent articles @xcite and @xcite that also discuss thermal effects in the gauge / string duality context .
for a very recent review of thermal properties of qcd related to the quark - gluon plasma and rhic experiment see @xcite . for other interesting results concerning gauge / string duality and qcd
see for instance @xcite-@xcite .
* acknowledgements : * we would thank andrey bytsenko , sebastio a. dias , jos a. helayl - neto and maria emilia x. guimares , organizers of `` fifth international conference on mathematical methods in physics , '' at cbpf , rio de janeiro , brazil , where this talk was presented , for the very warm atmosphere during the workshop .
the authors are partially supported by cnpq and faperj .
h. boschi - filho and n. r. f. braga , jhep * 0305 * ( 2003 ) 009 .
h. boschi - filho and n. r. f. braga , eur .
j. c * 32 * ( 2004 ) 529 .
g. f. de teramond and s. j. brodsky , phys .
* 94 * ( 2005 ) 201601 .
h. boschi - filho , n. r. f. braga and c. n. ferreira , phys .
d * 73 * ( 2006 ) 106006 . c. quigg and j. l. rosner , phys .
b * 71 * ( 1977 ) 153 .
e. eichten , k. gottfried , t. kinoshita , k. d. lane and t. m. yan , phys .
d * 17 * ( 1978 ) 3090 .
[ erratum - ibid .
d * 21 * ( 1980 ) 313 ] .
s. j. brodsky and g. f. de teramond , phys
* b 582 * ( 2004 ) 211 . o. andreev , phys .
* d 70 * ( 2004 ) 027901 .
o. andreev , phys .
d * 71 * ( 2005 ) 066006 . c. j. morningstar and m. peardon , phys .
d 56 ( 1997 ) 4043 .
e. caceres and r. hernandez , phys .
b * 504 * ( 2001 ) 64 .
r. apreda , d. e. crooks , n. evans , m. petrini , jhep * 0405 * ( 2004 ) 065 .
x. amador and e. caceres , jhep * 0411 * ( 2004 ) 022 .
e. caceres and c. nunez , jhep * 0509 * ( 2005 ) 027 .
p. v. landshoff , `` pomerons , '' , published in
elastic and difractive scattering " proceedings , ed . v. kundrat and p. zavada , 2002 , arxiv : hep - ph/0108156 .
r. d. pisarski and o. alvarez , phys .
d * 26 * ( 1982 ) 3735 .
de forcrand , g. schierholz , h. schneider and m. teper , phys . lett .
b * 160 * ( 1985 ) 137 . c. p. herzog , `` a holographic prediction of the deconfinement temperature , '' arxiv : hep - th/0608151 .
k. ghoroku and m. yahiro , phys .
d * 73 * ( 2006 ) 125010 .
o. andreev and v. i. zakharov , `` the spatial string tension , thermal phase transition , and ads / qcd , '' arxiv : hep - ph/0607026 .
l. a. pando zayas , j. sonnenschein and d. vaman , nucl .
b * 682 * ( 2004 ) 3 .
o. andreev and w. siegel , phys .
d * 71 * ( 2005 ) 086001 .
f. bigazzi , a. l. cotrone , l. martucci and l. a. pando zayas , phys .
d * 71 * ( 2005 ) 066002 .
j. erlich , e. katz , d. t. son and m. a. stephanov , phys .
* 95 * ( 2005 ) 261602 .
l. da rold and a. pomarol , nucl .
* b 721 * ( 2005 ) 79 .
n. evans , j. p. shock and t. waterson , phys.lett .
* b622 * ( 2005 ) 165 .
s. j. brodsky and g. f. de teramond , aip conf .
* 814 * ( 2006 ) 108 .
[ arxiv : hep - ph/0510240 ] .
s. j. brodsky and g. f. de teramond , phys .
lett . * 96 * ( 2006 ) 201601 . | the ads / cft correspondence is an exact duality between string theory in anti - de sitter space and conformal field theories on its boundary .
inspired in this correspondence some relations between strings and non conformal field theories have been found .
exact dualities in the non conformal case are intricate but approximations can reproduce important physical results . a simple approximation consists in taking just a slice of the ads space with a size that can be related to the qcd energy scale . here
we will discuss how this approach can be used to obtain the scaling of high energy qcd scattering amplitudes , glueball masses and regge trajectories , and the potential energy for a quark anti - quark pair .
this set up is modified to include finite temperature effects and obtain a deconfining phase trasition . |
let @xmath6 be a @xmath7-dimensional closed , connected , oriented manifold with an effective @xmath8-dimensional ( i.e. , half - dimensional ) torus @xmath9-action .
we call @xmath6 ( or @xmath10 ) a _ torus manifold _ if @xmath11 ( see @xcite ) , where @xmath12 is the set of fixed points . a toric manifold ( i.e. , a non - singular , complete toric variety viewed as a complex analytic space ) with the restricted @xmath9-action is a typical example of torus manifolds . recall that a toric manifold is a complex @xmath13-manifold with the dense orbit ( see @xcite , @xcite ) , and @xmath9 is the maximal compact subgroup of @xmath13
. a fundamental result of toric geometry tells us that there is a one - to - one correspondence between toric manifolds and combinatorial objects called fans .
thus , topological ( more precisely , geometric ) invariants of toric manifolds can be described in terms of combinatorial invarinats of fans , such as equivariant cohomology rings , equivariant characteristic classes and other topological invariants . in 2003 , hattori - masuda introduced a torus manifold as the topological generalization of a toric manifold in @xcite .
they also introduced the combinatorial objects , called multi - fans ( see @xcite , @xcite ) , and computed topological invariants ( such as equivariant characteristic classes or todd genus for unitary torus manifolds ) in terms of multi - fans .
however , unlike toric geometry , a multi - fan does not contain enough information to determine some topological invariants of torus manifolds ( e.g. equivariant cohomology ) .
so , in 2007 , maeda - masuda - panov introduced another combinatorial objects , called torus graphs motivated by gkm graphs introduced by guillemin - zara in @xcite .
the combinatorial information of torus graphs can completely determine the equivariant cohomology rings of torus manifolds with vanishing odd degree cohomology , i.e. , @xmath14 ( in this paper , we only consider the integer coefficient ) , see @xcite , @xcite ( and also see section [ sect3 ] in this paper about torus graphs ) .
however , in general , there is no one - to - one correspondence between torus manifolds with @xmath2 and torus graphs .
so , we are naturally led to ask the following two questions : ( 1 ) which subclasses of torus manifolds are completely determined by combinatorial objects ( like multi - fans or torus graphs ) ; ( 2 ) if we find such a subclass of torus manifolds , how we can classify such torus manifolds .
several mathematicians have answered to the 1st question ; for example , davis - januszkiewicz @xcite for the subclass called _ quasitoric manifolds _ ( see @xcite or section [ sect4.3 ] in this paper ) , ishida - fukukawa - masuda @xcite for the subclass called _ topological toric manifolds _ , and wiemeler for the class of simply connected @xmath0-dimensional torus manifolds with @xmath2 in @xcite ( see theorem [ wi2 ] ) .
the aim of this paper is to answer to the 2nd question for the class of simply connected @xmath0-dimensional torus manifolds with @xmath2 by using torus graphs .
let us briefly recall the classification results for torus manifolds with lower dimensions .
if @xmath15 acts on a compact @xmath16-dimensional manifold @xmath6 , then @xmath6 is the @xmath16-dimensional sphere @xmath5 , the @xmath16-dimensional real projectiove space @xmath17 or the @xmath16-dimensional torus @xmath18 . because @xmath11 and @xmath6 is oriented
, @xmath6 must be equivariantly diffeomorphic to @xmath5 with @xmath15-action ( also see @xcite ) .
when @xmath19 .
by orlik - raymond s theorem in @xcite , we have the following fact : [ or - thm ] let @xmath6 be a @xmath20-dimensional simply connected torus manifold .
then , @xmath6 is equivariantly diffeomorphic to the @xmath20-sphere @xmath21 or an equivariant connected sum of copies of complex projective spaces @xmath22 , @xmath23 ( reversed orientation ) or hirzebruch surfaces @xmath24 . here , in theorem [ or - thm ]
, a hirzebruch surface @xmath24 ( @xmath25 ) is a manifold which is defined by the projectivization of the complex @xmath16-dimensional vector bundle @xmath26 over @xmath27 , where @xmath28 is the tautological and @xmath29 is the trivial complex line bundles over @xmath27 . in this paper
, we prove that an orlik - raymond type theorem in theorem [ or - thm ] also holds for simply connected @xmath0-dimensional torus manifolds with @xmath2 .
before we state our main results , we introduce the result for non - simply connected torus manifolds .
as one of the consequences of masuda - panov s theorem ( see theorem [ 2-equiv ] in section [ sect2.2 ] ) , we have the following proposition ( also see @xcite ) : [ wiemeler ] let @xmath30 be a @xmath0-dimensional torus manifold with @xmath31 ( might not be simply connected ) . then , there are a simply connected , @xmath0-dimensional torus manifold @xmath6 with @xmath2 and a homology @xmath1-sphere @xmath32 such that @xmath33 up to equivariantly diffeomorphism . here , in proposition [ wiemeler ] ,
the product manifold @xmath34 is the product of @xmath32 and the @xmath1-dimensional torus @xmath35 with the free @xmath36-action on the @xmath16nd factor , and the symbol @xmath37 represents the equivariant gluing along two free orbits of @xmath6 and @xmath34 . because the fundamental groups @xmath38 and @xmath39 are isomorphic in proposition [ wiemeler ] , @xmath30 is simply connected if and only if @xmath32 is simply connected , i.e. , the standard sphere .
our main theorem is a classification of simply connected torus manifolds appeared in proposition [ wiemeler ] .
[ main1 ] let @xmath6 be a simply connected @xmath0-dimensional torus manifold with @xmath2 .
then , @xmath6 is equivariantly diffeomorphic to the @xmath0-sphere @xmath40 or obtained by an equivariant connected sum of copies of @xmath0-dimensional quasitoric manifolds or @xmath21-bundles over @xmath41 equipped with the structure of a torus manifold .
this type of classification , i.e. , classification by equivariant connected sum , may be regarded as the @xmath0-dimensional analogue of orlik - raymond s classification in theorem [ or - thm ] .
so , in this paper , we call this theorem an _ orlik - raymond type classification _
( also see the papers @xcite and @xcite ) . in the paper @xcite
, izmestiev proves an orlik - raymond type classification for some class of @xmath1-dimensional small covers ( i.e. , the real analogue of quasitoric manifolds , see section [ sect4.2 ] ) , called a linear model .
the organization of this paper is as follows . in section [ sect2 ] ,
we recall the basic facts about torus manifolds . in section [ sect3 ]
, we recall a torus graph .
in particular , corollary [ key2 ] is the key fact to prove theorem [ main1 ] . in section [ sect4 ] , we introduce the torus graphs of @xmath3 , quasitoric manifolds and @xmath4-bundles over @xmath5 .
these torus graphs will be the basic graphs to classify simply connected @xmath0-dimensional torus manifolds with @xmath2 . in section [ sect5 ]
, we introduce the `` oriented '' torus graphs and translate the equivariant connected sum around fixed points of torus manifolds to the connected sum around vertices of oriented torus graphs . in section [ sect6 ] and [ sect7 ] , we prove theorem [ main1 ] . the brief outline of the proof is as follows . due to corollary [ key2 ] ,
there is a one - to - one corresponds between @xmath0-dimensional simply connected torus manifolds with @xmath2 and @xmath1-valent torus graphs .
therefore , in order to prove theorem [ main1 ] , it is enough to prove that an ( oriented ) torus graph can be decomposed into basic torus graphs in section [ sect4 ] by the connected sum .
we prove this by using combinatorial arguments .
in this section , we recall some basic facts about torus manifolds ( see @xcite or @xcite for details ) . a @xmath7-dimensional torus manifold @xmath6 is said to be _ locally standard _ if every point in @xmath6 has a @xmath42-invariant open neighborhood @xmath43 which is weakly equivariantly homeomorphic to an open subset @xmath44 invariant under the standard @xmath9-action on @xmath45 , where two group actions @xmath46 and @xmath47 are said to be _ weakly equivariantly homeomorphic _ if there is an equivariant homeomorphism from @xmath43 to @xmath48 up to an automorphism on @xmath9 ( see e.g. ( * ? ? ?
* section 2.1 ) for details ) .
let @xmath49 , @xmath50 , be a codimension - two torus submanifold in a @xmath7-dimensional torus manifold @xmath6 which is fixed by some circle subgroup @xmath51 in @xmath42 .
such @xmath49 is a @xmath52-dimensional torus manifold with @xmath53-action , called a _
characteristic submanifold_. because a torus manifold @xmath6 is compact , the cardinality of all characteristic submanifolds in @xmath6 is finite . if @xmath6 is locally standard , each characteristic submanifold is also locally standard .
an _ omniorientation _
@xmath54 of @xmath6 is a choice of orientation for the torus manifold @xmath6 as well as for each characteristic submanifold .
if there are just @xmath55 characteristic submanifolds in @xmath6 , there are exactly @xmath56 omniorientations ( see @xcite , @xcite )
. if @xmath6 has a @xmath42-invariant almost complex structure @xmath57 ( in this case , @xmath6 is automatically locally standard ) , then there exists the canonical omniorientation @xmath58 determined by @xmath57 .
we call the torus manifold @xmath6 with a fixed omniorientation @xmath54 an _ omnioriented torus manifold _ and denote it by @xmath59 .
the orbit space @xmath60 of a locally standard torus manifold @xmath6 naturally admits the structure of a `` topological '' manifold with corners .
we next recall the basic facts of a topological manifold with corners ( cf .
the definition of a smooth manifold with corners in @xcite ) and introduce the structure on @xmath60 .
the following notations will be often used : @xmath61=\{0,\ 1,\ \ldots,\ n\},\end{aligned}\ ] ] and @xmath62 let @xmath63 be an @xmath8-dimensional topological manifold with boundary .
chart with corners _ for @xmath63 is a pair @xmath64 , where @xmath65 is an open subset of @xmath63 and @xmath66 is homeomorphic from @xmath65 to a ( relatively ) open subset @xmath67 .
two charts with corners @xmath68 are said to be _ ( topologically ) compatible _ if the composition of functions @xmath69 is a strata - preserving homeomorphism .
this implies that if @xmath70 contains exactly @xmath71 zero coordinates then @xmath72 also contains exactly @xmath71 zero coordinates for @xmath73 .
we call the collection of compatible charts with corners @xmath74 whose domains cover @xmath63 an _
atlas_. then , its maximal atlas is called a _ structure with corners _ of @xmath63 . a topological manifold with boundary together with a structure with corners
is called a _ ( topological ) manifold with corners_. let @xmath75 be a point of an @xmath8-dimensional manifold with corners @xmath63 .
a chart @xmath64 with corners such that @xmath76 defines the number @xmath77 $ ] by the number of zero - coordinates of @xmath72 . by the compatibility of charts , this number is independent on the choice of a chart with corners which contains @xmath78 .
therefore , the map @xmath79 $ ] is well - defined . the number @xmath80 is called a _ depth _ of @xmath78 .
we call the closure of a connected component of @xmath81 ( @xmath73 ) a _ codimension-@xmath71 face_. in particular , the codimension-@xmath82 face is @xmath63 itself
. moreover , codimension-@xmath83 , @xmath84 and @xmath8 faces are called _ facet _ , _ edge _ and _ vertex _ , respectively .
the set of all edges and vertices is called a _ one - skeleton of @xmath63 _ ( or a _ graph of @xmath63 _ ) .
by restricting the structure with corners on @xmath63 to faces , we may regard each codimension-@xmath71 face as an @xmath85-dimensional ( sub)manifold with corners .
an @xmath8-dimensional manifold with corners @xmath86 is said to be a _
manifold with faces _ ( or a _ nice manifold with corners _ ) if @xmath86 satisfies the following conditions : 1 . for every @xmath87 $ ]
, there exists a codimension-@xmath71 face ; 2 . for each codimension-@xmath71 face @xmath88
, there are exactly @xmath71 facets @xmath89 such that @xmath88 is a connected component of @xmath90 ; moreover , @xmath91 for any facet @xmath92 ( @xmath93 ) .
let @xmath10 be a torus manifold .
when @xmath10 is locally standard , by using the differentiable slice theorem , the orbit space @xmath60 has the structure of an @xmath8-dimensional manifold with faces .
on the other hand , when @xmath6 satisfies @xmath2 , its orbit space @xmath60 satisfies a stronger condition by the following masuda - panov theorem ( see ( * ? ? ?
* lemma 2.1 ) and ( * ? ? ? * theorem 2 ) ) : [ 2-equiv ] let @xmath6 be a @xmath7-dimensional torus manifold .
then , the following two conditions are equivalent : 1 .
@xmath2 ; 2 .
the @xmath42-action on @xmath6 is locally standard and its orbit space @xmath60 has the structure of an @xmath8-dimensional face acyclic manifold with corners . here , in theorem [ 2-equiv ] , an @xmath8-dimensional _ face acyclic _ manifold with corners @xmath86 is an @xmath8-dimensional manifold with faces such that all faces @xmath94 ( include @xmath86 ) of @xmath86 are acyclic , i.e. , @xmath95 .
for example , if @xmath86 is a simply connected @xmath1-dimensional face acyclic manifold with corners , then it is easy to check that the boundary of @xmath86 is homeomorphic to the @xmath16-sphere @xmath5 .
moreover , in this case , we can also check that @xmath86 itself is homeomorphic to the @xmath1-dimensional disk @xmath96 . therefore , as one of the conclusions of theorem [ 2-equiv ] , we have the following corollary : [ disk ] let @xmath6 be a simply connected @xmath0-dimensional torus manifold with @xmath2 . then , its orbit space @xmath60 is homeomorphic to the @xmath1-dimensional disk . by the definition of a manifold with faces
@xmath86 , we can define a simplicial poset ( partially ordered set ) @xmath97 , called a _ face poset _ of @xmath86 ( see @xcite ) , by the set of faces in @xmath86 with the empty - set @xmath98 ordered by inclusion , where the empty set @xmath98 is the smallest element by this order , say @xmath99 .
we often denote the face poset structure of @xmath86 as @xmath100 .
let @xmath101 and @xmath102 be @xmath8-dimensional manifolds with faces .
we call @xmath101 and @xmath102 are _ combinatorially equivalent _ if their face posets @xmath103 and @xmath104 are isomorphic as a poset ( i.e. , there is an order - preserving bijection between them ) .
we denote them by @xmath105 . by definition of weakly equivariant homeomorphism , if two locally standard torus manifolds @xmath106 and @xmath107 are weakly equivariantly homeomorphic then @xmath108 .
let @xmath6 be a @xmath7-dimensional locally standard torus manifold . by the argument demonstrated in section [ sect2.2 ] , the orbit map @xmath109
may be regarded as the projection onto some manifold with faces @xmath86 .
let @xmath110 be the set of all facets in @xmath86 . by the definition of facet @xmath111 ,
its preimage @xmath112 is a characteristic submanifold @xmath49 .
then , there exists the circle subgroup @xmath113 fixing @xmath114 ( recall that @xmath115 ) .
recall that @xmath51 is determined by a primitive element in @xmath116 ( the lattice of lie algebra of @xmath42 ) .
therefore , by taking this primitive element ( up to sign ) in @xmath117 , we can define the following map : @xmath118 where @xmath119 represents the quotient of @xmath117 by signs . we call @xmath120 a _
characteristic function_. now the choice of omniorientation @xmath54 of @xmath6 determines the sign of @xmath120 as follows .
fix an omniorientation @xmath54 of @xmath6 .
namely , we fix the orientation of the tangent bundle of @xmath6 ( resp .
@xmath49 ) , say @xmath121 ( resp .
@xmath122 ) . restricting @xmath121 to the submanifold @xmath49 , say @xmath123 , we obtain the @xmath9-equivariant decomposition @xmath124 , where @xmath125 is the @xmath51-equivariant normal bundle of @xmath49 . therefore , because we fix the orientation of @xmath123 ( induced from the orientation of @xmath121 ) and that of @xmath122 , we may choose an orientation of @xmath125 such that the orientation of @xmath123 coincides with that of @xmath126 ( thus , we may regard @xmath125 as the complex line bundle over @xmath49 ) . because @xmath51 acts on @xmath125 , we may choose an orientation of @xmath51 such that the @xmath51-action preserves the orientation of @xmath125 .
this orientation of @xmath51 determines the sign of @xmath127 for @xmath50 . by this way
, we have the following function : @xmath128 this is called an _ omnioriented characteristic function _ ( of @xmath59 ) , in this paper . [ rem - chfct ]
the characteristic function defined in @xcite may be regarded as the characteristic function @xmath120 as above .
on the other hand , the characteristic function defined in @xcite may be regarded as the characteristic function @xmath129 as above by taking an appropriate omniorientation ( also see ( * ? ? ? * section 5.2 ) ) .
let @xmath130 .
we define the subset @xmath131 $ ] as follows : @xmath132\ |\ p\in m_{i}\}.\ ] ] by the differentiable slice theorem around @xmath130 , we have that its cardinality @xmath133 for every @xmath130 . put @xmath134 . because the @xmath42-action on @xmath6 is effective , @xmath135 spans @xmath136 , i.e. , the determinant of the induced @xmath137-matrix @xmath138 satisfies that @xmath139 similarly , we have @xmath140 for each @xmath8 facets
such that @xmath141 for some vertex @xmath142 ( called _ the facets around a vertex _ ) . motivated by the observations as above
, we may abstractly define the characteristic function on a manifold with faces as follows ( see @xcite , @xcite for simple polytopes and @xcite , @xcite for manifold with faces ) : let @xmath86 be an @xmath8-dimensional manifold with faces and @xmath143 be the set of its facets .
let @xmath117 be the lattice of lie algebra of @xmath9 and @xmath119 be its quotient by @xmath144 . then
, a function @xmath145 is said to be a _
characteristic function _
if @xmath120 satisfies the relation for the facets around every vertex , and a function @xmath146 is said to be an _ omnioriented characteristic function _ if @xmath129 satisfies the relation for the facets around every vertex .
we denote an @xmath8-dimensional manifold with faces @xmath86 with its characteristic function @xmath120 ( resp .
omnioriented characteristic function @xmath129 ) by @xmath147 ( resp .
@xmath148 ) .
let @xmath101 and @xmath102 be manifolds with faces and @xmath149 ( resp .
@xmath150 ) and @xmath151 ( resp .
@xmath152 ) be their ( resp .
omnioriented ) characteristic functions , respectively .
assume that @xmath153 and it is induced by the bijective map @xmath154 .
denote its restriction onto the set of facets as @xmath155 then , we call that @xmath156 and @xmath157 are _ combinatorially equivalent _ if the following diagram commutes : @xmath158^{f } \ar[r]^{\lambda_{1 } } & { \mathfrak{t}}_{{{\mathbb{z}}}}/\{\pm 1\ } \ar[d]^{id } \\ & \mathcal{f}(q_{2 } ) \ar[r]^{\lambda_{2 } } & { \mathfrak{t}}_{{{\mathbb{z}}}}/\{\pm 1\ } } \ ] ] similarly , @xmath159 and @xmath160 are _ combinatorially equivalent _ if the following diagram commutes : @xmath158^{f } \ar[r]^{\quad \lambda_{\mathcal{o}_{1 } } } & { \mathfrak{t}}_{{{\mathbb{z } } } } \ar[d]^{id } \\ & \mathcal{f}(q_{2 } ) \ar[r]^{\quad \lambda_{\mathcal{o}_{2 } } } & { \mathfrak{t}}_{{{\mathbb{z } } } } } \ ] ] note that the characteristic function @xmath120 can be obtained by ignoring sings from the omnioriented characteristic function @xmath129 ; we call such @xmath120 an _ induced characteristic function _ from @xmath129 . on the other hand , by choosing a sign for each facet , we can obtain an omnioriented characteristic function @xmath129 from the characteristic function @xmath120 ; we call such @xmath129 an _ induced oriented characteristic function _ from @xmath120
. therefore , we have the following lemma : [ omni->nonomni ] if @xmath159 and @xmath160 are combinatorially equivalent , then their induced @xmath156 and @xmath157 are also combinatorially equivalent .
if @xmath156 and @xmath157 are combinatorially equivalent , then there are induced omnioriented characteristic functions @xmath150 and @xmath152 such that @xmath159 and @xmath160 are combinatorially equivalent .
now we may introduce one of the key facts to prove our main theorem ( see ( * ? ? ?
* theorem 1.3 , theorem 6.1 ) ) : [ wi2 ] let @xmath106 and @xmath107 be a @xmath0-dimensional simply connected torus manifold with @xmath2 and @xmath156 and @xmath157 be their orbit spaces with characteristic functions .
then , the following three statements are equivalent : 1 .
@xmath156 and @xmath157 are combinatorially equivalent ; 2 . @xmath106 and @xmath107 are equivariantly homeomorphic ; 3 .
@xmath106 and @xmath107 are equivariantly diffeomorphic .
therefore , by corollary [ disk ] and theorem [ wi2 ] , in order to classify all @xmath0-dimensional simply connected torus manifolds with @xmath2 , it is enough to classify all @xmath147 s up to combinatorial equivalent , where @xmath86 is a @xmath1-dimensional disk equipped with the structure of a manifold with faces .
let @xmath59 be an omnioriented locally standard @xmath7-dimensional torus manifold and @xmath148 be its orbit space with an omnioriented characteristic function .
from the one - skeleton of @xmath148 , we can define a labelled graph , called a _
torus graph_. one of the key arguments to prove the main theorem is to classify all possible torus graphs ( see section [ sect7 ] ) . to do that , in this section , we recall a torus graph defined by maeda - masuda - panov @xcite .
let @xmath161 be the graph of @xmath86 .
let @xmath162 be its vertices and @xmath163 be its oriented edges , i.e. , we distinguish two edges @xmath164 and @xmath165 . for @xmath166 , we denote the set of outgoing edges from @xmath78 as @xmath167 . because @xmath86 is an @xmath8-dimensional manifold with faces , @xmath168 and each edge
@xmath169 is a connected component of @xmath170 , for some @xmath171 .
moreover , for @xmath166 which is one of two vertices on @xmath172 , there is another facet @xmath173 such that @xmath174 is a connected component of @xmath175 . in other words ,
@xmath176 may be regarded as a normal facet of @xmath169 on @xmath166 .
put @xmath177 . then , there exists unique @xmath178 such that @xmath179 where @xmath180 represents the pairing of @xmath181 and @xmath182 .
therefore , by this way , we can define a map @xmath183 from the omnioriented characteristic function @xmath129 .
this map @xmath184 is called an _ axial function _ on @xmath161 .
we call the labelled graph @xmath185 a _ torus graph _ induced from @xmath148 ( or equivalently @xmath59 ) .
we denote such a torus graph as @xmath186 ( or @xmath187 ) .
we can easily check the following proposition by definition of torus graphs ( also see @xcite ) : [ connection ] let @xmath185 be a torus graph induced from @xmath148 .
then , @xmath161 is an @xmath8-valent regular graph , i.e. , @xmath168 for all @xmath166 , and @xmath185 satisfies the following conditions : 1 .
@xmath188 , where @xmath189 is the orientation reversed edge of @xmath172 ; 2 .
@xmath190 spans @xmath191 for all vertices @xmath166 ; 3 .
there is a bijection @xmath192 for all edges whose initial vertex is @xmath78 and terminal vertex is @xmath193 such that the following conditions hold : 1 .
@xmath194 ; 2 .
@xmath195 ; 3 .
@xmath196 for all @xmath197 .
we call @xmath198 a _ connection _ on @xmath185 . the original torus graph ( induced from an omnioriented torus manifold ) is defined by using the _ tangential representations _ , see @xcite , @xcite .
the definition of torus graph as above is essentially same with this definition . in @xcite , motivated by the gkm graph introduced by guillemin - zara in @xcite , an @xmath8-valent graph @xmath161 with a label @xmath183 which satisfies three conditions in proposition [ connection ]
is called an _ ( abstract ) torus graph _ ( i.e. , there might be no geometric objects which define @xmath185 ) .
we next define the equivalence relation between two torus graphs .
we call the map @xmath199 a _ graph isomorphism _ , if the restricted map @xmath200 and @xmath201 are bijective and the following map commutes : @xmath202^{\pi_{v_{1 } } } \ar[r]^{f|_{e } } & e(\gamma_{2 } ) \ar[d]^{\pi_{v_{2 } } } \\ & v(\gamma_{1 } ) \ar[r]^{f|_{v } } & v(\gamma_{2 } ) } \ ] ] where @xmath203 is the map projecting onto the initial vertex , i.e. , @xmath204 . in other words
, the bijection @xmath205 preserves the edges .
now we may define the equivalence relation .
[ def_torus - graph ] let @xmath206 and @xmath207 be torus graphs .
we say @xmath206 and @xmath207 are _ equivalent _ if there is a graph isomorphism @xmath208 such that the following diagram commutes : @xmath202^{f|_{e } } \ar[r]^{\quad \mathcal{a}_{1 } } & { \mathfrak{t}}_{{{\mathbb{z}}}}^ { * } \ar[d]^{id } \\ & e(\gamma_{2 } ) \ar[r]^{\quad \mathcal{a}_{2 } } & { \mathfrak{t}}_{{{\mathbb{z}}}}^ { * } } \ ] ] assume @xmath209 .
let @xmath210 be the set of @xmath71-valent torus subgraphs in @xmath185 , i.e. , @xmath71-valent subgraphs in @xmath161 closed under the connection @xmath211 , where @xmath212 and we define @xmath213 .
then , the set @xmath214 admits the structure of a simplicial poset by inclusion ( see @xcite ) .
we denote this structure by @xmath215 .
let @xmath97 be the face poset of @xmath86 ( see section [ sect2.2 ] ) and @xmath216 be the set of all @xmath71-dimensional faces , where @xmath212 and @xmath217 .
then , each element of @xmath210 is nothing but the graph of an element in @xmath216 .
this implies that the poset @xmath215 is equivalent to the poset @xmath218 .
therefore , we have the following lemma : [ equiv<->equiv ] the following two statements are equivalent : 1 . two manifolds with faces with omnioriented characteristic functions @xmath159 and @xmath160 are combinatorially equivalent ; 2 .
their induced torus graphs @xmath219 and @xmath220 are equivalent . by lemma [ omni->nonomni ] , theorem [ wi2 ] and lemma [ equiv<->equiv ]
, we have the following corollary : [ key2 ] let @xmath221 and @xmath222 be @xmath0-dimensional simply connected torus manifolds with vanishing odd degree cohomology .
then , the following statements are equivalent : 1 .
@xmath221 and @xmath222 are equivariantly diffeomorphic ; 2 .
their orbit spaces , i.e. , @xmath1-dimensional disks with the structures of manifolds with faces , with characteristic functions @xmath223 and @xmath224 are combinatorially equivalent ; 3 .
there are omnioriented characteristic functions @xmath150 and @xmath152 such that their induced @xmath1-valent torus graphs @xmath225 and @xmath226 are equivalent .
therefore , in order to prove our main theorem ( theorem [ main-1 ] ) , it is enough to classify all @xmath1-valent torus graphs @xmath185 , induced from @xmath59 , up to equivalent .
let @xmath10 be a simply connected , @xmath0-dimensional torus manifold with @xmath2 , and @xmath227 be its torus graph induced by some omniorientation . as a preliminary to prove the main theorem ( theorem [ main-1 ] ) , in this section
, we will introduce some of basic torus graphs @xmath185 and their corresponding @xmath0-dimensional torus manifolds @xmath10 .
because the induced torus graphs from @xmath10 are @xmath1-valent , if there is a @xmath1-multiple edge , i.e. , three edges that are incident to the same two vertices , then it follows from proposition [ connection ] that such torus graph must be the torus graph in figure [ tg - s^6 ] .
( we denote the torus graph in figure [ tg - s^6 ] as @xmath228 . ) , where @xmath229 are @xmath230-basis . ] put @xmath231 , @xmath232 and @xmath233 , by the standard basis @xmath234 in @xmath235 .
then , the following equation holds : @xmath236 let @xmath237 be the unit sphere , i.e. , the set @xmath238 such that @xmath239 . define the @xmath36-action on the first three complex coordinates in @xmath3 by @xmath240 where @xmath241 and @xmath242 , @xmath243 , is a one - dimensional complex representation defined by @xmath244 then , by choosing an appropriate omniorientation on @xmath3 , we have that its induced torus graph is equivalent to @xmath228 .
therefore , by using corollary [ key2 ] , we have the following lemma : [ s^6 ] let @xmath59 be an omnioriented @xmath0-dimensional simply connected torus manifold with @xmath2 .
if its induced torus graph is @xmath228 , then @xmath10 is equivariantly diffeomorphic to one of @xmath245 defined by .
assume that a @xmath1-valent torus graph @xmath185 does not have @xmath1-multiple edges but have multiple edges , i.e. , two edges that are incident to the same two vertices .
in this section , we classify the easiest case of such torus graphs . because @xmath161 is a one - skeleton of @xmath1-dimensional manifold with faces @xmath86 , the number of vertices @xmath246 .
assume that @xmath247 .
then , we can easily check that such torus manifold is the one - skeleton of the @xmath1-simplex ( see figure [ tg - cp^3 ] in section [ sect4.3 ] ) or the graph drawn in figure [ tg - bundles ] , say @xmath248 . as is well known that the torus manifold whose torus graph is the one - skeleton of the @xmath1-simplex is equivariantly diffeomorphic to the complex projective space with some @xmath42-action ( see e.g. @xcite , and also see figure [ tg - cp^3 ] in section [ sect4.3 ] ) .
so , we only study the torus manifold which induces the graph @xmath248 .
because @xmath86 is homeomorphic to @xmath96 , we may regard @xmath86 whose one - skeleton is @xmath248 as the product of @xmath249 , where @xmath250 is the @xmath16-dimensional disk and @xmath251 is the interval . by considering all functions on facets of @xmath86 which satisfies
, we can classify all omnioriented characteristic functions @xmath129 on @xmath86 .
then , by the way to induce the axial function @xmath252 from @xmath148 demonstrated in section [ sect3 ] , we can obtain all possible axial functions on @xmath248 as shown in figure [ tg - bundles ] . , where @xmath253 and @xmath254 and @xmath255 are @xmath230-basis of @xmath191 .
] the torus graph @xmath256 in figure [ tg - bundles ] can be induced from an @xmath21-bundle over @xmath41 as follows .
first , by choosing @xmath253 , we may define two free @xmath15-actions on @xmath257 as follows : @xmath258 we denote @xmath259 with the above @xmath260-action by @xmath261 .
note that @xmath262 is diffeomorphic to the @xmath16-sphere @xmath41 , and a complex line bundle over @xmath41 can be denoted by @xmath263 where @xmath264 is the complex @xmath83-dimensional @xmath260-representation space by @xmath71-times rotation for some @xmath25 .
let @xmath265 be the trivial real line bundle over @xmath41 .
take the unit sphere bundle of the following whitney sum of three vector bundles for @xmath254 : @xmath266 then , we obtain the @xmath4-bundle over @xmath5 denoted by @xmath267 for @xmath253 , @xmath254 .
namely , we can identify elements in @xmath268 by @xmath269=[(t^{-1}w , t^{\epsilon}z),(t^{a}x , t^{b}y , r)]\ ] ] for any @xmath270 such that @xmath271 and @xmath272 .
define a @xmath35-action on @xmath268 by @xmath269\mapsto [ ( t_{1}w , z),(t_{2}x , t_{3}y , r)],\ ] ] where @xmath273 . fix an omniorientation on @xmath268 by the induced orientations from @xmath274 .
then , considering the tangential representations around each fixed point , it is easy to check that the induced torus graph is @xmath275 , where @xmath234 are the standard basis of @xmath276
. therefore , by taking the appropriate automorphism of @xmath35 , we can construct each torus graph @xmath256 in figure [ tg - bundles ] from @xmath268 . note that if @xmath277 and @xmath278 , then this is nothing but one of the torus manifolds which appeared in the classifications of torus manifolds with codimension one extended actions in @xcite . by the argument as above and corollary [ key2 ]
, we establish the following lemma : [ s^4-s^2(geom ) ] let @xmath59 be an omnioriented @xmath0-dimensional simply connected torus manifold with @xmath2 . if its induced torus graph has @xmath20-vertices , then @xmath10 is equivariantly diffeomorphic to one of the followings : 1 .
@xmath279 with the standard @xmath35-action up to automorphism of @xmath35 ; 2 .
@xmath268 for some @xmath253 and @xmath254 .
assume that there are no multiple edges in a @xmath1-valent torus graph @xmath185 , i.e. , there are no two edges that are incident to the same two vertices .
a graph @xmath161 is called a _
simple _ if @xmath161 does not have both of multiple edges and loops . in this and section [ sect5 ] , we study simple torus graphs which can be realized as the one - skeleton of manifold with faces homeomorphic to @xmath96 .
the typical example of such torus manifolds whose torus graphs are simple is a quasitoric manifold ( introduced by davis - januszkiewicz in @xcite ( also see @xcite ) ) .
a _ quasitoric manifold _
is defined by a torus manifold whose orbit space is a _ simple convex polytope _ , i.e. , a convex polytope admitting the structure of a manifold with faces .
for example , the complex projective space @xmath280 with the standard @xmath281-action is the quasitoric manifold whose orbit space is the @xmath8-dimensional simplex .
the figure [ tg - cp^3 ] shows the torus graph induced from @xmath282 , i.e. , the omniorientation @xmath283 induced from the standard complex structure on @xmath284 and the standard @xmath42-action on @xmath284 . .
] we next characterize when torus graphs are induced from simple convex polytopes , i.e. , induced from quasitoric manifolds .
the steinitz theorem ( see ( * ? ? ? * chapter 4 ) ) tells us that a graph @xmath161 is the one skeleton of a @xmath1-dimensional convex polytope if and only if @xmath161 is a simple , planner and @xmath1-connected graph , where @xmath161 is called a _
@xmath1-connected _ graph if it remains connected whenever fewer than @xmath1 vertices are removed .
it easily follows from the steinitz theorem that we have the following lemma : [ polytope ] let @xmath86 be a manifold with faces and @xmath161 be its graph .
assume that @xmath86 is homeomorphic to the @xmath1-disk @xmath96 and there are no multiple edges .
then , the following two statements are equivalent : 1 .
@xmath86 is combinatorially equivalent to a @xmath1-dimensional simple convex polytope @xmath285 ; 2 .
@xmath161 is a @xmath1-connected graph .
therefore , together with corollary [ key2 ] , we have the following fact : [ quasi - toru ] let @xmath59 be an omnioriented @xmath0-dimensional simply connected torus manifold with @xmath2 .
then , the following two statements are equivalent : 1 .
@xmath10 is equivariantly diffeomorphic to a quasitoric manifold ; 2 .
its induced torus graph @xmath161 is a @xmath1-connected graph with no multiple edges .
by the arguments in section [ sect4 ] , only the following case remains : the simply connected @xmath0-dimensional torus manifolds with @xmath2 whose induced torus graphs are simple but not @xmath1-connected .
such torus manifolds can be constructed by using the connected sum of `` oriented '' torus graphs .
the purpose of this section is to introduce oriented torus graphs and their connected sum .
we first recall the equivariant connected sum of torus manifolds .
let @xmath106 , @xmath107 be @xmath7-dimensional torus manifolds and @xmath286 , @xmath287 be fixed points . by using the slice theorem , we may take @xmath42-invariant open neighborhoods @xmath288 of @xmath78 and @xmath289 of @xmath193 .
assume that @xmath290 and @xmath291 are equivariantly diffeomorphic .
then , @xmath292 and @xmath293 are equivariantly diffeomorphic to @xmath294 , where @xmath295 with some effective @xmath9-action and @xmath296 with the trivial @xmath9-action for some @xmath297 .
then , we glues these two neighborhood by @xmath298 defined by the identity on @xmath299 and the map @xmath300 on @xmath251 for @xmath301 .
namely , we can glue @xmath302 and @xmath303 by the following identification : @xmath304 the @xmath9-manifold obtained by this way is denoted by @xmath305 or @xmath306 ( if we emphasize fixed points @xmath286 and @xmath287 ) .
because each torus manifold has more than two fixed points , @xmath305 is again a torus manifold .
we call this operation an _ equivariant connected sum_. the following lemma holds : [ lem - easy ] if two torus manifolds @xmath106 and @xmath107 are simply connected and @xmath307 , then @xmath305 is also simply connected and @xmath308 .
it is easy to check the statement by using van - kampen s theorem and the mayer - vietoris exact sequence .
assume that @xmath309 , @xmath310 are @xmath0-dimensional omnioriented simply connected torus manifolds with @xmath307 .
let @xmath206 and @xmath207 be their induced @xmath1-valent torus graphs .
assume that we can glue @xmath311 and @xmath287 by the connected sum .
then , by considering the restriction of @xmath298 in onto @xmath295 , i.e. , the identity map , the axial functions around @xmath312 and @xmath313 must satisfy @xmath314 however , at this stage , torus graphs @xmath206 and @xmath207 do not have the information of orientations of @xmath106 and @xmath107 .
note that , to do connected sum , we need the orientations around @xmath286 and @xmath287 . to encode the orientations around fixed points , we need the following notion
: let @xmath185 be a torus graph .
we call a triple @xmath315 with a map @xmath316 an _ oriented torus graph _ , if @xmath316 satisfies the following condition for all @xmath169 : @xmath317 where @xmath318 is the initial vertex of @xmath169 , i.e. , for @xmath319 , @xmath320 and @xmath321 .
we call a map @xmath322 an _ orientation _ of @xmath185 .
let @xmath59 be an omnioriented torus manifold .
the oriented torus graph @xmath315 of @xmath59 is defined as follows .
let @xmath130 .
then , there exist exactly @xmath8 characteristic submanifolds @xmath323 such that @xmath78 is a connected component of @xmath324 . now
the fixed orientations of @xmath325 determine the decomposition of the tangential representation , i.e. , @xmath326 is determined by fixing the orientations of @xmath325 . on the other hand
, the orientation of @xmath6 determines the orientation of @xmath327 .
so , we define the map @xmath328 by @xmath329 let @xmath330 and @xmath331 be the induced oriented torus graphs from @xmath309 and @xmath310 .
if we can glue @xmath311 and @xmath287 by the connected sum , then both of the relation and the relation @xmath332 hold ( corresponds to that the orientations on @xmath333 and @xmath334 are different ) . note the induced ( oriented ) torus graph by @xmath335 is nothing but the one - skeleton of the connected sum @xmath336 of manifolds with faces , where @xmath337 is the orbit space of @xmath49 , @xmath338 ( also see ( * ? ? ?
* definition 3 ) , ( * ? ? ?
* section 3.1 ) for details about the connected sum of polytopes ) .
therefore , conversely , if @xmath312 and @xmath313 satisfy and , then we can do the _ connected sum of ( oriented ) torus graphs _ between @xmath330 and @xmath331 , say @xmath339 or @xmath340 ( if we emphasize the vertices @xmath312 and @xmath313 ) .
more precisely , @xmath339 is defined as follows ( also see figure [ operations1 ] ) .
1 . @xmath341 ; 2 .
@xmath342 , where @xmath343 for @xmath243 ; 3 .
@xmath344 such that @xmath345 and @xmath346 for @xmath347 and @xmath348 , and @xmath349 and @xmath350 ; 4 .
@xmath351 is defined by @xmath352 for @xmath353 and @xmath354 for @xmath355 .
of torus manifold ( from left to right ) and its inverse @xmath356 ( from right to left ) , where @xmath357 .
here , @xmath358 are @xmath230-basis of @xmath359 and @xmath360 . ]
then , we can easily check that @xmath315 is an oriented torus graph . using corollary [ key2 ] and the arguments as above
, we have the following lemma : [ con - sum canonical model ] let @xmath106 , @xmath107 be @xmath0-dimensional simply connected torus manifolds with @xmath307 , and @xmath330 , @xmath331 be their induced oriented torus graphs from some omniorientations , respectively .
if @xmath361 , then @xmath315 is the oriented torus graph induced from @xmath362 with some omniorientation . by using the connected sum
, we can construct the torus manifolds which are not appeared in section [ sect4 ] .
for example , the following torus manifold is one of such examples : @xmath363 where @xmath364 is the reversed orientation of @xmath279 .
the figure [ others ] shows the torus graph induced from @xmath365 ( see the axial functions on figure [ tg - bundles ] , [ tg - cp^3 ] for details ) .
we can easily check that this graph is @xmath1-valent , simple and planner but not @xmath1-connected ; therefore , due to lemma [ quasi - toru ] , this manifold is not a quasitoric manifold . ) induced from @xmath365 . ]
to prove the main theorem ( theorem [ main-1 ] ) , in this section , we show the following two lemmas .
[ one - pt ] let @xmath86 be a @xmath1-dimensional manifold with faces which is homeomorphic to @xmath96 and @xmath161 be its graph .
then , @xmath366 is connected , for all vertices @xmath166 . because @xmath86 is homeomorphic to the @xmath1-disk @xmath96 ,
@xmath161 may be regarded as a planner graph by the stereographic projection of @xmath367 .
assume @xmath366 is not connected .
because @xmath86 is a @xmath1-dimensional manifold with faces , there are exactly three out - going edges from @xmath78 , say @xmath368 , @xmath369 and @xmath370 .
therefore , we may assume that there exists a connected component @xmath371 in @xmath366 such that @xmath372 but @xmath373 ( see figure [ one - pt - con ] ) . .
the facet @xmath94 has the self - intersection on the edge @xmath374 . ]
since @xmath371 is also a planner @xmath1-valent graph excepts on the vertex @xmath375 ( because @xmath376 ) , there is a @xmath16-valent subgraph in @xmath371 , say @xmath377 , such that @xmath377 splits @xmath367 into two connected components @xmath378 and @xmath379 , where @xmath380 but @xmath381 .
this implies that there is a facet @xmath94 in @xmath86 such that @xmath382 contains @xmath377 and @xmath374 .
however , in this case , @xmath374 must be a self - intersection edge of @xmath94 ( see figure [ one - pt - con ] ) .
this gives a contradiction to that @xmath86 is a manifold with faces .
this establishes the statement . by lemma [ one - pt ] , if @xmath161 is not @xmath1-connected , then there are two vertices @xmath383 such that @xmath384 is not connected but both of @xmath366 and @xmath385 are connected
. more precisely , we have the following lemma : [ separate - points ] let @xmath86 be a @xmath1-dimensional manifold with faces which is homeomorphic to @xmath96 and @xmath161 be its graph .
assume that there are two vertices @xmath383 such that @xmath386 for any facets @xmath94 , i.e , @xmath78 and @xmath193 are not on the same facet @xmath94 .
then , @xmath384 is connected .
assume that @xmath387 are not on the same facet of @xmath86 . because @xmath86 is a manifold with faces , there are
mutually distinct facets @xmath388 such that @xmath174 is a component of @xmath389 and @xmath390 is that of @xmath391 and we can take vertices @xmath392 and @xmath393 such that @xmath394 and @xmath395 are all outgoing edges from @xmath78 and @xmath193 , for @xmath243 .
take two vertices @xmath396 from @xmath397 . by lemma [ one - pt ]
, @xmath398 is connected .
so there is a path @xmath28 from @xmath399 to @xmath400 in @xmath398 .
if @xmath28 does not go through @xmath78 , then @xmath399 and @xmath400 are connected in @xmath397 .
assume that this path @xmath28 goes through @xmath78 .
then @xmath28 goes through exactly two vertices @xmath401 , @xmath402 ( we may assume @xmath375 and @xmath403 ) .
moreover , one of the facets @xmath404 , say @xmath405 , contains both of @xmath375 , @xmath403 .
note that @xmath405 corresponds to the @xmath16-valent subgraph in @xmath161 .
therefore , we can take the path @xmath406 connecting @xmath375 , @xmath403 on @xmath405 which is not the path @xmath407 . because @xmath387 are not on the same facet , in particular @xmath408
, the path @xmath406 does not contain @xmath193 .
hence , the connected subgraph @xmath409 contains both of @xmath410 but does not contain both of @xmath411 .
thus , we can take the path @xmath412 from @xmath399 to @xmath400 in @xmath413 .
this establishes that @xmath397 is connected . in summary , by lemma [ one - pt ] and [ separate - points ] , we have the following fact . [ key3 ]
let @xmath161 be a one - skeleton of @xmath1-dimensional manifold with faces @xmath86 .
then , for all @xmath166 , @xmath366 is connected .
furthermore , if @xmath384 is not connected , then @xmath411 is on the same facet .
the main theorem of this paper can be stated as follows : [ main-1 ] let @xmath6 be a simply connected , @xmath0-dimensional torus manifold with @xmath2 . then , @xmath6 is equivariantly diffeomorphic to one of the following manifolds : 1 .
@xmath414 with a torus action induced from a ( faithful ) representation of @xmath35 on @xmath415 ; 2 .
@xmath0-dimensional quasitoric manifold @xmath416 ; 3 .
@xmath21-bundle over @xmath41 which is equivariantly diffeomorphic to @xmath268 for some @xmath417 , or otherwise , there are some @xmath0-dimensional quasitoric manifolds @xmath418 , for some @xmath419 , and some @xmath21-bundles over @xmath41 , say @xmath420 ( for some @xmath421 , @xmath422 and @xmath423 ) , such that @xmath6 is equivariantly diffeomorphic to @xmath424 where @xmath425 represents the equivariant connected sum around fixed points , @xmath426 for @xmath427 , and the case @xmath428 means that there is no @xmath418 factor . in this final section ,
we prove theorem [ main-1 ] .
let @xmath6 be a simply connected , @xmath0-dimensional torus manifold with @xmath2 , @xmath86 be its orbit space which is homeomorphic to @xmath96 and @xmath187 be its induced oriented torus graph ( we omit the orientation ) . because @xmath429 is a one - skeleton of a manifold with faces which is homeomorphic to @xmath96 , it is easy to check that @xmath430 .
if @xmath431 , by lemma [ s^6 ] , we have that @xmath6 is equivariantly diffeomorphic to @xmath40 , i.e. , the statement ( 1 ) .
if @xmath432 , it follows from lemma [ s^4-s^2(geom ) ] that @xmath6 is equivariantly diffeomorphic to a quasitoric manifold @xmath284 or @xmath268 for some @xmath253 , @xmath254 , i.e. , the statement ( 2 ) or ( 3 ) occurs .
so we may only prove the case when @xmath433 .
we first claim the following lemma : [ 1-claim ] assume that @xmath433 and there is a multiple edge in @xmath429 .
then , @xmath187 can be decomposed into the connected sum of the following torus graphs : @xmath434 or @xmath435 where @xmath436 is a torus graph without multiple edges and @xmath420 for @xmath437 .
assume two vertices @xmath78 , @xmath193 are connected by a multiple edge , i.e. , two edges ( see the bottom graph in figure [ essential discussion to prove this theorem ] ) . as any generators in @xmath359 and @xmath438 and @xmath439 . here
, the bottom graph is @xmath187 and the left upper graph is @xmath440 and the right upper graph is @xmath441 .
note that if we fix the orientation of @xmath187 then the orientations of @xmath440 and @xmath441 are automatically determined . ] then , by the connection of the torus graph ( see proposition [ connection ] ) , it is easy to check that the axial functions around the vertex @xmath399 of the bottom graph in figure [ essential discussion to prove this theorem ] satisfy the axial functions expressed in figure [ essential discussion to prove this theorem ] , where we can take @xmath442 , @xmath443 , @xmath28 as @xmath230-basis @xmath359 . in this case , we can do ( inverse ) connected sum such as expressed in figure [ essential discussion to prove this theorem ] ( from the bottom to the top in figure [ essential discussion to prove this theorem ] ) .
then , the induced torus graph @xmath187 is decomposed into two induced torus graphs @xmath440 and @xmath441 , where @xmath444 is some simply connected @xmath0-dimensional torus manifold with @xmath445 by lemma [ lem - easy ] .
namely , we have @xmath446 if there is no multiple edges in @xmath447 , then we may put @xmath448 .
assume that there is a multiple edge in @xmath447 .
if there are only @xmath20 vertices in @xmath447 , then we may put @xmath444 as @xmath449 by lemma [ s^4-s^2(geom ) ] .
when there are more than @xmath20 vertices in @xmath447 , with iterating the arguments as above , finally we establish the statement of this lemma .
therefore , to prove theorem [ main-1 ] , it is enough to show the following lemma : [ 2-claim ] assume that @xmath433 and there are no multiple edges in @xmath429 .
then , @xmath187 can be decomposed into the connected sum of the following torus graphs : @xmath450 where @xmath451 for @xmath419 is the torus graph induced from a quasitoric manifold @xmath418 , and @xmath420 for @xmath452 .
if @xmath453 is @xmath1-connected , then it follows from lemma [ quasi - toru ] that the statement holds , i.e. , @xmath454 .
therefore , we may assume @xmath161 is not @xmath1-connected .
in this case , by corollary [ key3 ] , there is a @xmath16-valent torus subgraph @xmath455 such that some @xmath456 satisfy that @xmath397 is not connected .
if @xmath94 is a triangle ( i.e. , @xmath457 ) , with the method similar to that demonstrated in the proof of lemma [ one - pt ] , we have that there is a face in @xmath86 which has the self - intersection edge .
this gives a contradiction to that @xmath86 is a manifold with faces .
therefore , we may assume @xmath458 .
we first assume that @xmath164 is an edge of @xmath94 . then , there are two graphs @xmath371 and @xmath459 which are the connected components of @xmath397 expressed in figure [ edge - con ] . and
@xmath193 from @xmath161 instead of @xmath78 and @xmath193 , the graph is also disconnected . ]
if we remove the two vertices @xmath399 and @xmath193 from @xmath161 instead of @xmath78 , @xmath193 , where @xmath460 such that @xmath461 is an edge , then @xmath462 is also not connected ( see figure [ edge - con ] ) . therefore , from now , we may assume the followings : 1 .
@xmath463 satisfy @xmath397 is not connected ; 2 . @xmath464
there is a @xmath16-valent torus subgraph ( facet ) @xmath94 with @xmath458 in @xmath161 such that @xmath456 .
we call such facet @xmath94 a
_ singular facet_. let @xmath94 be a singular facet .
assume @xmath465 .
when @xmath465 , where @xmath466 and @xmath467 . here
, @xmath468 is a facet which intersects with @xmath94 on @xmath461 and @xmath469 . ] in this case , by the similar argument just before , we may take @xmath387 as in the position of figure [ 6-vertices ] , i.e. , @xmath78 and @xmath193 are on two separated edges @xmath470 and @xmath471 which are common edges of two facets @xmath94 and @xmath468 in figure [ 6-vertices ] ( note that @xmath399 and @xmath400 might be connected by an edge ) .
moreover , by considering the omnioriented characteristic functions of the facets @xmath94 and @xmath468 , we may take the axial functions around the facet @xmath94 as in figure [ 6-vertices ] . by taking an appropriate orientation
, we can do the connected sum as in figure [ 6-vertices2 ] ; here we denote the ( oriented ) torus graph containing @xmath371 as @xmath472 and that containing @xmath459 as @xmath473 . in figure [ 6-vertices ]
splits into two torus graphs @xmath472 ( upper ) and @xmath473 ( lower ) .
here , we omit the axial functions around the vertices @xmath474 because they are exactly same with that of figure [ 6-vertices ] . ]
the torus graph obtained by this connected sum is nothing but the torus graph @xmath185 in figure [ 6-vertices ] .
note that @xmath475 is simple and planner , and @xmath476 is just planner . with the method similar to that demonstrated in figure [ essential discussion to prove this theorem ]
, @xmath473 can be obtained from the connected sum of @xmath256 and the simple , planner graph @xmath477 ( containing @xmath459 ) , where @xmath256 is one of the torus graphs ( by taking the appropriate axial functions ) in figure [ tg - bundles ] .
namely , the torus graph in figure [ 6-vertices ] can be obtained from the following connected sum : @xmath478 here , it is easy to check that @xmath479 consists of @xmath480 and the other two facets , say @xmath481 and @xmath482 ( induced from @xmath94 and @xmath468 in @xmath161 ) . because of figure [ 6-vertices2 ] ,
the number of vertices of other two facets @xmath481 and @xmath482 are reduced ; in particular , the number of vertices of the facet @xmath481 induced from the singular facet @xmath94 is strictly less than @xmath0 .
if both of @xmath472 and @xmath477 are @xmath1-connected , then these torus graphs are induced from quasitoric manifolds , i.e , the statements of lemma [ 2-claim ] hold .
assume that @xmath472 is not @xmath1-connected .
then , by the arguments before , there is a singular facet @xmath94 in @xmath472 .
if @xmath465 , then @xmath472 also decomposes into @xmath483 by using the similar arguments explained in figure [ 6-vertices2 ] .
iterating this arguments , we may reduce all singular facets with @xmath465 .
more precisely , we may decompose @xmath185 in figure [ 6-vertices ] into @xmath484 where @xmath485 , for @xmath423 , is a torus graph in figure [ tg - bundles ] and @xmath486 , for @xmath487 , is a @xmath1-valent simple and planner torus graph which satisfies one of the followings : * @xmath1-connected ( in this case , @xmath486 is induced from a quasitoric manifold ) ; * otherwise , all singular facets @xmath94 satisfy @xmath488 or @xmath489 .
assume that the number of vertices in every singular facet of the torus graph @xmath185 is less than or equal to @xmath489 .
then , such torus graph is one of torus graphs expressed in figure [ 45-vertices ] .
with @xmath490 or @xmath20 . here , @xmath468 is a facet which intersects with @xmath94 on @xmath461 and @xmath469 . ] however , because @xmath161 is the one - skeleton of a manifold with faces and not @xmath1-connected , it is easy to check that there exists the singular facet @xmath468 such that @xmath491 and @xmath492 .
this gives a contradiction .
hence , this case does not occur .
this establishes lemma [ 2-claim ] .
consequently , by lemma [ con - sum canonical model ] , [ 1-claim ] and [ 2-claim ] , we have the statement of theorem [ main-1 ] finally , by theorem [ main-1 ] and the mayer - vietoris exact sequence , we also have the following well - known result . [ 2nd deg ]
let @xmath6 be a simply connected @xmath0-dimensional torus manifold whose cohomology ring is generated by the @xmath16nd degree cohomology .
then , @xmath6 is a quasitoric manifold .
the author would like to thank professors yael karshon and takashi tsuboi for providing him excellent circumstances to do research . | the aim of this paper is to classify simply connected @xmath0-dimensional torus manifolds with vanishing odd degree cohomology .
it is shown that there is a one - to - one correspondence between equivariant diffeomorphism types of these manifolds and @xmath1-valent labelled graphs , called torus graphs introduced by maeda - masuda - panov . using this correspondence and combinatorial arguments
, we prove that a simply connected @xmath0-dimensional torus manifold with @xmath2 is equivariantly diffeomorphic to the @xmath0-dimensional sphere @xmath3 or an equivariant connected sum of copies of @xmath0-dimensional quasitoric manifolds or @xmath4-bundles over @xmath5 . |
theoretical studies of the cmb have shown that the accurate measurement of the cmb anisotropy spectrum @xmath0 with future space missions such as planck will allow for tests of cosmological scenarios and the determination of cosmological parameters with unprecedented accuracy .
nevertheless , some near degeneracies between sets of cosmological parameters yield very similar cmb temperature anisotropy spectra . the measurement of the cmb polarization and the computation of its power spectrum @xcite may lift to some extent some of these degeneracies .
it will also provide additional information on the reionization epoch and on the presence of tensor perturbations , and may also help in the identification and removal of polarized astrophysical foregrounds @xcite .
a successful measurement of the cmb polarization stands as an observational challenge ; the expected polarization level is of the order of @xmath1 of the level of temperature fluctuations ( @xmath2 ) .
efforts have thus gone into developing techniques to reduce or eliminate spurious non - astronomical signals and instrumental noise which could otherwise easily wipe out real polarization signals . in a previous paper @xcite , we have shown how to configure the polarimeters in the focal plane in order to minimize the errors on the measurement of the stokes parameters . in this paper , we address the problem of low frequency noise . low frequency noise in the data streams can arise due to a wide range of physical processes connected to the detection of radiation . @xmath3 noise in the electronics , gain instabilities , and temperature fluctuations of instrument parts radiatively coupled to the detectors , all produce low frequency drifts of the detector outputs . the spectrum of the total noise can be modeled as a superposition of white noise and components behaving like @xmath4 where @xmath5 , as shown in fig .
[ spectrebruit ] .
this noise generates stripes after reprojection on maps , whose exact form depends on the scanning strategy . if not properly subtracted , the effect of such stripes is to degrade considerably the sensitivity of an experiment . the elimination of this `` striping '' may be achieved using redundancies in the measurement , which are essentially of two types for the case of planck : * each individual detector s field of view scans the sky on large circles , each of which is covered consecutively many times ( @xmath6 ) at a rate of about @xmath7 rpm .
this permits a filtering out of non scan - synchronous fluctuations in the circle constructed from averaging the consecutive scans . *
a survey of the whole sky ( or a part of it ) involves many such circles that intersect each other ( see fig .
[ inter ] ) ; the exact number of intersections depends on the scanning strategy but is of the order of @xmath8 for the planck mission : this will allow to constrain the noise at the intersection points .
one of us @xcite has proposed to remove low frequency drifts for unpolarized data in the framework of the planck mission by requiring that all measurements of a single point , from all the circles intersecting that point , share a common sky temperature signal .
the problem is more complicated in the case of polarized measurements since the orientation of a polarimeter with respect to the sky depends on the scanning circle .
thus , a given polarimeter crossing a given point in the sky along two different circles will not measure the same signal , as illustrated in fig .
[ deuxpol ] .
the rest of the paper is organized as follows : in sect .
[ noise ] , we explain how we model the noise and how low frequency drifts transform into offsets when considering the circles instead of individual scans . in sect .
[ skypol ] , we explain how polarization is measured .
the details of the algorithm for removing low - frequency drifts are given in sect .
we present the results of our simulations in sect .
[ simul ] and give our conclusions in sect .
[ resul ] .
as shown in fig . [ spectrebruit ] , the typical noise spectrum expected for the planck high frequency instrument ( hfi ) features a drastic increase of noise power at low frequencies @xmath9 .
we model this noise spectrum as : @xmath10 the knee frequency @xmath11 is defined as the frequency at which the power spectrum due to low frequency contributions equals that of the white noise .
the noise behaves as pure white noise with variance @xmath12 at high frequencies .
the spectral index of each component of the low - frequency noise , @xmath13 , is typically between 1 and 2 , depending on the physical process generating the noise .
the fourier spectrum of the noise on the circle obtained by combining @xmath14 consecutive scans depends on the exact method used .
the simplest method , setting the circle equal to the average of all its scans , efficiently filters out all frequencies save the harmonics of the spinning frequency @xcite . since the noise power mainly resides at low frequencies ( see fig .
[ spectrebruit ] ) , the averaging transforms to first order low frequency drifts into constant offsets different for each circle and for each polarimeter .
this is illustrated in the comparison between figs .
[ stream1sf ] and [ average1sf ] .
more sophisticated methods for recombining the data streams into circles can be used , as @xmath15 minimization , wiener filtering , or any map - making method projecting about @xmath16 samples onto a circle of about @xmath17 points . for simplicity
, we will work in the following with the circles obtained by simple averaging of all its consecutive scans .
we thus model the effect of low frequency drifts as a constant offset for each polarimeter and each circle .
this approximation is excellent for @xmath18 .
the remaining white noise of the @xmath19 polarimeters is described by one constant @xmath20 matrix .
the measurement with one polarimeter of the linear polarization of a wave coming from a direction @xmath21 on the sky , requires at least three measurements with different polarimeter orientations . since the stokes parameters @xmath22 and @xmath23 are not invariant under rotations
, we define them at each point @xmath21 with respect to a reference frame of tangential vectors @xmath24 .
the output signal given by a polarimeter looking at point @xmath21 is : @xmath25 where @xmath26 is the angle between the polarimeter and @xmath27 stokes parameter since no net circular polarization is expected through thomson scattering . ] . in the following , we choose the longitude - latitude reference frame as the fixed reference frame on the sky ( see fig .
[ frame ] ) .
the destriping method consists in using redundancies at the intersections between circle pairs to estimate , for each circle @xmath28 and each polarimeter @xmath29 , the offsets @xmath30 on polarimeter measurements . for each circle intersection
, we require that all three stokes parameters _ in a fixed reference frame _ in that direction of the sky , as measured on each of the intersecting circles , be the same .
a @xmath15 minimization leads to a linear system whose solution gives the offsets . by subtracting these offsets
, we can recover the stokes parameters corrected for low - frequency noise .
we consider a mission involving @xmath31 circles .
the set of all circles that intercept circle @xmath28 is denoted by @xmath32 and contains @xmath33 circles .
for any pair of circles @xmath28 and @xmath34 , we denote the two points where these two circles intersect ( if any ) by @xmath35 . in this notation @xmath28
is the circle currently scanned , @xmath34 the intersecting circle in set @xmath32 , and @xmath36 indexes the two intersections ( @xmath37 indexes the first ( second ) point encountered from the northernmost point on the circle ) so that the points @xmath38 and @xmath35 on the sky are identical .
the stokes parameters at point @xmath39 , with respect to a fixed global reference system , are denoted by a @xmath40vector @xmath41 , with @xmath42 at intersection @xmath39 , the set of measurements by @xmath19 polarimeters travelling along the scanning circle @xmath28 is a @xmath43vector denoted by @xmath44 , and is related to the stokes parameters at this point by ( see eq .
[ mesurepol ] ) : @xmath45 where @xmath46 is the @xmath47 matrix : @xmath48 @xmath49 $ ] is the angle between the orientation of polarimeter @xmath29 and the reference axis in the fixed global reference frame ( see fig .
[ frame ] ) .
the matrix @xmath46 can be factorised as @xmath50 the constant @xmath47 matrix @xmath51 characterizes the geometrical setup of the @xmath19 polarimeters in the focal reference frame : @xmath52 where @xmath53 is the angle between the orientations of polarimeters @xmath29 and @xmath54 , so we have @xmath55 and @xmath56 .
the rotation matrix @xmath57 brings the focal plane to its position at intersection @xmath39 when scanning along circle @xmath28 : @xmath58
to extract the offsets from the measurements , we use a @xmath15 minimization .
this @xmath15 relates the measurements @xmath44 to the offsets @xmath59 and the stokes parameters @xmath41 , using the redundancy condition . in order to take into account the two contributions of the noise ( see sect . [ noise ] ) and of the stokes parameters ( see eq .
[ stokesm ] ) , we model the measurement as : @xmath60 so that we write @xmath61 where @xmath62 is the @xmath63 matrix of noise correlation between the @xmath19 polarimeters .
minimization with respect to @xmath59 and @xmath41 yields the following equations : @xmath64 we can work with a reduced set of transformed measurements and offsets which can be viewed as the stokes parameters in the focal reference frame and the associated offsets which are the @xmath65 dimensional vectors : @xmath66 eqs . and then simplify to : @xmath67 @xmath68 in eq .
can be solved for and the result inserted in eq . .
after a few algebraic manipulations , one gets the following linear system for the offsets @xmath69 as functions of the data @xmath70 : @xmath71^{-1 } \times \\ \!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\ ! \left[\boldsymbol{\delta}_i - \widetilde{\boldsymbol{r}}(i , j,\delta)\,\boldsymbol{\delta}_j - \mathscr{s}_{i , j,\delta } + \widetilde{\boldsymbol{r}}(i , j,\delta)\,\mathscr{s}_{j , i,-\delta}\right ] = 0 , \label{eqgd3}\end{gathered}\ ] ] where the rotation @xmath72 brings the focal reference frame from its position along scan @xmath34 at intersection @xmath39 to its position along scan @xmath28 at the same intersection ( remember that @xmath73 ) . note that @xmath74 . in this linear system , we need to know the measurements of the polarimeters at the points @xmath75 and @xmath76 .
these two points on circles @xmath28 and @xmath34 respectively will unlikely correspond to a sample along these circles .
so we have linearly interpolated the value of the intersection points from the values measured at sampled points . for a fixed circle @xmath28 ,
this is a @xmath77 linear system . in eq .
, @xmath28 runs from @xmath54 to @xmath31 , therefore the total matrix to be inverted has dimension @xmath78 .
however , because the rotation matrices are in fact two dimensional ( see eq . [ arot ] )
, the intensity components @xmath79 of the offsets only enter eq . through their differences @xmath80 so that the linear system is not invertible : the rank of the system is @xmath81 . in order to compute the offsets
, we can fix the intensity offset on one particular scanning circle or add the additionnal constraint that the length of the solution vector is minimized . once the offsets @xmath69 are known , the stokes parameters in the global reference frame @xmath82 at a generic sampling @xmath83 of the circle @xmath28 , labeled by @xmath84 are estimated as @xmath85 where @xmath86 is the rotation matrix which transforms the focal frame stokes parameters into those of the global reference frame .
the quantities @xmath87 are the stokes parameters measured in the focal frame of reference at this point and are simply given in terms of the measurements @xmath88 ( see eq .
[ defg ] ) by : @xmath89 the matrix @xmath90 is the variance matrix of the stokes parameters on circle @xmath28 .
note that this algorithm is totally independent of the pixelization chosen which only enters when reprojecting the stokes parameters on the sphere .
when the polarimeters are uncorrelated with identical noise , the variance matrix reduces to @xmath91 and the matrices @xmath92 can all be written as @xmath93 we have shown @xcite that the polarimeters can be arranged in `` optimized configurations '' , where the @xmath19 polarimeters are separated by angles of @xmath94 .
if the noise level of each of the @xmath19 polarimeters is the same and if there are no correlation between detector noise , then the errors of the stokes parameters are also decorrelated and the matrix @xmath95 has the simple form : @xmath96 because this matrix commutes with all rotation matrices @xmath97 , eq .
simplifies further to @xmath98 where the sum over @xmath36 is explicit on the left side of the equation and we have defined an average error @xmath99 along circle @xmath28 by @xmath100 and where the rotation matrix @xmath97 is defined by eq . .
rotations @xmath97 and @xmath101 correspond to the two intersections between circles @xmath28 and @xmath34 .
eq . can be simplified further .
we can separate the @xmath69 and the @xmath70 into scalar components related to the intensity : @xmath102 and 2-vectors components related to the polarization : @xmath103 .
we obtain then two separate equations , one for the intensity offsets @xmath104 , which is exactly the same as in the unpolarized case ( see [ appendixa ] ) : @xmath105 and one for the polarization offsets @xmath106 : @xmath107 where @xmath108 is the angle of the rotation that brings the focal reference frame along scan @xmath34 on the focal reference frame along scan @xmath28 at intersection @xmath39 .
the two dimensional matrix @xmath109 is the rotation sub - matrix contained in @xmath97 ( see eq .
[ frot ] ) .
note that all mixing between polarization components have disappeared from the left side of eq . ) .
therefore we are left with two different @xmath110 matrices to invert in order to solve for the offsets @xmath69 , instead of one @xmath111 matrix . as in eq .
, the linear system in eq . involves differences between the offsets @xmath79 , the matrix is not invertible , and we find the solution in the same way as in the general case . on the other hand , for the polarized offsets @xmath112 , the underlying matrix of eq .
is regular as expected .
we now discuss how we test the method using numerical simulations . for each simulated mission
, we produce several maps .
the first , which we use as the standard `` reference '' of comparison , is a projected map of a mission with only white noise , @xmath113 .
the remaining maps include @xmath3 plus white noise streams with @xmath114 , @xmath115 .
the first of these is an `` untreated '' image which is projected with no attempt to remove striping affects . in the second `` zero - averaged '' map
, we attempt a crude destriping by subtracting its average to each circle .
the final `` destriped '' map is constructed using the algorithm in this paper .
we subtract the input maps ( @xmath116 , @xmath22 and @xmath23 ) from the final maps in order to get maps of noise residuals .
note that in case of a zero signal sky , setting the average of each circle to zero is better than destriping by nature because the offsets are only due to the noise . with a real sky ,
both signal and noise contribute to the average so that zeroing circle not only removes the noise but also the signal .
giard et al @xcite have attempted to refine their method by fitting templates of the dipole and of the galaxy before subtracting a baseline from each circle .
they concluded that an additional destriping ( they used the algorithm of delabrouille @xcite ) is needed . in order to test its efficiency ,
the destriping algorithm has been applied to raw data streams generated from simulated observations using various circular scanning strategies representative of a satellite mission as planck , different `` optimized configurations '' , and various noise parameters .
the resulting maps were then compared with input and untreated maps to test the quality of the destriping . the input temperature ( @xmath116 ) maps are the sum of galaxy , dipole , and a randomly generated standard cdm anisotropy map ( we used healpix and cmbfast ) .
similarly , the polarization maps @xmath22 and @xmath23 are the sum of the galaxy and cmb polarizations .
the cmb polarization maps are randomly generated assuming a standard cdm scenario . for the galactic polarization maps , we constructed a random , continuous and correlated vector field defined on the 2-sphere with a correlation length of @xmath117 and a maximum polarization rate of @xmath118 ( @xmath119 gave similar results ) .
given the temperature map of the sky ( not including cmb contribution ) , we can thus construct two polarization maps for @xmath22 and @xmath23 .
we first consider the case of destriping pure white noise and check that the destriping algorithm does not introduce spurious structure .
once this is verified , we apply the destriping algorithm to low frequency noise .
we find that the quality of the destriping is significantly dependent on @xmath120 only .
to demonstrate visually the quality of the destriping , we produce projected sky maps with the input galaxy , dipole and cmb signal subtracted . for temperature maps , we can compare figs .
7,8 , and 9 .
the eye is not able to see any differences between the white noise map and the noise residual on the destriped map .
we will see in the following how to quantify the presence of structures . for the `` zero - averaged circles '' map ,
the level of the structure is very high and make it impossible to compute the power spectrum of the cmb ( see fig .
13 ) . for the @xmath22 stokes parameter , fig .
11 shows the destriped map for @xmath121 .
12 shows a map where the offsets are calculated as the average of each circle .
the maps for @xmath23 are very similar . as for the @xmath116 maps ,
the destriped map is very similar to the white noise map .
there exist some residual structure on the `` zero - averaged circles '' map .
to assess quantitatively the efficiency of the destriping algorithm , we have first studied the power spectra @xmath122 , @xmath123 , @xmath124 , @xmath125 and @xmath126 calculated from the @xmath116 , @xmath22 and @xmath23 maps @xcite .
the reference sensitivity of our simulated mission is evaluated by computing the average spectra of 1000 maps of reprojected mission white noise .
this reference sensitivity falls , within sample variance , between the two dotted lines represented in figs . 13 , 14 , 15 and 16 and below the dotted line in figs .
17 and 18 . figs . 13 and 14 show the spectra @xmath127 corresponding to @xmath128 respectively , for the @xmath129 field .
similarly , figs . 15 and 16 are the spectra @xmath130 .
the @xmath131 field is not represented because it is very similar to @xmath132 .
17 and 18 represent the correlation between @xmath132 and @xmath129 : @xmath133 . for @xmath134
, we see that we are able to remove very efficiently low frequency drifts in the noise stream : the destriped spectra obtained are compatible with the white spectrum ( within sample variance ) .
similar quality destriping is achieved for any superposition of @xmath3 , @xmath135 and white noise , provided that the knee frequency is lower than or equal to the spin frequency . in the case of @xmath136 , the method as implemented here leaves some striping noise on the maps at low values of @xmath137 .
modeling the noise as an offset is no longer adequate and a better model of the averaged low - frequency noise is required ( superposition of sine and cosine functions for instance ) , or a more sophisticated method for constructing one circle from 60 scans .
we again note that the value of the ratio @xmath138 for both planck hfi and lfi is likely to be very close to unity ( in fig .
[ spectrebruit ] , @xmath139 hz and @xmath140 hz ) . to quantify the presence of stripes in the maps of residuals
, we can compute the value of the `` striping '' estimator @xmath141 , because stripes tend to appear as structure grossly parallel to the iso - longitude circles . in the case of pure white noise with a uniform sky coverage ,
this value is 1 . here , because of the scanning strategy , the sky coverage is not uniform and the value of this estimator is greater than 1 , showing that it is not specific of the striping . in order to get rid of the effect of non - uniform sky coverage , we express the estimator @xmath141 in units of @xmath141 for the white noise .
this new estimator is specific to the striping .
the results in table [ table1 ] show the improvement achieved by the destriping algorithm although the result is still not perfect .
.values of @xmath141 as a function of @xmath142 in units of @xmath143 we have checked that the systematic difference between the zero - averaged @xmath132 and @xmath131 fields is randomly in favor of @xmath132 and @xmath131 depending on the particular sky simulation . [ cols="^,^,^,^,^",options="header " , ]
[ [ comparison - with - other - methods . ] ] comparison with other methods .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + although no other method has yet been developped specifically for destriping polarized data , many methods exist for destriping unpolarized cmb data , which could be adapted to polarized data as well .
we first comment on the classical method which consists in modelling the measurement as @xmath144 where @xmath145 is the so - called `` pointing matrix '' , @xmath129 a vector of temperatures in pixels of the sky , @xmath146 the data and @xmath147 the noise .
the problem is solved by inversion , yielding an estimator of the signal : @xmath148^{-1 } a^t n^{-1 } \ , m,\ ] ] where @xmath149 is the noise correlation matrix and @xmath150 is the transposed matrix of @xmath145 .
this method can be extended straightforwardly to polarized measurements , at the price of extending by a factor of @xmath151 the size of the matrix @xmath152 , by 3 that of vector @xmath153 ( replaced by @xmath154 ) , and by @xmath19 that of the data stream ( remember that @xmath19 is the number of polarimeters ) .
the implementation of this formally simple solution may turn into a formidable problem when megapixel maps are to be produced .
numerical methods have been proposed by a variety of authors @xcite , that use properties of the noise correlation matrix ( symmetry , band - diagonality ) and of the pointing matrix ( sparseness ) . such methods , however , rely critically on the assumption that the noise is a gaussian , stationary random process , which has been a reasonable assumption for cmb mission as cobe where the largest part of the uncertainty comes from detector noise , but is probably not so for sensitive missions such as planck . our method requires only inverting a @xmath155 matrix where @xmath31 is the number of circles involved , and does not assume anything on the statistical properties of low frequency drifts .
it just assumes a limit frequency ( the knee frequency ) above which the noise can be considered as a white gaussian random process .
another interesting method is the one that has been used by ganga in the analysis of firs data @xcite , which is itself adapted from a method developed originally by cottingham @xcite . in that method ,
coefficients for splines fitting the low temperature drifts are obtained by minimising the dispersion of measurements on the pixels of the map .
such a method , very similar in spirit to ours , could be adapted to polarization .
splines are natural candidates to replace our offsets in refined implementations of our algorithm . here , we have assumed that the averaged noise can be modeled as circle offsets plus white noise ( eq . [ eq : offsetsmodel ] ) , i.e. that the noise between different measurements from the same bolometer is uncorrelated after removal of the offset .
this allowed us to simplify the @xmath15 to that shown in eq . .
in reality , the circular offsets do not completely remove the low frequency noise and there does remain some correlation between the measurements .
the amount of correlation is directly related to the value of @xmath138 ; the smaller @xmath138 the smaller the remaining correlation . figs .
13 - 18 already contain the errors induced from the fact that we did not include these correlations in the covariance matrix , and thus demonstrate that the effect is small for @xmath156 .
[ [ conclusion . ] ] conclusion .
+ + + + + + + + + + + the destriping as implemented in this paper removes low frequency drifts up to the white noise level provided that @xmath157 . for larger @xmath11
, the simple offset model for the averaged noise could be replaced with a more accurate higher - order model that destripes to better precision provided the scan strategy allows to do so , as discussed in delabrouille et al @xcite .
we are currently working on improving our algorithm to account for these effects .
however , despite the shortcomings of our model , it still appears to be robust for small @xmath11 and can serve as a first order analysis tool for real missions .
in particular , our technique can not only be used for the planck hfi and lfi , but can also be adopted for other cmb missions with circular scanning strategies , such as cosmosomas for instance @xcite .
we would like to acknowledge our referee s very useful suggestions .
we give here the formul for the simpler case of destriping temperature measurements with bolometers .
the assumptions are the same as in the polarized case and we adopt the same notation for the common quantities . instead of polarimeters , we have @xmath19 bolometers . since the measurement is no longer dependent on the orientation of the bolometer , the model of the measurement is given by : @xmath158 where @xmath159 is the @xmath43vector made of measurements by the @xmath19 bolometers , @xmath160 is an @xmath43vector containing the offsets for the ith circle and @xmath161 is the temperature in the direction of the intersection point labeled by @xmath39 .
@xmath162 is an @xmath19-vector , corresponding to the @xmath51 matrix of the polarized case .
the @xmath15 can be written as : @xmath163 in this case , the physical quantity uniquely defined at an intersection point is the temperature of the sky at this point .
the constraint used here for removing low - frequency noise is then : @xmath164 given this relation , the minimization of the @xmath15 with respect to @xmath59 and @xmath165 leads to the linear system : @xmath166 where @xmath167 corresponds to the @xmath92 matrix of the polarized case , @xmath168 corresponds to the @xmath70 local stokes parameters of the polarized case and the scalar @xmath169 corresponds to the 3-vector @xmath170 of the polarized case .
temperature offsets appear through their differences @xmath171 so this linear system is not invertible and we can use the same methods to invert the system than in the polarized case .
the size of the matrix to invert is @xmath110 .
+ after evaluation of the offsets @xmath170 , one can recover the value of temperature for any sample @xmath83 along circle @xmath28 : @xmath172
the results presented in this paper correspond to the `` optimized configuration '' involving @xmath65 polarimeters and to an angular step of @xmath173 : the angle between two consecutive samples along a circle is @xmath173 and there is one circle every @xmath173 . in this case , the number of circles is @xmath174 and the number of samples on each circle is @xmath175 .
the spin axis has a sinusoidal motion around the ecliptic plane with an amplitude of @xmath176 and with @xmath177 cycles during the mission .
the aperture angle of the circles is @xmath178 .
the simulated noise has @xmath114 with @xmath179 .
all the maps ( cmb , dipole , galaxy , simulation ) are healpix maps with 196608 pixels of @xmath181 .
only 12 pixels ( @xmath182 of the map ) are not seen by the mission : their values are set to the average of the map .
the signal maps have been smoothed by a gaussian beam with a fwhm set to @xmath183 .
we have run other simulations with `` optimized configuration '' involving @xmath65 or @xmath184 polarimeters , and with cycloidal , sinusoidal or anti - solar spin axis trajectories ( see bersanelli et al @xcite and the planck web page[multiblock footnote omitted ] for additionnal information about proposed scanning strategies ) .
the aperture angle of the circles have been taken in @xmath185 $ ] . in all these cases ,
the results are similar for @xmath156 . for more pessimistic noise cases ,
the choice of the scanning strategy may have a strong impact on the quality of the final maps .
a quantitative study of this point is deffered to a forthcoming publication .
rebolo , r. , gutirrez , c. , watson , r. , and gallegos , j. : 1999 , in _ proceedings of the workshop `` the cmb and the planck mission '' held in santander , to appear in astrophysical letters and communications _ | a major problem in cosmic microwave background ( cmb ) anisotropy mapping , especially in a total - power mode , is the presence of low - frequency noise in the data streams .
if unproperly processed , such low - frequency noise leads to striping in the maps . to deal with this problem ,
solutions have already been found for mapping the cmb temperature fluctuations but no solution has yet been proposed for the measurement of cmb polarization .
complications arise due to the scan - dependent orientation of the measured polarization . in this paper , we investigate a method for building temperature and polarization maps with minimal striping effects in the case of a circular scanning strategy mission such as the planck mission . |
chaotic behavior in hamiltonian systems with many degrees of freedom has been the subject of intense investigation in the last fifty years , see e.g. [ lichtenberg & lieberman , 1991 ; mackay & meiss , 1987 ; wiggins , 1988 ] and @xcite . by degrees of freedom ( dof )
we are referring to the number of canonically conjugate pairs of positions and momentum variables , @xmath20 and @xmath21 respectively , with @xmath22 .
the relevance of these systems to problems of practical concern can not be overemphasized .
their applications range from the stability of the solar system @xcite and the containment of charged particles in high intensity magnetic fields @xcite to the blow up of hadron beams in high energy accelerators @xcite and the understanding of the properties of simple molecules and hydrogen bonded systems @xcite .
one of the most fundamental areas in which the dynamics of multi degree of freedom hamiltonian systems has played ( and continues to play ) a crucial role is the study of transport phenomena in one dimensional ( @xmath23d ) lattices and the role of chaos in providing a link between deterministic and statistical behavior @xcite . in this context
, a lattice of @xmath0 dof is expected , in the thermodynamic limit ( @xmath6 at fixed energy density @xmath12 ) to exhibit chaotic behavior for almost all initial configurations , satisfying at least the property of ergodicity .
this would allow the use of probability densities , leading from the computation of orbits to the study of statistical quantities like ensemble averages and transport coefficients .
chaotic regions , where nearby solutions diverge exponentially from each other , provide an excellent `` stage '' on which such a desired transition from classical to statistical mechanics can occur . however , the presence of significantly sized islands ( or tori ) of quasiperiodic motion , in which the dynamics is `` stable '' for long times , preclude the success of this scenario and make any attempt at a globally valid statistical description seriously questionable .
these tori occur e.g. around _ simple periodic orbits _ which are stable under small perturbations and ,
if their size does not shrink to zero as @xmath0 or @xmath1 increases , their presence can attribute global consequences to a truly local phenomenon . that such important periodic orbits do exist in hamiltonian lattices , even in the @xmath6 limit ,
could not have been more dramatically manifested than in the remarkable discovery of discrete breathers ( see e.g. @xcite ) , which by now have been observed in a great many experimental situations [ eisenberg _ et al . _ , 1998 ; schwarz _ et al .
_ , 1999 ; fleischer _ et al .
_ , 2003 ; sato _ et al .
_ , 2003 ] .
discrete breathers are precisely one such kind of stable periodic orbits , which also happen to be localized in space , thus representing a very serious limitation to energy transport in nonlinear lattices . then , there were , of course , the famous numerical experiments of fermi , pasta and ulam of the middle @xmath24 s @xcite , which demonstrated the existence of recurrences that prevent energy equipartition among the modes of certain @xmath23d lattices , containing nonlinear interactions between nearest neighbors .
these finite , so
called fpu hamiltonian systems were later shown to exhibit a transition to `` global '' chaos , at high enough energies where major resonances overlap @xcite . before that transition , however , an energy threshold to a `` weak '' form of chaos was later discovered that relies on the interaction of the first few lowest frequency modes and , at least for the fpu system , does appear to ensure equipartition among all modes [ de luca _ et al .
_ , @xmath25 ; de luca & lichtenberg , @xmath26 .
interestingly enough , very recently , this transition to `` weak '' chaos was shown to be closely related to the destabilization of one of the lowest frequency nonlinear normal mode of this fpu system @xcite .
thus , today , @xmath27 years after its famous discovery , the fermi pasta
ulam problem and its transition from recurrences to true statistical behavior is still a subject of ongoing investigation @xcite . in this paper , we have sought to approach the problem of global chaos in hamiltonian systems , by considering two paradigms of @xmath0 dof , @xmath23d nonlinear lattices , with very different origins .
one is the famous fpu lattice mentioned above , with quadratic and quartic nearest neighbor interactions , described by the hamiltonian @xmath28 where @xmath29 is the displacement of the @xmath30th particle from its equilibrium position , @xmath31 is the corresponding canonically conjugate momentum of @xmath29 , @xmath32 is a positive real constant and @xmath1 is the value of the hamiltonian representing the total energy of the system .
the other one is obtained by a discretization of a partial differential equation ( pde ) of the nonlinear schrdinger type referred to as the gross pitaevskii equation @xcite , which in dimensionless form reads @xmath33 where @xmath34 is the planck constant , @xmath35 is a positive constant ( repulsive interactions between atoms in the condensate ) and @xmath36 is an external potential . equation ( [ gross pitaevskii equation ] ) is related to the phenomenon of bose
einstein condensation ( bec ) @xcite .
here we consider the simple case @xmath37 , @xmath38 and discretize the @xmath39dependence of the complex variable @xmath40 in ( [ gross pitaevskii equation ] ) , approximating the second order derivative by @xmath41 .
setting then @xmath42 and @xmath43 , one immediately obtains from the above pde ( [ gross pitaevskii equation ] ) a set of ordinary differential equations ( odes ) for the canonically conjugate variables , @xmath44 and @xmath45 , described by the bec hamiltonian @xcite @xmath46 where @xmath47 and @xmath48 are constant parameters , @xmath49 with @xmath50 and @xmath1 is the total energy of the system . in sec .
[ sec_2 ] , we study these hamiltonians , focusing on some simple periodic orbits ( spos ) , which are known in closed form and whose local ( linear ) stability analysis can be carried out to arbitrary accuracy . by spos
, we refer here to periodic solutions where all variables return to their initial state after only one maximum and one minimum in their oscillation , i.e. all characteristic frequencies have unit ratios . in particular ,
we examine first their bifurcation properties to determine whether they remain stable for arbitrarily large @xmath1 and @xmath0 , having perhaps finitely sized islands of regular motion around them .
this was found to be true only for the so called in phase mode ( ipm ) of the bec hamiltonian ( [ bec_hamiltonian ] ) .
spos corresponding to out of phase motion ( opm ) of either the fpu ( [ fpu_hamiltonian ] ) or the bec system ( [ bec_hamiltonian ] ) destabilize at energy densities @xmath51 , with @xmath52 or @xmath53 ( for the spos we studied ) , as @xmath6 .
the same result was also obtained for what we call the ohs mode of the fpu system @xcite , where all even indexed particles are stationary and all others execute out of phase oscillation , under fixed or periodic boundary conditions .
all these are in agreement with detailed analytical results obtained for families of spos of the same fpu system under periodic boundary conditions ( see e.g. @xcite ) .
then , in sec
. [ sec_3 ] , we vary the values of @xmath1 and @xmath0 and study the behavior of the lyapunov exponents of the opms of the fpu and bec hamiltonians .
we find that , as the eigenvalues of the monodromy matrix exit the unit circle on the real axis , at energies @xmath54 , for fixed @xmath0 , all positive lyapunov exponents @xmath2 , increase following two distinct power laws , @xmath3 , @xmath4 , with the @xmath55 s as reported in the literature , @xcite and @xcite .
furthermore , as the energy @xmath1 grows at fixed @xmath0 , the real eigenvalues of the opm orbit of fpu continue to move away from @xmath9 , unlike the opm of the bec hamiltonian , where for very large @xmath1 all these eigenvalues tend to return to @xmath10 .
interestingly enough , the ipm of the bec hamiltonian remains stable for all the energies and number of dof we studied ! in sec .
[ sec_4 ] , we turn to the question of the `` size '' of islands around stable spos and use the smaller alignment index ( sali ) , introduced in earlier papers @xcite to distinguish between regular and chaotic trajectories in our two hamiltonians .
first , we verify again in these multi degree of freedom systems the validity of the sali dependence on the two largest lyapunov exponents @xmath56 and @xmath57 in the case of chaotic motion , to which it owes its effectiveness and predictive power .
then , computing the sali , at points further and further away from stable spos , we determine approximately the `` magnitude '' of these islands and find that it vanishes ( as expected ) at the points where the corresponding opms destabilize .
in fact , for the opm of the fpu system the size of the islands decreases monotonically before destabilization while for the bec orbit the opposite happens !
even more remarkably , for the ipm of the bec hamiltonian ( [ bec_hamiltonian ] ) , not only does the `` size '' of the island not vanish , it even grows with increasing energy and remains of considerable magnitude for all the values of @xmath1 and @xmath0 we considered . finally , in sec .
[ sec_5 ] , using as initial conditions the unstable spos , we compute the lyapunov spectra of the fpu and bec systems in the so called thermodynamic limit , i.e. as the energy @xmath1 and the number of dof @xmath0 grow indefinitely , with energy density @xmath58 fixed .
first , we find that lyapunov exponents , fall on smooth curves of the form @xmath59 , for both systems .
then , computing the kolmogorov
sinai entropy @xmath60 , as the sum of the positive lyapunov exponents @xcite , in the thermodynamic limit , we find , for both hamiltonians , that @xmath60 is an extensive quantity as it grows linearly with @xmath0 , demonstrating that in their chaotic regions the fpu and bec hamiltonians behave as ergodic systems of statistical mechanics .
we consider a @xmath23d lattice of @xmath0 particles with nearest neighbor interactions given by the hamiltonian @xcite @xmath61 where @xmath29 is the displacement of the @xmath30th particle from its equilibrium position , @xmath31 is the corresponding canonically conjugate momentum of @xmath29 and @xmath32 is a positive real constant .
imposing fixed boundary conditions @xmath62 one finds a simple periodic orbit first studied by @xcite , taking @xmath0 odd , which we shall call the ohs mode ( using ooyama s , hirooka s and sait s initials ) @xmath63 here , we shall examine analytically the stability properties of this mode and determine the energy range @xmath64 over which it is linearly stable .
the equations of motion associated with hamiltonian ( [ fpu_hamiltonian_2 ] ) are @xmath65 whence , using the boundary condition ( [ fpu_fixed_boundary_conditions_ohs ] ) and the expressions ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) for every @xmath66 , we arrive at a single equation @xmath67 describing the anharmonic oscillations of all odd particles of the initial lattice .
the solution of ( [ fpu_single_equation_ohs ] ) is , of course , well known in terms of jacobi elliptic functions @xcite and can be written as @xmath68 where @xmath69 and @xmath70 is the modulus of the @xmath71 elliptic function .
the energy per particle of this spo is then given by @xmath72 setting @xmath73 in ( [ fpu_eq_motion_ohs ] ) and keeping up to linear terms in @xmath74 we get the corresponding variational equations for the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) @xmath75 where @xmath76 .
using the standard method of diagonalization of linear algebra , we can separate these variational equations to @xmath0 uncoupled independent lam equations [ abramowitz & stegun , 1965 ] @xmath77 where the @xmath78 variations are simple linear combinations of @xmath74 s . using ( [ sol_fpu_single_equation_ohs ] ) and changing variables to @xmath79 , eq .
( [ fpu_lame_equations_ohs ] ) takes the form @xmath80 where we have used the identity @xmath81 and primes denote differentiation with respect to @xmath82 .
equation ( [ fpu_simplified_lame_equations_ohs ] ) is an example of hill s equation [ copson , @xmath83 ; magnus & winkler , @xmath84 @xmath85 where @xmath86 is a @xmath87periodic function ( @xmath88 ) with @xmath89 and @xmath90 is the elliptic integral of the first kind . according to floquet theory @xcite the solutions of eq .
( [ fpu_floquet_equation_sample_ohs ] ) are bounded ( or unbounded ) depending on whether the floquet exponent @xmath8 , given by @xmath91 is real ( or imaginary ) .
the matrix @xmath92 is called hill s matrix and in our case its entries are given by @xmath93_{n , m}\equiv\frac{a_{n - m}}{a_{0}-\bigl(\alpha+\frac{n\pi}{\mathcal{k}(\kappa^{2})}\bigr)^{2}}+\delta_{n , m}\ ] ] where @xmath94 , is the kronecker delta with @xmath95 , @xmath96 and the @xmath97 s are the coefficients of the fourier series expansion of @xmath86 , @xmath98 thus , eq .
( [ fpu_floquet_relation_ohs ] ) gives a stability criterion for the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) , by the condition @xmath99 in this case , the fourier coefficients of hill s matrix @xmath100 are given by the relations [ copson , @xmath83 ] @xmath102 where @xmath103 , @xmath90 and @xmath104 are the elliptic integrals of the first and second kind respectively and @xmath105 with @xmath106 .
in the evaluation of @xmath100 in eq .
( [ fpu_stability_criterion_ohs ] ) we have used @xmath107 terms in the fourier series expansion of @xmath108 ( that is , @xmath109 hill s determinants of @xmath100 ) .
thus , we determine with accuracy @xmath110 the @xmath111 values at which the floquet exponent @xmath8 in ( [ fpu_floquet_relation_ohs ] ) becomes zero and the @xmath112 in ( [ fpu_simplified_lame_equations_ohs ] ) become unbounded .
we thus find that the first variation @xmath112 to become unbounded as @xmath113 increases is @xmath114 and the energy values @xmath5 at which this happens ( see eq .
( [ fpu_energy_per_perticle_ohs ] ) ) are listed in table [ table_fpu_ohs ] for @xmath115 .
the @xmath112 variation with @xmath116 has floquet exponent @xmath8 equal to zero for every @xmath117 $ ] , that is @xmath118 corresponds to variations along the orbit .
.the energies @xmath5 , for @xmath115 , at which the @xmath119 in ( [ fpu_simplified_lame_equations_ohs ] ) becomes unbounded for some odd values of @xmath0 .
[ cols="^,^,^,^,^,^,^",options="header " , ] plotting the results in fig .
[ fig_3 ] we observe again that @xmath120 decreases with @xmath0 following a power
law @xmath121 as in the case of the opm of the fpu model . on the other hand ,
the ipm orbit ( [ bec_non_lin_mode_periodic_boundary_conditions_ipm ] ) was found to remain stable for all the values of @xmath0 and @xmath1 we studied ( up to @xmath122 and @xmath123 ) .
let us now study the chaotic behavior in the neighborhood of our unstable spos , starting with the well known method of the evaluation of the spectrum of lyapunov exponents ( les ) of a hamiltonian dynamical system , @xmath124 where @xmath125 .
the les measure the rate of exponential divergence of initially nearby orbits in the phase space of the dynamical system as time approaches infinity . in hamiltonian systems ,
the les come in pairs of opposite sign , so their sum vanishes , @xmath126 and two of them are always equal to zero corresponding to deviations along the orbit under consideration .
if at least one of them ( the largest one ) @xmath127 , the orbit is chaotic , i.e. almost all nearby orbits diverge exponentially in time , while if @xmath128 the orbit is stable ( linear divergence of initially nearby orbits ) .
_ , [ @xmath129 studied in detail the problem of the computation of all les and proposed an efficient algorithm for their numerical computation , which we use here . in particular , @xmath130 for a given orbit @xmath131 is computed as the limit for @xmath132 of the quantities @xmath133 where @xmath134 and @xmath135 are deviation vectors from the given orbit @xmath131 , at times @xmath136 and @xmath137 respectively .
the time evolution of @xmath138 is given by solving the so called variational equations , i.e. the linearized equations about the orbit .
generally , for almost all choices of initial deviations @xmath134 , the limit @xmath139 of eq .
( [ lyap._exps._definition ] ) gives the same @xmath140 . in practice
, of course , since the exponential growth of @xmath141 occurs for short time intervals , one stops the evolution of @xmath141 after some time @xmath142 , records the computed @xmath143 , orthogonormalizes the vectors @xmath141 and repeats the calculation for the next time interval @xmath144 , etc .
obtaining finally @xmath140 as an average over many @xmath145 , @xmath146 given by @xmath147 next , we varied the values of the energy @xmath1 keeping @xmath0 fixed and studied the behavior of the lyapunov exponents , using as initial conditions the opms ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) and ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu and bec hamiltonians respectively . first , we find that the values of the maximum lyapunov exponent @xmath56 increase by two distinct power law behaviors ( @xmath148 ) as is clearly seen in fig .
[ fig_4 ] for the opm of the fpu system .
the result for the @xmath56 for the power law behavior shown by solid line in fig .
[ fig_4 ] is in agreement with the results in @xcite and @xcite , where they obtain @xmath149 and @xmath150 for low dimensional systems and differs slightly from the one obtained in @xcite for the higher dimensional case of @xmath151 .
we also find the same power
law behaviors with similar exponents @xmath152 for the other positive lyapunov exponents as well .
for example , for the @xmath57 we obtain @xmath153 and @xmath154 and @xmath155 and @xmath156 , with the transition occurring at @xmath157 for all of them .
turning now to the full lyapunov spectrum in fig .
[ fig_5 ] we see that , for fixed @xmath0 , as the energy is increased ( and more eigenvalues of the monodromy matrix exit the unit circle ) the lyapunov spectrum tends to fall on a smooth curve for the opm orbits of fpu and bec , see fig .
[ fig_5](a ) , ( b ) , as well as for the ohs mode with periodic boundary conditions ( fig . [ fig_5](c ) ) .
observe that in fig .
[ fig_5](c ) we have plotted the lyapunov spectrum of both the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) and of the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) for @xmath158 and periodic boundary conditions at the energy @xmath159 where both of them are destabilized and their distance in phase space is such that they are far away from each other .
we clearly see that the two lyapunov spectra are almost identical suggesting that their chaotic regions are somehow `` connected '' , as orbits starting initially in the vicinity of one of these spos visit often in the course of time the chaotic region of the other one . in fig .
[ fig_5](d ) we have plotted the positive lyapunov exponents spectra of three neighboring orbits of the ohs mode for @xmath160 dof in three different energies and observe that the curves are qualitatively the same as in fig .
[ fig_5](a ) and ( c ) . finally , in fig .
[ fig_6 ] we have plotted the eigenvalues of the monodromy matrix of several spos and have observed the following : for the opm of the fpu hamiltonian the eigenvalues exit from @xmath9 ( as the orbit destabilizes via period doubling ) and continue to move away from the unit circle , as @xmath1 increases further , see fig .
[ fig_6](a ) .
by contrast , the eigenvalues of the opm of the bec hamiltonian exit from @xmath10 by a symmetry breaking bifurcation and for very large @xmath1 tend to return again to @xmath10 , see fig . [ fig_6](b ) .
this does not represent , however , a return to globally regular motion around this spo , as the lyapunov exponents in its neighborhood remain far from zero . finally , in fig .
[ fig_6](c ) , we show an example of the fact that the eigenvalues of the ipm orbit of the bec system , remain all on the unit circle , no matter how high the value of the energy is . here , @xmath161 , but a similar picture occurs for all the other values of @xmath0 we have studied up to @xmath122 and @xmath123 .
in this section , we estimate the `` size '' of islands of regular motion around stable spos using the smaller alignment index ( sali ) method @xcite , to distinguish between regular and chaotic orbits in the fpu and bec hamiltonians .
the computation of the sali has proved to be a very efficient method in revealing rapidly and with certainty the regular vs. chaotic nature of orbits , as it exhibits a completely different behavior for the two cases : it fluctuates around non
zero values for regular orbits , while it converges exponentially to zero for chaotic orbits .
the behavior of the sali for regular motion was studied and explained in detail by @xcite , while a more analytical study of the behavior of the index in the case of chaotic motion can be found in @xcite . as a first step ,
let us verify in the case of chaotic orbits of our @xmath0degree of freedom systems , the validity of sali s dependence on the two largest lyapunov exponents @xmath162 and @xmath57 proposed and numerically checked for @xmath163 and @xmath164 , in @xcite @xmath165 this expression is very important as it implies that chaotic behavior can be decided by the exponential decay of this parameter , rather than the often questionable convergence of lyapunov exponents to a positive value . to check the validity of ( [ sali_mle_relation ] ) let us take as an example the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) using fixed boundary conditions , with @xmath160 dof and @xmath115 , at the energy @xmath166 and calculate the lyapunov exponents , as well as the corresponding sali evolution . plotting sali as a function of time @xmath167 ( in linear scale ) together with its analytical formula ( [ sali_mle_relation ] ) in fig . [
fig_7](a ) , we see indeed an excellent agreement . increasing further the energy to the value @xmath168 , it is in fact possible to verify expression ( [ sali_mle_relation ] ) , even in the case where the two largest lyapunov exponents are nearly equal , as fig .
[ fig_7](b ) evidently shows ! exploiting now the different behavior of sali for regular and chaotic orbits , we estimate approximately the `` size '' of regions of regular motion ( or , `` islands '' of stability ) in phase space , by computing sali at points further and further away from a stable periodic orbit checking whether the orbits are still on a torus ( sali@xmath169 ) or have entered a chaotic `` sea '' ( sali@xmath170 ) up to the integration time @xmath171 .
the initial conditions are chosen perturbing all the positions of the stable spo by the same quantity @xmath172 and all the canonically conjugate momenta by the same @xmath173 while keeping always constant the integral @xmath174 , given by eq .
( [ bec_hamiltonian_second_integral ] ) , in the case of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) and the energy @xmath1 in the case of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) . in this way
, we are able to estimate the approximate `` magnitude '' of the islands of stability for the opm of hamiltonian ( [ fpu_hamiltonian_2 ] ) and for the ipm and opm of hamiltonian ( [ bec_hamiltonian_2 ] ) varying the energy @xmath1 and the number of dof @xmath0 . in the case of the opm solutions of both hamiltonians , as the number of dof @xmath0 increases , for fixed energy @xmath1
, the islands of stability eventually shrink to zero and the spos destabilize .
for example , this is seen in fig . [ fig_8](a ) for the islands around the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) .
a surprising behavior , however , is observed for the same spo , if we keep @xmath0 fixed and increase the energy : instead of diminishing , as expected from the fpu and other examples , the island of stability actually grows , as shown in fig .
[ fig_8](b ) , for the case of @xmath161 dof .
in fact , it remains of considerable size until the spo is destabilized for the first time , through period doubling bifurcation at @xmath175 , whereupon the island ceases to exist ! but
what happens to the island of stability around the ipm solution of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) , which does not become unstable for all values of @xmath0 and @xmath1 we studied ? does it shrink to zero at sufficiently large @xmath1 or @xmath0 ?
from fig . [ fig_8](c ) we see that for a fixed value of @xmath0 , the size of this island also increases as the energy increases .
in fact , this spo has large islands about it even if the energy is increased , keeping the ratio @xmath12 constant ( see fig .
[ fig_8](d ) ) .
this was actually found to be true for considerably larger @xmath1 and @xmath0 values than shown in this figure .
finally , choosing again as initial conditions the unstable opms of both hamiltonians , we determine some important statistical properties of the dynamics in the so called thermodynamic limit of @xmath1 and @xmath0 growing indefinitely , while keeping @xmath12 constant . in particular , we compute the spectrum of the lyapunov exponents of the fpu and bec systems starting at the opm solutions ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) and ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) for energies where these orbits are unstable . we thus find that the lyapunov exponents are well approximated by smooth curves of the form @xmath176 , for both systems , with @xmath177 , @xmath178 respectively and @xmath179 where @xmath180 ( see fig . [ fig_9 ] ) .
specifically , in the case of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) for fixed energy density @xmath181 we find @xmath182 while , in the case of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) for fixed energy density @xmath183 , a similar behavior is observed , @xmath184 these exponential formulas were found to hold quite well , up to @xmath185 .
for the remaining exponents , the spectrum is seen to obey different decay laws , which are not easy to determine . as this appears to be a subtle matter , however , we prefer to postpone it for a future publication .
the functions ( [ l_i_fpu_opm ] ) and ( [ l_i_bec_opm ] ) , provide in fact , invariants of the dynamics , in the sense that , in the thermodynamic limit , we can use them to evaluate the average of the positive lyapunov exponents ( i.e. the kolmogorov
sinai entropy per particle ) for each system and find that it is a constant characterized by the value of the maximum lyapunov exponent @xmath56 and the exponent @xmath8 appearing in them . in fig .
[ graph_16 ] we compute the well known kolmogorov
sinai entropy @xmath186 @xcite ( solid curves ) , which is defined as the sum of the @xmath187 positive lyapunov exponents , @xmath188 in this way , we find , for both hamiltonians , that @xmath186 is an extensive thermodynamic quantity as it is clearly seen to grow linearly with @xmath0 ( @xmath189 ) , demonstrating that in their chaotic regions the fpu and bec hamiltonians behave as ergodic systems of statistical mechanics . finally , using eqs .
( [ l_i_fpu_opm ] ) and ( [ l_i_bec_opm ] ) , as if they were valid for all @xmath190 , we approximate the sum of the positive lyapunov exponents @xmath140 , and calculate the @xmath186 entropy from eq .
( [ ks_entropy ] ) as @xmath191 and @xmath192 in fig .
[ graph_16 ] , we have plotted eqs .
( [ h_ks_fpu_opm ] ) and ( [ h_ks_bec_opm ] ) with dashed curves ( adjusting the proportionality constants appropriately ) and obtain nearly straight lines with the same slope as the data computed by the numerical evaluation of the @xmath186 , from eq .
( [ ks_entropy ] ) .
in this paper we have investigated the connection between local and global dynamics of two @xmath0 dof hamiltonian systems , describing @xmath23d nonlinear lattices with different origins known as the fpu and bec systems .
we focused on solutions located in the neighborhood of simple periodic orbits and showed that as the energy increases beyond the destabilization threshold , all positive lyapunov exponents increase monotonically with two distinct power
law dependencies on the energy .
we also computed the destabilization energy per particle of the ohs mode and of the two opm orbits and found that it decays with a simple power law of the form @xmath7 , @xmath193 1 or 2 .
one notable exception is the ipm orbit of the bec hamiltonian , which is found to be stable for any energy and number of dof we considered !
furthermore , we found that as we increase the energy @xmath1 of both hamiltonians for fixed @xmath0 , the spos behave in very different ways : in the opm case of the fpu hamiltonian the eigenvalues of the monodromy matrix always move away from @xmath9 with increasing energy , while in the opm of the bec hamiltonian these eigenvalues return to @xmath10 , for very high energies .
nevertheless , this behavior does not represent a return of the system to a globally regular motion around the spo as one might have thought .
it simply reflects a local property of the spo , as the orbits in its neighborhood still have positive lyapunov exponents which are far from zero .
we have also been able to estimate the `` size '' of the islands of stability around the spos with the help of a recently introduced very efficient indicator called sali and have seen them to shrink , as expected , when increasing @xmath0 for fixed @xmath1 in the case of the opm of the fpu hamiltonian .
of course , when we continue increasing the energy , keeping the number of dof fixed , the opms destabilize and the islands of stability are destroyed .
unexpectedly , however , in the case of the opm of the bec hamiltonian these islands were seen to _ grow _ in size if we keep @xmath0 fixed and increase @xmath1 up to the destabilization threshold , a result that clearly requires further investigation .
the peculiarity of our bec hamiltonian is even more vividly manifested in the fact that the islands of stability about its ipm orbit never vanish , remaining actually of significant size , even for very high values of @xmath1 and @xmath0 ( keeping @xmath12 fixed ) .
starting always near unstable spos , we also calculated the lyapunov spectra characterizing chaotic dynamics in the regions of the ohs and opm solutions of both hamiltonians .
keeping @xmath0 fixed , we found energy values where these spectra were practically the same near spos which are far apart in phase space , indicating that the chaotic regions around these spos are visited ergodically by each other s orbits
. finally , using an exponential law accurately describing these spectra , we were able to show , for both hamiltonians , that the associated kolmogorov
sinai entropies per particle increase linearly with @xmath0 in the thermodynamic limit of @xmath194 and @xmath6 and fixed @xmath12 and , therefore , behave as extensive quantities of statistical mechanics .
our results suggest , however , that , even in that limit , there may well exist hamiltonian systems with significantly sized islands of stability around stable spos ( like the ipm of the bec hamiltonian ) , which must be excluded from a rigorous statistical description .
it is possible , of course , that these islands are too small in comparison with the extent of the chaotic domain on a constant energy surface and their `` measure '' may indeed go to zero in the thermodynamic limit .
but as long as @xmath1 and @xmath0 are finite , they will still be there , precluding the global definition of probability densities , ensemble averages and the validity of the ergodic hypothesis over all phase space .
clearly , therefore , their study is of great interest and their properties worth pursuing in hamiltonian systems of interest to physical applications .
this work was partially supported by the european social fund ( esf ) , operational program for educational and vocational training ii ( epeaek ii ) and particularly the program herakleitos , providing a ph .
d scholarship for one of us ( c. a. ) . c. a. also acknowledges with gratitude the @xmath164 month hospitality , march
june @xmath195 , of the `` center for nonlinear phenomena and complex systems '' of the university of brussels .
in particular , he thanks professor g. nicolis , professor p. gaspard and dr .
v. basios for their instructive comments and useful remarks in many discussions explaining some fundamental concepts treated in this paper .
the second author ( t. b. ) wishes to express his gratitude to the max planck institute of the physics of complex systems at dresden , for its hospitality during his @xmath164 month visit march
june @xmath195 , when this work was completed .
in particular , t. b. wants to thank dr .
sergej flach for numerous lively conversations and exciting arguments on the stability of multi dimensional hamiltonian systems .
useful discussions with professors a. politi , r. livi , r. dvorak , f. m. izrailev and dr .
t. kottos are also gratefully acknowledged . the third author ( c. s. )
was partially supported by the research committee of the academy of athens .
alpha abramowitz , m. and stegun , i. [ 1965 ] , `` handbook of mathematical functions '' , ( dover , new york ) , chap . 16 .
aubry , s. , flach , s. , kladko , k. and oldbrich , e. [ 1996 ] , `` manifestation of classical bifurcation in the spectrum of the integrable quantum dimer '' , _ phys .
_ , * 76 * , ( 10 ) , pp
. 16071610 .
benettin , g. , galgani , l. , giorgilli , a. and strelcyn , j. m. [ 1980 ] , `` lyapunov characteristic exponents for smooth dynamical systems and for hamiltonian systems ; a method for computing all of them , part 1 : theory '' , _ meccanica _ , * 15 * , * march * , pp .
9 - 20 .
benettin , g. , galgani , l. , giorgilli , a. and strelcyn , j. m. [ 1980 ] , `` lyapunov characteristic exponents for smooth dynamical systems and for hamiltonian systems ; a method for computing all of them , part 2 : numerical applications '' , _ meccanica _ , * 15 * , * march * , pp .
21 - 30 .
benettin , g. [ 1984 ] , `` power law behavior of lyapunov exponents in some conservative dynamical systems''_physica d _ , * 13 * , pp .
211220 .
berman , g. p. and izrailev , f. m. [ 2005 ] , `` the fermi pasta
ulam problem : 50 years of progress '' , chaos * 15 * 015104 .
bountis , t. ed .
[ 1992 ] , `` proton transfer in hydrogen
bonded systems '' , _ proceedings of nato arw _ ,
plenum , london .
budinsky , n. and bountis , t. [ 1983 ] , `` stability of nonlinear modes and chaotic properties of @xmath23d fermi pasta ulam lattices '' , _ physica d _ , * 8 * , pp .
445 452 .
cafarella , a. , leo , m. and leo , r. a. [ 2003 ] , `` numerical analysis of the one mode solutions in the fermi pasta
ulam system '' , _ phys .
, * 69 * , pp .
046604 .
chirikov , b. v. [ 1979 ] , `` a universal instability of many dimensional oscillator systems '' , _ phys .
_ , * 52 * , ( 5 ) , pp .
263379 .
contopoulos , g. [ 2002 ] , _ order and chaos in dynamical astronomy _ , ( springer ) , astronomy and astrophysics library .
copson , e. t. [ 1935 ] , _ an introduction to the theory of functions of a complex variable _ , ( oxford univ . press , oxford ) ,
chap . 14 .
dalfovo , f. , giorgini , s. , pitaevskii , l. p. and stringari , s. [ 1999 ] , `` theory of bose - einstein condensation in trapped gases '' , _ rev .
_ , * 71 * , pp .
463 - 512 .
de luca , j. , lichtenberg , a. j. and lieberman , m. a. [ 1995 ] , `` time scale to ergodicity in the fermi pasta ulam system '' , _ chaos _ , * 5 * , ( 1 ) , pp .
283297 .
de luca , j. and lichtenberg , a. j. [ 2002 ] , `` transitions and time scales to equipartition in oscillator chains : low frequency initial conditions '' , _ phys .
e _ , * 66 * , ( 2 ) , pp .
026206 .
eisenberg , h. s. , silberberg , y. , morandotti , r. , boyd , a. r. & aitchison , j. s. [ 1998 ] , `` discrete spatial optical solitons in waveguide arrays '' , _ phys .
lett . _ * 81*(16 ) , pp .
33833386 .
fermi , e. , pasta , j. and ulam , s. [ 1955 ] , `` studies of nonlinear problems '' , _ los alamos document la1940_. see also : `` nonlinear wave motion '' , [ 1974 ] , _ am .
soc . providence _ ,
* 15 * , lectures in appl .
math . , ed .
newell a. c. flach , s. and willis , c. r. [ 1998 ] , `` discrete breathers '' , _ phys .
_ , * 295 * , ( 5 ) , pp .
181264 .
flach , s. , ivanchenko , m. v. and kanakov , o. i. [ 2005 ] , `` q - breathers and the fermi pasta
ulam problem '' , _ phys .
_ , * 95 * , pp .
064102 - 1064102 - 4 . fleischer , j. w. , segev , m. , efremidis , n. k. & christodoulides , d. n. [ 2003 ] , `` observation of two dimensional discrete solitons in optically induced nonlinear photonic lattices '' , _ nature _ * 422*(6928 ) , pp . 147150 .
ford , j. [ 1992 ] , `` the fermi pasta
ulam problem : paradox turned discovery '' , _ physics reports _ , * 213 * , pp .
271310 .
hilborn , r. c. [ 1994 ] , _ chaos and nonlinear dynamics _ , ( oxford university press ) .
izrailev , f. m. and chirikov , b. v. [ 1966 ] , `` statistical properties of a nonlinear string '' , _ sov .
_ , * 11 * , pp . 30 .
ketterle , w. , durfee , d. s. and stamper kurn , d. m. [ 1999 ] , `` bose einstein condensation in atomic gases '' , in _ international school of physics _ , eds .
fermi , e. , course @xmath196 , edited by inguscio , m. , stringari , s. and wieman , c. , ( ios press , amsterdam ) .
lichtenberg , a. j. and lieberman , m. a [ 1991 ] , `` regular and stochastic motion '' , 2nd edition ( springer verlag , new york ) .
livi , r. , politi , a. and ruffo , s. , [ 1986 ] , `` distribution of characteristic exponents in the thermodynamic limit '' , _ j. phys .
_ , * 19 * , pp . 20332040 .
mackay , r. s. and meiss , j. d. [ 1987 ] , `` hamiltonian dynamical systems '' , ( adam hilger , bristol ) .
magnus , w. and winkler , s. [ 1966 ] , `` hill s equation '' , ( interscience , new york ) .
ooyama , n. , hirooka , h. and sait , n. [ 1969 ] , `` computer studies on the approach to thermal equilibrium in coupled anharmonic oscillators .
ii . one dimensional case '' , _ j. phys . soc . of japan _ ,
* 27 * , ( 4 ) , pp . 815 824 .
pesin , ya .
b. [ 1976 ] , `` lyapunov characteristic indexes and ergodic properties of smooth dynamic systems with invariant measure '' , _ dokl .
sssr _ , * 226 * , ( 4 ) , pp .
774777 .
poggi , p. & ruffo , s. [ 1997 ] , `` exact solutions in the fpu oscillator chain '' , _ physica d _ , * 103 * , pp .
251272 .
prosmiti , r. & farantos , s. c. [ 1995 ] , `` periodic orbits , bifurcation diagrams and the spectroscopy of @xmath197 system '' , _ j. chem .
_ , * 1039 * , pp .
3299 3314 .
rechester , a. b. , rosenbluth , m. n. & white , r. b. [ 1979 ] , `` calculation of the kolmogorov entropy for motion along a stochastic magnetic field '' , _ phys .
_ , * 42 * , pp . 1247 .
scandale , w. & turchetti , g. , eds .
[ 1991 ] , `` nonlinear problems in future particle accelerators '' , ( world scientific , singapore ) . sato , m. hubbard , b. e. , sievers , a. j. , ilic , b. , czaplewski , d. a. & craighead , h. g. [ 2003 ] , `` observation of locked intrinsic localized vibrational modes in a micromechanical oscillator array '' _ phys . rev .
_ , * 90*(4 ) , pp . 044102 .
schwarz , u. t. , english , l. q. , & sievers a. j. [ 1999 ] , `` experimental generation and observation of intrinsic localized spin wave modes in an antiferromagnet '' , _ phys .
_ , * 83*(1 ) , pp
. 223226 .
sim , c. ed .
[ 1999 ] , _ hamiltonian systems with three or more degrees of freedom _ , ( kluwer academic publishers ) , * 533 * , nato asi .
skokos , ch .
[ 2001 ] , `` alignment indices : a new , simple method for determining the ordered or chaotic nature of orbits '' , _
a _ , * 34 * , pp .
10029 10043 .
skokos , ch . ,
antonopoulos , ch . ,
bountis , t. c. and vrahatis , m. n. [ 2003 ] , `` smaller alignment index ( sali ) : determining the ordered or chaotic nature of orbits in conservative dynamical systems '' , in _ proceedings of the conference libration point orbits and applications _ , eds .
gomez , g. , lo , m. w. and masdemont , j. j. , ( world scientific ) , pp . 653 664 .
skokos , ch . ,
antonopoulos , ch . ,
bountis , t. c. and vrahatis , m. n. [ 2003 ] , `` how does the smaller alignment index ( sali ) distinguish order from chaos ? '' , _ prog . theor .
_ , * 150 * , pp .
439 443 skokos , ch . ,
antonopoulos , ch . , bountis , t. c. and vrahatis , m. n. [ 2004 ] , `` detecting order and chaos in hamiltonian systems by the sali method '' , _ j. phys .
a _ , * 37 * , pp .
6269 6284 .
smerzi , a. and trombettoni , a. [ 2003 ] , `` nonlinear tight binding approximation for bose
einstein condensates in a lattice '' , _ phys .
_ , * a68 * , 023613 .
trombettoni , a. and smerzi , a. [ 2001 ] , `` discrete solitons and breathers with dilute bose
einstein condensates '' , _ phys .
_ , * 86 * , pp . 2353 .
wiggins , s. [ 1988 ] , `` global bifurcations and chaos : analytical methods '' , ( new york , springer verlag ) .
the solid curve corresponds to the energy per particle @xmath198 , for @xmath115 , of the first destabilization of the ohs nonlinear mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) of the fpu system ( [ fpu_hamiltonian_2 ] ) obtained by the numerical evaluation of the hill s determinant in ( [ fpu_stability_criterion_ohs ] ) , while the dashed line corresponds to the function @xmath199 . note that both axes are logarithmic .
the solid curve corresponds to the energy per particle @xmath198 , for @xmath200 , of the first destabilization of the nonlinear opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu system ( [ fpu_hamiltonian_2 ] ) obtained by the numerical evaluation of the hill s determinant , while the dashed line corresponds to the function @xmath201 . note that both axes are logarithmic .
the solid curve corresponds to the energy per particle @xmath198 of the first destabilization of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) obtained by the numerical evaluation of the eigenvalues of the monodromy matrix of eq .
( [ bec_linearized_equations ] ) , while the dashed line corresponds to the function @xmath202 .
note that both axes are logarithmic .
4 . the two distinct power law behaviors in the evolution of the maximum lyapunov exponent @xmath56 as the energy grows for the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) for @xmath203 .
a similar picture is obtained for the @xmath57 and @xmath204 also , with similar exponents and the transition occurring at the same energy value ( see text ) . note that both axes are logarithmic .
\(a ) the spectrum of the positive lyapunov exponents for fixed @xmath203 of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) as the energy grows .
( b ) also for @xmath203 , the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) yields a similar picture as the energy is increased .
( c ) the lyapunov spectrum of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) for @xmath158 and the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) of the same hamiltonian and @xmath0 , for periodic boundary conditions practically coincide at @xmath159 where both of them are destabilized .
( d ) the lyapunov spectrum of the fpu ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) with fixed boundary conditions for @xmath160 as the energy grows presents as shape which is qualitatively similar to what was found for the spos of panel ( c ) .
the eigenvalues @xmath205 ( a ) of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) with @xmath203 dof . (
b ) the eigenvalues of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) with the same number of dof @xmath0 .
( c ) the eigenvalues of the ipm ( [ bec_non_lin_mode_periodic_boundary_conditions_ipm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) with @xmath161 dof . 7 .
\(a ) the time evolution of the sali ( solid curve ) and of the eq .
( [ sali_mle_relation ] ) ( dashed line ) at @xmath166 of the ohs mode ( [ fpu_non_lin_mode_fixed_boundary_conditions_ohs ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) with fixed boundary conditions .
( b ) similar plot to panel ( a ) but for the larger energy @xmath168 for which the two largest lyapunov exponents @xmath56 and @xmath57 are almost equal while all the other positive ones are very close to zero . in both panels
the agreement between the data ( solid curve ) and the derived function of eq .
( [ sali_mle_relation ] ) ( dashed line ) is remarkably good .
note that the horizontal axes in both panels are linear .
\(a ) `` size '' of the islands of stability of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) for @xmath206 and @xmath207 dof and spos constant energy @xmath208 before the first destabilization ( see table [ table_bec_opm ] ) .
( b ) `` size '' of the islands of stability of the same hamiltonian and spo as in ( a ) for @xmath161 dof and three different energies of the spo before the first destabilization ( see table [ table_bec_opm ] ) .
( c ) `` size '' of the islands of stability of the same hamiltonian as in ( a ) of the ipm ( [ bec_non_lin_mode_periodic_boundary_conditions_ipm ] ) for @xmath161 dof and four different energies of the spo .
( d ) `` size '' of the islands of stability of the same hamiltonian as in ( a ) of the ipm ( [ bec_non_lin_mode_periodic_boundary_conditions_ipm ] ) for @xmath209 . here
@xmath1 corresponds to the energy of the ipm .
\(a ) positive lyapunov exponents spectrum of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) for fixed @xmath181 .
( b ) positive lyapunov exponents spectrum of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) for fixed @xmath183 . in both panels
@xmath210 runs from @xmath23 to @xmath0 .
\(a ) the @xmath186 entropy of the opm ( [ fpu_non_lin_mode_periodic_boundary_conditions_opm ] ) of the fpu hamiltonian ( [ fpu_hamiltonian_2 ] ) for fixed @xmath181 ( solid curve ) and the approximated formula ( [ h_ks_fpu_opm ] ) ( dashed curve ) .
( b ) the @xmath186 entropy of the opm ( [ bec_non_lin_mode_periodic_boundary_conditions_opm ] ) of the bec hamiltonian ( [ bec_hamiltonian_2 ] ) for fixed @xmath183 ( solid curve ) and the approximated formula ( [ h_ks_bec_opm ] ) ( dashed curve ) . | we investigate the connection between local and global dynamics of two @xmath0degree of freedom hamiltonian systems with different origins describing one dimensional nonlinear lattices : the fermi pasta ulam ( fpu ) model and a discretized version of the nonlinear schrdinger equation related to the bose einstein condensation ( bec ) .
we study solutions starting in the immediate vicinity of simple periodic orbits ( spos ) representing in phase ( ipm ) and out of phase motion ( opm ) , which are known in closed form and whose linear stability can be analyzed exactly .
our results verify that as the energy @xmath1 increases for fixed @xmath0 , beyond the destabilization threshold of these orbits , all positive lyapunov exponents @xmath2 , exhibit a transition between two power laws , @xmath3 , @xmath4 , occurring at the same value of @xmath1 .
the destabilization energy @xmath5 per particle goes to zero as @xmath6 following a simple power
law , @xmath7 , with @xmath8 being 1 or 2 for the cases we studied .
however , using the sali , a very efficient indicator we have recently introduced for distinguishing order from chaos , we find that the two hamiltonians have very different dynamics near their _ stable _ spos : for example , in the case of the fpu system , as the energy increases for fixed @xmath0 , the islands of stability around the opm decrease in size , the orbit destabilizes through period doubling bifurcation and its eigenvalues move steadily away from @xmath9 , while for the bec model the opm has islands around it which _ grow _ in size before it bifurcates through symmetry breaking , while its real eigenvalues return to @xmath10 at very high energies .
furthermore , the ipm orbit of the bec hamiltonian _ never destabilizes _ , having finite
sized islands around it , even for very high @xmath0 and @xmath1 .
still , when calculating lyapunov spectra , we find for the opms of both hamiltonians that the lyapunov exponents decrease following an exponential law and yield extensive kolmogorov sinai entropies per particle @xmath11 , in the thermodynamic limit of fixed energy density @xmath12 with @xmath1 and @xmath0 arbitrarily large .
_ @xmath13department of mathematics + and + center for research and applications of nonlinear systems ( crans ) , + university of patras , + gr@xmath14 , rio , patras , greece + @xmath15research center for astronomy and applied mathematics + academy of athens , + soranou efesiou @xmath16 , gr@xmath17 , athens , greece + @xmath18department of applications of informatics and management in finance , + technological institute of mesologhi , + gr@xmath19 , mesologhi , greece _ _ keywords : _ hamiltonian systems , simple periodic orbits , regular and chaotic behavior , lyapunov spectra , kolmogorov entropy , sali method . |
realization of a quantum computer would be very important for the solution of hard problems in classical mathematics @xcite and even more important for the solution of many quantum problems in field theory , quantum chemistry , material physics , etc.@xcite however , despite a large effort of many groups the realization remains far away because of the conflicting requirements imposed by scalability and decoupling from the environment .
this dichotomy can be naturally resolved by the error free computation in the topologically protected space @xcite if the physical systems realizing such spaces are identified and implemented . though difficult
, this seems more realistic than attempts to decouple simple physical systems implementing individual bits from the environment .
thus , the challenge of the error free quantum computation resulted in a surge of interest to physical systems and mathematical models that were considered very exotic before .
the main idea of the whole approach , due to kitaev @xcite , is that the conditions of long decoherence rate and scalability can be in principle satisfied if elementary bits are represented by anyons , the particles that indergo non - trivial transformations when moved adiabatically around each other ( braided ) @xcite and that can not be distinguished by local operators .
one of the most famous examples of such excitations is provided by the fractional quantum hall effect @xcite .
the difficult part is , of course , to identify a realistic physical system that has such excitations and allows their manipulations . to make it more manageable
, this problem should be separated into different layers .
the bottom layer is the physical system itself , the second is the theoretical model that identifies the low energy processes , the third is the mathematical model that starts with the most relevant low energy degrees of freedom and gives the properties of anyons while the fourth deals with construction of the set of rules on how to move the anyons in order to achieve a set of universal quantum gates ( further lies the layer of algorithms and so on ) .
ideally , the study of each layer should provide the one below it a few simplest alternatives and the one above the resulting properties of the remaining low energy degrees of freedom . in this paper
we focus on the third layer : we discuss a particular set of mathematical models that provides anyon excitations , namely the chern simons gauge theories with the discrete gauge groups .
generally , an appealing realization of the anyons is provided by the fluxes in non - abelian gauge theories .
@xcite . the idea is to encode individual bits in the value of fluxes belonging to the same conjugacy class of the gauge group : such fluxes can not be distinguished locally because they are transformed one into another by a global gauge transformation and would be even completely indistiguishable in the absence of other fluxes in the system .
alternatively , one can protect anyons from the adverse effect of the local operators by spreading the fluxes over a large region of space . in this case
one does not need to employ a non - abelian group : the individual bits can be even encoded by the presence / absence of flux in a given area , for instance a large hole in the lattice .
such models @xcite are much easier to implement in solid state devices but they do not provide a large set of manipulation that would be sufficient for universal quantum computation .
thus , these models provide a perfect quantum memory but not a quantum processor . on the other hand
, the difficulty with the flux representation is that universal computation can be achieved only by large non - abelian groups ( such as @xmath1 ) that are rather difficult to implement , or if one adds charge motion to the allowed set of manipulations .
because the charges form a non - trivial representation of the local gauge group , it is difficult to protect their coherence in the physical system which makes the latter alternative also difficult to realize in physical systems . the last alternative is to realize a chern - simons model where on the top of the conjugacy transformations characteristic of non - abelian theories , the fluxes acquire non - trivial phase factors when moved around each other .
we explore this possibility in this paper .
chern - simons theories with finite non - abelian gauge groups have been extensively studied for a continuous @xmath2 dimensional space - time . unlike continous gauge - group , discrete groups on a continous space allow non - trivial fluxes only around cycles which can not be contracted to a single point .
so in a path - integral approach , and by contrast to the case of continuous groups , the integration domain is restricted to gauge - field configurations for which the local flux density vanishes everywhere .
such path integrals were introduced , analyzed , and classified in the original paper by dijkgraaf and witten @xcite .
they showed that for a given finite gauge group @xmath3 , all possible chern - simons actions in @xmath2 dimensions are in one to one correspondence with elements in the third cohomology group @xmath4 of @xmath3 with coefficients in @xmath5 .
they also provided a description in terms of a @xmath2 lattice gauge theory , where space - time is tiled by tetrahedra whose internal states are described by just three independent group elements @xmath6 because fluxes through all triangular plaquettes vanish .
elements of @xmath4 are then identified with functions that play the role of the elementary action for a tetrahedron .
this description turns out to be rather cumbersome because does not have specific symmetry properties , so that the definition of the total action requires to choose an ordering for all the lattice sites , which can not be done in a natural way . as a result
, it seems very difficult to take the limit of a continuous time that is necessary for our purposes because physical implementations always require an explicit hamiltonian form . in principle , the knowledge of an elementary action @xmath7 in @xmath4 allows to derive all braiding properties of anyonic excitations ( i.e. local charges and fluxes ) .
this has been done directly from the original path - integral formulation @xcite , using as an intermediate step the representation theory of the so - called quasi - quantum double associated to the group @xmath3 @xcite .
this mathematical struture is completely defined by the group @xmath3 and a choice of any element @xmath7 in @xmath4 . using this later formalism , which bypasses the need to construct microscopic hamiltonians ,
many detailed descriptions of anyonic properties for various finite groups , such as direct products of cyclic groups , or dihedral groups , have been given by de wild propitius @xcite .
unfortunately , these general results do not provide a recipe how to contruct a microscopic hamiltonian with prescribed braiding properties . to summarize , in spite of a rather large body of theoretical knowledge , we still do not know which chern - simons models with a finite gauge group can _ in principle _ be realized in a physical system with a local hamiltonian , that is which one can appear as low energy theory . to answer
this question is the purpose of the present paper .
the main ingredients of our construction are the following .
we shall be mostly interested in a hilbert space spanned by _
dilute _ classical configurations of fluxes because it is only these configurations that are relevant for quantum computation that involves flux braiding and fusion .
furthermore , we expect that a finite spacial separation between fluxons may facilitate various manipulations and measurements .
notice in this respect that in theories with discrete group symmetry in a continuous space@xcite , non - trivial fluxes appear only thanks to a non - trivial topology of the ambiant space @xcite and thus are restricted to large well separated holes in a flat 2d space .
the second feature of our construction is that gauge generators are modified by phase - factors which depend on the actual flux configuration in a local way .
this is a natural generalization of the procedure we have used recently for the construction of chern - simons models with @xmath8 symmetry group @xcite .
note that in the most general situations , the representation of the local gauge group in a discrete chern - simons theory becomes _ projective _
this would be inappropriate for a robust implementation because projective representations lead to degenerate multiplets that are strongly coupled to local operators , and therefore become very fragile in the presence of external noise .
we shall therefore restrict ourselves to the models were no projective representations occur .
in practice , they are associated with the non - trivial elements of the group @xmath9 , and it turns out that for some interesting classes of groups such as the dihedral groups @xmath10 with @xmath11 odd , @xmath9 is trivial @xcite , so this restriction is not too important . as we show in section [ sectiongenerators ] , these assumptions allow us to find all the possible phase factors associated with the gauge transformations in terms of homomorphisms from the subgroups of @xmath3 that leave invariant a fixed element of @xmath3 under conjugacy into @xmath5 . the last step is to construct a set of local gauge - invariant operators corresponding to the following elementary processes : moving a single fluxon , creating or annihilating a pair of fluxons with vacuum quantum numbers , and branching processes were two nearby fluxons merge into a single one ( or conversely ) .
we shall see that the requirement of local gauge invariance leads to a relatively mild constraint for the possible phase - factors , leaving a rather large set of solutions .
the main result of this work is twofold .
first , we provide an explicit computation of holonomy properties of fluxes in a set a models based on dihedral groups . of special interest
is the simplest of them , @xmath12 which is also the permutation group of 3 elements @xmath13 .
this group belongs to a class which is in principle capable of universal quantum computation @xcite .
this part is therefore connected to the upper layer ( designing a set of univeral gates from properties of anyons ) in the classification outlined above .
but our construction of a local hamiltonian version for a chern - simons model on a lattice provides some guidelines for possible desirable physical implementations .
the plan of the paper is the following .
section ii is mostly pedagogical , providing the motivation for our general construction through the simplest example of a chern - simons gauge theory , namely the non - compact @xmath5 model . in section iii
we formulate general conditions on local chern - simons phase factors that satisfy gauge invariance conditions . in section
iv we discuss the construction of the electric field operator and derive condition on the phase factor that allows one a gauge invariant fluxon dynamics . in section v we discuss the fluxon braiding properties and derive the chern - simons phase factors associated with the braiding . in section
vi we apply our results to the simplest non - abelian groups @xmath0 .
finally , section vii gives conclusions .
some technical points relevant for the general construction have been relegated to appendix a , and appendix b discusses some interesting features of the torus geometry .
although this geometry is not easy to implement in experiments , it is conceptually very interesting since it is the simplest example of a two - dimensional space without boundary and with topologically non - trivial closed loops .
to motivate the construction of the present paper , it is useful to discuss some important features of chern - simons gauge theories in the simpler case of abelian groups . for this purpose
, we shall consider here as an illustration the model based on the _ continuous _ abelian group with one generator in its non - compact version .
of course , our main purpose here is to address _ finite _ groups , but as we shall discuss , this non - compact @xmath5 model contains already the key ingredients . on a @xmath2 dimensional space - time
, it is defined from the following lagrangian density : @xmath14 where @xmath15 is the local magnetic field ( a pseudo - scalar in @xmath2 dimensions ) and dots stand for time - derivatives .
we have used the gauge in which the time component @xmath16 of the vector potential is zero . because of this , we shall consider only invariance under time - independent gauge transformations in this discussion .
these are defined as usual by , where @xmath17 is any time - independent smooth scalar function of spacial position .
under such a gauge transformation , the action associated to the system evolution on a two - dimensional space manifold @xmath18 during the time interval @xmath19 $ ] varies by the amount @xmath20 where : @xmath21 because @xmath22 is time - independent , the integrand is a total time - derivative , so we may write , where @xmath23 denotes the field configuration at time @xmath24 , @xmath25 , and : @xmath26 in the case where @xmath18 has no boundary ( and in particular no hole ) , we may integrate by parts and get : @xmath27 when @xmath28 , this modifies in a non - trivial way the behavior of the corresponding quantized model under time - independent gauge transformations . one way to see
this is to consider the time - evolution of the system s wave - functional @xmath29 . in a path - integral approach , the probability amplitude to go from an initial configuration @xmath30 at time @xmath31 to a final @xmath32 at time @xmath33
is given by : @xmath34 where fields @xmath35 are required to satisfy boundary conditions for @xmath36 .
after the gauge - transformation , and using the above expression for @xmath20 , we see that the probability amplitude connecting the transformed field configurations @xmath37 and @xmath38 is : @xmath39 it is therefore natural to define the gauge - transformed wave - functional @xmath40 by : @xmath41 this definition ensures that @xmath29 and @xmath42 evolve according to the same set of probability amplitudes .
in a hamiltonian formulation , we associate to any classical field configuration @xmath43 a basis state @xmath44 .
the gauge - transformation corresponding to @xmath22 is now represented by a unitary operator @xmath45 defined by : @xmath46 the presence of the phase - factor is one of the essential features of the chern - simons term ( i.e. the term proportional to @xmath47 ) added to the action . note that when @xmath22 varies , the family of operators @xmath45 gives rise to a representation of the full group of local gauge transformations .
indeed , at the classical level , the composition law in this group is given by the addition of the associated @xmath22 functions , and because @xmath48 given in eq .
( [ iwithoutboundary ] ) is itself gauge - invariant , we have .
it is interesting to give an explicit expression for @xmath45 .
it reads : @xmath49 where @xmath50 and @xmath51 are the canonically conjugated variables to @xmath52 and @xmath53 .
note that this no longer holds in the case of a manifold @xmath18 with a boundary , as will be discussed in a few paragraphs . in the hamiltonian quantization , a very important role
is played by the gauge - invariant electric operators @xmath54 and @xmath55 . in the absence of chern - simons term , they are simply equal to @xmath50 and @xmath51 . when @xmath28 , because of eq .
( [ explicitu(f ) ] ) , the transformation law for @xmath50 and @xmath51 becomes : @xmath56 to compensate for this new gauge sensitivity of conjugated momenta , the gauge - invariant electric field becomes : @xmath57 any classical gauge - invariant lagrangian gives rise , after legendre transformation , to a hamiltonian which is a functional of @xmath54 , @xmath55 , and @xmath58 fields .
if we add the chern - simons term to the original lagrangian and perform the legendre transformation , we get a new hamiltonian which is expressed in terms of the new gauge - invariant @xmath54 , @xmath55 , and @xmath58 fields through the _ same _ functional as without the chern - simons term . for the special example of the maxwell - chern - simons lagrangian ( [ l_cs ] ) , this functional reads : @xmath59 but although the chern - simons term preserves the hamiltonian functional , it does modify the dynamical properties of the system through a modification of the basic commutation rules between @xmath54 and @xmath55 .
more precisely , we have : @xmath60=-i\hbar(2\nu)\delta ( \mathbf{r } -\mathbf{r^{\prime } } ) \ ] ] so it appears that finding the appropriate deformations of electric field operators plays a crucial role in constructing the hamiltonian version of a chern - simons theory .
we have also seen that such deformations are strongly constrained by the requirement of invariance under local gauge transformations .
this discussion shows that most of the relevant information is implicitely encoded in the additional phase - factor @xmath48 involved in quantum gauge transformations , as in eqs .
( [ gaugetransformedwavefunctional ] ) and ( [ gaugetransformedbasisstate ] ) .
let us now briefly discuss what happens when the model is defined on a two - dimensional space manifold @xmath18 with a boundary @xmath61 . using stokes formula , we may recast the phase factor @xmath48 attached to a gauge - transformation as : @xmath62 in this situation , the phase factor @xmath48 is no longer gauge - invariant , and this implies that two gauge transformations attached to functions @xmath22 and @xmath63 do not commute because : @xmath64 in more mathematical terms , this reflects the fact that the phase - factor @xmath48 , as used in eqs .
( [ gaugetransformedwavefunctional ] ) and ( [ gaugetransformedbasisstate ] ) defines only a projective representation of the classical gauge group , that is : @xmath65 as first shown by witten @xcite , this may be understood in terms of a chiral matter field attached to the boundary of @xmath18 .
an explicit example of boundary degrees of freedom induced by a chern - simons term has been discussed recently in the case of a @xmath66 model on a triangular lattice @xcite . to close this preliminary section ,
it is useful to discuss the case of a finite cylic group @xmath67 . in the @xmath5 case , for a pair of points @xmath68 and @xmath69 , we have a natural group element defined by @xmath70 where the integral is taken along the segment joining @xmath68 and @xmath71 , and @xmath72 is the flux quantum in the model . for a finite group @xmath3 ,
the notion of a lie algebra is no longer available , so it is natural to define the model on a lattice . in a classical gauge theory , each oriented link @xmath73 carries a group element @xmath74 .
we have the important constraint @xmath75 , where @xmath76 is the neutral element of the group @xmath3 . in the quantum version ,
the hilbert space attached to link @xmath73 is the finite dimensional space generated by the orthogonal basis @xmath77 where @xmath78 runs over all elements of @xmath3 . for a lattice ,
the corresponding hilbert space is obtained by taking the tensor product of all these finite dimensional spaces associated to links . in the @xmath67 model , @xmath78 becomes an integer modulo @xmath11 , @xmath79 .
the connection with the continuous case is obtained through the identification . on each link @xmath73
, we introduce the unitary operator @xmath80 which sends @xmath81 into @xmath82 . in the absence of a chern - simons term , the generator of the gauge transformation based at site @xmath83 ( which turns @xmath84 into @xmath85 ) is @xmath86 , where the product involves all the nearest neighbors of site @xmath83 . by analogy with the continuous case ,
the presence of a chern - simons term is manifested by an additional phase - factor whose precise value depends on the lattice geometry and is to some extent arbitrary , since fluxes are defined on plaquettes , not on lattice sites . on a square lattice , a natural choice is to define @xmath87 according to @xcite : @xmath88 where @xmath89 is the oriented loop defined by the outer boundary of the four elementary plaquettes adjacent to site @xmath83 .
this expression has exactly the same structure as eq .
( [ explicitu(f ) ] ) , but somehow , the local magnetic field at site @xmath83 is replaced by a smeared field on the smallest available loop centered around @xmath83 . it has been shown @xcite that a consistent quantum theory can be constructed only when @xmath90 is an integer multiple of @xmath91 .
as discussed in the previous section , the most important step is to construct a non - trivial phase factor which appears in the definition of unitary operators associated to local gauge transformations , generalizing eq .
( [ gaugetransformedbasisstate ] ) . for this , let us first define the operator @xmath92 which is the left multiplication of @xmath78 by @xmath63 , namely : @xmath93 .
for any site @xmath83 and group element @xmath63 , we choose the generator of a local gauge transformation based at @xmath83 to be of the following form : @xmath94 where @xmath95 denotes any nearest neighbor of @xmath83 and @xmath96 is the flux around any of the four square plaquettes , centered at @xmath68 , adjacent to @xmath83 . here , and in all this paper , we shall focus on the square lattice geometry , to simplify the presentation . but adaptations of the basic construction to other lattices are clearly possible . since we are dealing with a non - abelian group , we have to specify an origin in order to define these fluxes , and it is natural to choose the same site @xmath83 , which is expressed through the notation @xmath96 . since we wish @xmath97 to be unitary , we require @xmath98 .
it is clear from this construction that two generators @xmath97 and @xmath99 based on different sites commute , since the phase factors @xmath100 are gauge invariant .
this form is a simple generalization of the lattice chern - simons models for the cyclic groups @xmath67 discussed in the previous section . in this example , for a square plaquette @xmath101 , the flux @xmath102 is equal to @xmath103 modulo @xmath11 , and @xmath63 is simply any integer modulo @xmath11 .
( [ defui ] ) above may be interpreted as : @xmath104 this is a well defined function for @xmath102 and @xmath63 modulo @xmath11 only if @xmath90 is an integer multiple of @xmath105 .
we have not succeeded to cast odd integer multiples of @xmath91 for @xmath90 in the framework of the general construction to be presented now .
this is not too surprising since these models were obtained by imposing special periodicity conditions on an infinite - dimensional hilbert space where @xmath79 can take any integer value @xcite .
our goal here is not to write down all possible chern - simons theories with a finite group , but to easily construct a large number of them , therefore allowing for non - trivial phase - factors when two localized flux excitations are exchanged . as discussed in the introduction , a very desirable property , at least for the sake of finding possible physical implementations ,
is that these deformed generators define a unitary representation of the group @xmath3 .
so we wish to impose : @xmath106 or equivalently : @xmath107 to solve these equations let us first choose a group element @xmath102 .
let us denote by @xmath108 the stabilizor of @xmath102 under the operation of conjugacy , namely @xmath109 belongs to @xmath108 whenever .
this notion is useful to describe the elements of the conjugacy class of @xmath102 .
indeed , we note that if @xmath109 belongs to @xmath108 .
therefore , the elements in the conjugacy class of @xmath102 are in one to one correspondence with the left cosets of the form @xmath110 .
let us pick one representative @xmath111 in each of these cosets .
we shall now find all the functions .
first we may specialize eq .
( [ defgaugegroup ] ) to the case where @xmath109 belongs to @xmath108 , giving : @xmath112 in particular , it shows that the function defines a group homomorphism from @xmath108 to @xmath5 .
once this homomorphism is known , we can specify completely @xmath100 for any group element @xmath63 once the values @xmath113 for the coset representatives are known .
more explicitely , we have : @xmath114 where @xmath115 . finally , eq . ( [ defgaugegroup ] )
yields : @xmath116 let us now show that for any choice of homomorphism , @xmath115 , and unit complex numbers for @xmath113 , eqs .
( [ defchi1 ] ) , ( [ defchi2 ] ) reconstruct a function @xmath100 which satisfies the condition ( [ defgaugegroup ] ) .
any element @xmath117 in @xmath3 may be written as @xmath118 , with @xmath115 .
we have : @xmath119 which is exactly eq .
( [ defgaugegroup ] ) .
note that there are many equivalent ways to choose these functions @xmath120 .
let us multiply the system wave - function by a phase - factor of the form where . under this unitary transformation , @xmath121
is changed into @xmath122 given by : @xmath123 in particular , it is possible to choose the values of @xmath124 so that @xmath125 .
although this does not seem to be required at this stage of the construction , it is also necessary to assume that when the flux @xmath102 is equal to the neutral element @xmath76 , @xmath126
. this will play an important role later in ensuring that the phase - factor accumulated by the system wave - function as a fluxon winds around another is well defined .
( shown here as middle vertical bond ) the surrounding sites are labeled @xmath127 as indicated here .
the fluxes @xmath128 , @xmath129 , ( @xmath130 , @xmath131 ) are counted counterclockwise , starting from site @xmath83 , ( @xmath95 ) , e.g. @xmath132 , width=192 ] our goal here is to construct local gauge invariant operations for basic fluxon processes : fluxon motion , creation of a pair with vacuum quantum numbers , branching of a single fluxon into two fluxons and their time reversions .
these three types of elementary processes can all be derived from a single operation , the electric field operator that at the level of classical configurations , is simply a left multiplication @xmath92 attached to any link @xmath73 . to show this ,
let us consider a pair of two adjacent plaquettes as shown on fig .
[ twoplaquettes ] .
we denote by @xmath128 ( resp .
@xmath133 ) the local flux through the left ( resp .
right ) plaquette , with site @xmath83 chosen as origin .
similarly , we define fluxes @xmath130 and @xmath131 .
changing the origin from @xmath83 to @xmath95 simply conjugates fluxes , according to
. the left multiplication @xmath92 changes @xmath78 into .
therefore , it changes simultaneously both fluxes on the left and right plaquettes adjacent to link @xmath73 .
more specifically , we have : @xmath134 in particular , this implies : @xmath135 note that transformation laws for fluxes based at site @xmath95 are slightly more complicated since they read : @xmath136 this asymmetry between @xmath83 and @xmath95 arises because , so we have : @xmath137 where @xmath138 denotes the right multiplication of @xmath139 by the group element @xmath109 . in the absence of chern - simons term
, @xmath92 commutes with all local gauge generators with the exception of @xmath140 since : @xmath141 we now apply these general formulas to the elementary processes involving fluxes .
suppose that initially a flux @xmath102 was localized on the left plaquette , and that the right plaquette is fluxless .
applying @xmath142 on such configuration gives : @xmath143 which shows that a @xmath102-fluxon has moved from the left to the right plaquette .
a second interesting situation occurs when both plaquettes are initially fluxless .
the action of @xmath144 on such state produces a new configuration characterized by : @xmath145 so we have simply created a fluxon and antifluxon pair from the vacuum .
of course , applying @xmath146 on the final state annihilates this pair . finally , a single flux originaly located on the left plaquette may split into a pair @xmath147 on the left and @xmath148 on the right .
this is achieved simply by applying @xmath149 . in order to incorporate these elementary processes into a hamiltonian chern - simons theory
, we have to modify @xmath92 into an electric field operator @xmath150 by introducing phase - factors so that it commutes for all generators @xmath151 if @xmath152 and that : @xmath153 as explained in the introduction , we shall need only to construct @xmath150 for special types of configurations for which at least one of the four fluxes @xmath154 vanishes . nevertheless , it is useful to first construct @xmath150 for an arbitrary flux background .
the requirement of local gauge - symmetry induces strong constraints on the phase - factors @xmath155 as we shall now show .
these constraints are less stringent when we restrict the action of @xmath150 to the limited set of configurations just discussed . .
this operator transforms the pair of fluxes ( @xmath156 , @xmath157 ) of the left and right plaquettes into the pair ( @xmath158 , @xmath159 ) .
we define the amplitude of the transition between two reference states shown at the top of the figure by @xmath160 .
the amplitude of the process starting from a generic initial state shown in the lower left can be related to @xmath160 using the gauge invariance .
this is done in two steps .
first , the gauge transformation is used to relate the amplitude of the process starting from the generic state to the amplitude of the transition @xmath161 between reference state and a special state shown in the middle right .
second , we use gauge transformation @xmath162 to relate the special state to the reference state on the upper right .
site labeling is the same as in figure 1 .
, width=288 ] we now construct the electric field operators @xmath150 attached to links . in the absence of chern - simons term the electric field operator is equivalent to the left multiplication of the link variable : @xmath163 in this case , @xmath150 commutes with all local gauge generators with the exception of @xmath140 , and eq .
( [ fieldselfcharge ] ) is satisfied .
we have also the group property , namely : @xmath164 the chern - simons term gives phase factors @xmath100 to the gauge generators , so we expect some phase factor , @xmath165 to appear in the electric field operators as well : @xmath166 .
the gauge invariance condition allows us to relate the phase factors associated with different field configurations to each other .
specifically , we introduce the reference states ( shown on the top of fig .
[ electricfieldconstruction ] ) in which only two bonds carry non - trivial group elements .
we define the transition amplitude induced by the electric field between these reference states by @xmath160 . in order to find the phases @xmath167 for arbitrary field configuration
we first transform both the initial and final state by @xmath168 .
this relates the amplitude of the generic process to the amplitude , @xmath161 , of the process that starts with the reference state but leads to the special final state shown in fig .
[ electricfieldconstruction ] middle right .
collecting the phase factors associated with the gauge transformation @xmath169 before and after the electric field moved the flux we get@xmath170 where @xmath128 denotes the flux in the left plaquette counted from site @xmath171 @xmath131 denotes flux in the right plaquette counted from site @xmath95 , and prime refers to the flux configuration after the action of the electric field .
finally , we employ the gauge transformation @xmath162 to relate this special final state to the reference state .
the phase factor associated with this gauge transformation is @xmath172 so @xmath173 in order to express the phase factors , @xmath174 , through the initial field configuration we relate the parameters , @xmath175 , of the gauge transformations to the bond variables @xmath176 by @xmath177 and the fluxes in the left and right plaquettes before and after electric field operator has changed them .
before the electric field operator has acted the fluxes were @xmath178 while afterwards they become @xmath179 combining the preceding equations and using the relation ( [ defgaugegroup ] ) a few times , we get the final expression for the phase - factor , @xmath180 where we have used the definition @xmath181 ( ) to make it more symmetric .
commutation of @xmath150 with @xmath182 , @xmath183 , @xmath184 , @xmath185 , and @xmath186 follows directly from this construction .
it can also be checked directly from ( [ elphasefactor ] ) , using the condition ( [ defgaugegroup ] ) on the elementary phase - factors @xmath100 .
note that sites @xmath83 and @xmath95 play different roles , which is expected because @xmath150 acts by _ left _
multiplication on @xmath78 whereas @xmath187 acts by _ right _ multiplication on the same quantity . .
the state shown at the top and botton left of the figure can be transformed to the reference state ( middle ) in two ways : by applying a gauge transformation @xmath188 on all shaded sites or by applying transformation @xmath189 on site @xmath83 . if @xmath190 and @xmath191 these transformations lead to the same reference state shown in the middle .
furthermore , the action of electric field on these states can be found by making a gauge transformation of the final state shown on the middle right if @xmath192 .
the phases of electric field operator obtained in these two ways should be equal giving additional condition on the phase associated with the gauge field transformation.,width=288 ] although the electric field operator commutes with @xmath182 , @xmath183 , @xmath184 , @xmath185 , and @xmath186 , it does not necessarily commute with @xmath140 even if @xmath193 .
in fact , the requirement of this commutation leads to an important constraint on possible choices of phases @xmath100 .
the appearance of the new constraints becomes clear when one considers a special field configuration shown in fig .
[ electricfieldgaugeinvariance ] .
two identical field configurations shown on the top and bottom left of this figure can be obtained by two different gauge transformations from the reference state if both @xmath156 and @xmath157 commute with @xmath109 : in one case one applies gauge transformation _ only _ at site @xmath83 , in the other one applies gauge transformation on all sites _ except _ @xmath83 .
provided that the resulting states are the same , i.e. @xmath192 , the total phase factor @xmath174 obtained by these two different ways should be the same .
the phase factors associated with these gauge - transformations are the following @xmath194 putting all these factors together we conclude that the gauge invariance of the electric field operator implies that @xmath195 this condition can be further simplified by using the main phase factor equation ( [ defgaugegroup ] ) .
we start by noting that because @xmath109 commutes with @xmath156 , @xmath157 , and @xmath63 , @xmath196 and @xmath197 .
this gives @xmath198 furthermore , combining the identities
@xmath199 and@xmath200 we get @xmath201 this reduces the condition ( [ almostrelation ] ) to a much simpler final form @xmath202 we emphasize that constraint ( [ phasefactorcondition ] ) on the phase factors has to be satisfied only for fluxes satisfying the condition @xmath203 .
although we have derived this condition imposing only the gauge invariance of the electrical field acting on a very special field configuration , a more lengthy analysis shows that it is sufficient to ensure that in a general case@xmath204 the details of the proof are presented in appendix [ amplsol ] . unlike eq .
( [ defgaugegroup ] ) , the constraint ( [ phasefactorcondition ] ) relates functions @xmath100 and @xmath205 for @xmath102 and @xmath206 belonging to _ different _ conjugacy classes . as shown in section [ classmodels ] this constraint
strongly reduces the number of possible chern - simons theories .
note that if both @xmath207 and @xmath208 are solutions of the fundamental relations ( [ defgaugegroup ] ) and ( [ phasefactorcondition ] ) , their product @xmath209 is also a solution .
so there is a natural group structure on the set of possible chern - simons models based on the group @xmath3 , which is transparent in the path - integral description : this means that the sum of two chern - simons action is also a valid chern - simons action .
is this construction also compatible with the group property ( [ electricgroup ] ) ? from eq .
( [ phasefactorcondition ] ) , we obtain : @xmath210 where : @xmath211 it does not seem that @xmath212 are always equal to unity for any choice of @xmath100 that in turns determines the amplitudes @xmath213 ( see appendix [ amplsol ] ) .
but this is not really a problem because this group property does not play much role in the construction of gauge - invariant hamiltonians .
we now specialize the most general constraint arising from local gauge - invariance to the various physical processes which are required for fluxon dynamics . for the single fluxon moving operation , we have , , and , so the condition eq .
( [ phasefactorcondition ] ) is always satisfied . for the pair creation process
, we have , and .
the constraint becomes : @xmath214 finally , let us consider the splitting of an isolated fluxon into two nearby ones .
this is described by , , and .
we need then to impose : @xmath215 it is clear that condition ( [ vacuumpaircond ] ) is a special case of the stronger condition ( [ generalpaircond ] ) .
furthermore , multiplying the conditions ( [ generalpaircond ] ) for pairs of fluxes @xmath216 and @xmath217 we get the most general condition ( [ phasefactorcondition ] ) ; this shows that constraint ( [ generalpaircond ] ) is necessary and sufficient condition for the gauge invariant definition of electric field operator acting on any flux configuration .
let us consider two isolated fluxes carrying group elements @xmath218 and @xmath219 , and move the first one counterclockwise around the second one , as shown on fig .
[ fluxbraiding ] .
this can be done by successive applications of local gauge - invariant electric field operators as discussed in the previous section .
although we wish to work in the gauge - invariant subspace , it is very convenient to use special configurations of link variables to illustrate what happens in such a process .
we simply have to project all special states on the gauge invariant subspace , which is straightforward since the fluxon moving operator commutes with this projector .
the initial fluxes are described by two vertical strings of links carrying the group elements @xmath218 and @xmath219 , see fig .
[ fluxbraiding ] .
when several other fluxes are present , besides the two to be exchanged , it is necessary to choose the location of the strings in such a way that no other flux is present in the vertical band delimited by the strings attached to the two fluxons . during the exchange process ,
the first fluxon collides with the second string , and after this event , it no longer carries the group element @xmath218 , but its conjugate .
after the process is completed , the first flux has recovered its original position , but the configuration of group elements has changed . if we measure the second flux using a path starting at point o shown on fig .
[ fluxbraiding ] we find that it has also changed into .
the final state can be reduced to its template state built from two strings carrying groups elements @xmath220 and @xmath221 by the gauge transformation @xmath222 where @xmath223 is locally constant in the two following regions : the core region inside the circuit followed by the first fluxon ( @xmath224 ) , the intermediate region delimited by the two initial vertical strings and the upper part of the circuit ( @xmath225 ) .
note that because we do not wish to modify external fluxes , we can not perform gauge transformations in the bulk , outside of these regions .
( right ) and @xmath219 ( left ) connected to the edge by the strings of @xmath218 ( purple arrow ) or @xmath219 ( black ) group elements .
moving the right flux around the left leaves behind a string of @xmath218 elements until its path crosses the vertical string of @xmath219 . upon crossing the string changes to the string of @xmath226 ( red arrows ) .
performing the gauge transformations with @xmath227 on all sites indicated by full ( red ) dots and with @xmath228 on sites indicated by empty ( orange ) dots reduces the configuration to the template shown in the last pane with the new fluxes @xmath229 and @xmath230 ( see text , eq .
( [ braiding ] ) ) .
, width=192 ] group elements @xmath231 and @xmath232 have to satisfy the following conditions , which may be obtained readily upon inspection of fig .
[ fluxbraiding ] @xmath233 these equations are mutually compatible , and we get . since fluxes are present in the core region , they will contribute a phase - factor @xmath234 when the gauge transformation from the template to the actual final state is performed .
the final result may be summarized as follows : @xmath235 the new phase - factor @xmath234 appears not to depend on the detailed path taken by the first fluxon , but just on the fact that it winds exactly once around the second . in this sense
, our construction really implements a topological theory .
the dihedral groups @xmath10 are among the most natural to consider , since they contain a normal cyclic group @xmath67 which is simply extended by a @xmath66 factor to form a semi - direct product . in this respect
, one may view this family as the most weakly non - abelian groups .
@xmath10 can be described as the isometry group of a planar regular polygon with @xmath11 vertices .
the @xmath67 subgroup corresponds to rotations with angles multiples of @xmath236 .
we shall denote by @xmath237 the generator of this subgroup , so @xmath237 may be identified with the @xmath236 rotation .
@xmath10 contains also @xmath11 reflections , of the form @xmath238 .
the two elements @xmath237 and @xmath239 generate @xmath10 , and they are subjected to the following minimal set of relations : @xmath240 this last relation shows that indeed @xmath10 is non - abelian .
the next useful information about these groups is the list of conjugacy classes .
if @xmath241 , @xmath10 contains @xmath242 classes which are : , , ... , , . if @xmath243 , there are @xmath244 classes whose list is : , , , ... , , ,
as shown in section [ sectiongenerators ] , in order to construct possible phase factors associated to gauge transformations , we need to know the stabilizors of group elements for the conjugacy operation of @xmath10 acting on itself .
here is a list of these stabilizors : for @xmath11 odd : @xmath245 for @xmath11 even : @xmath246 finally , we need to choose homomorphisms from these stabilizors into @xmath5 . in the case of a cyclic group @xmath67 generated by @xmath237 , homomorphisms @xmath247 are completely determined by @xmath248 , so that , with the constraint : . for the group @xmath10 itself , we have : , with the following constraints : @xmath249 these are direct consequences of generator relations ( [ genrel1]),([genrel2]),([genrel3 ] ) .
again , the parity of @xmath11 is relevant . for @xmath11 odd , , which leaves only two homomorphisms from @xmath10 into @xmath5 . for @xmath11 even , , and there are four such homomorphisms .
the last possible stabilizor to consider is the four element subgroup of @xmath250 , .
this abelian group has four possible homomorphisms into @xmath5 , which are characterized as follows : @xmath251 let us first consider conjugacy classes of the form .
since the stabilizor of @xmath252 for the conjugacy action of @xmath10 is @xmath67 , we have : , where . choosing , we have for fluxes in the cyclic group generated by @xmath237 : @xmath253 for the remaining conjugacy class , we have , with . choosing , and using eqs .
( [ defchi1 ] ) and ( [ defchi2 ] ) , we obtain : @xmath254 all these possible phase - factors satisfy the following property : @xmath255 so that eq .
( [ vacuumpaircond ] ) is always satisfied .
so no new constraint is imposed by the requirement to create or annihilate a pair of fluxons .
what about the stronger condition eq .
( [ generalpaircond ] ) ?
its form suggests that we should first determine pairs of fluxes such that their stabilizors @xmath256 and @xmath256 have a non - trivial intersection .
this occurs if both @xmath147 and @xmath148 are in the @xmath8 normal subgroup generated by @xmath237 , or if .
the second case simply implies @xmath257 , which is not a new condition . in the first case , choosing @xmath109 in @xmath8 as well shows that @xmath258 is a homomorphism from @xmath8 to @xmath5 with respect to both @xmath102 and @xmath109 .
this is satisfied in particular if @xmath259 itself is a group homomorphism in both arguments .
this sufficient ( but possibly not necessary ) condition simplifies algebraic considerations ; it can be also justified from physical argument that the theory should allow for the sites with different number of neighbours , @xmath260 , which would change @xmath261 .
replacing this the constraint on @xmath258 by the constraint on @xmath259 , we get @xmath262 therefore , the class of possible phase - factors ( which is stable under multiplication ) is isomorphic to the group .
this group of phase - factors is identical to the group of possible chern - simons actions since in this case , @xcite .
very likely , the coincidence of these two results can be traced to the absence of projective representations for @xmath10 for @xmath11 odd @xcite : as explained above , the projective representations are not allowed in our construction but are allowed in the classification of all possible chern - simons actions @xcite .
the conjugacy classes of the form behave in the same way as for @xmath11 odd .
so writing @xmath243 , we have : @xmath263 the conjugacy class @xmath264 is special since its stabilizor is @xmath10 itself . as discussed in section [ genpropdihedral ] above , there are four homomorphisms form @xmath10 into @xmath5 that we denote by : @xmath265 let us now turn to with the corresponding stabilizor equal to .
as seen in section [ genpropdihedral ] , the four possible homomorphisms may be written as : @xmath266 from this , we derive the expression of @xmath267 , in the following form ( ): @xmath268 furthermore , because , eq . ( [ defchi2 ] ) implies : @xmath269 the last conjugacy class to consider contains @xmath270 , with the stabilizor . in this case
, we may set ( ): @xmath271 here again , the constraint eq .
( [ vacuumpaircond ] ) is always satisfied . to impose eq .
( [ generalpaircond ] ) , we have to consider pairs of fluxes such that @xmath272 is non - trivial .
as before , choosing @xmath147 and @xmath148 in the @xmath67 subgroup imposes @xmath273 , with @xmath274 .
a new constraint arises by choosing @xmath275 and @xmath276 . in this case
, @xmath277 belongs to their common stabilizor .
( [ generalpaircond ] ) implies : @xmath278 but using eq .
( [ defgaugegroup ] ) this yields : @xmath279 therefore we have the constraint : , which enables us to simplify drastically the above expression for phase - factors : @xmath280 and @xmath281 but eq .
( [ ctauconstraint ] ) now implies that @xmath282 for any @xmath283 , which is satisfied only when @xmath284 . specializing to @xmath285
, we see that the common stabilizor contains now two more elements , namely @xmath286 and @xmath287 .
( [ ctauconstraint ] ) now requires that : @xmath288 it is then easy to check that considering the common stabilizor of @xmath289 and @xmath286 , which may contain two or four elements , does not bring any new constraint .
finally , @xmath290 belongs to the @xmath291 group and among the three binary variables @xmath292 , @xmath293 , and @xmath294 , only two are independent .
therefore , the set all all possible phase - factors for the @xmath250 group is identical to .
this contains only half of @xmath295 which is equal to @xcite . but
@xmath250 admits non trivial projective representations since @xmath296 which can not appear in our construction .
the important result is that in spite of this restriction , we get a non - trivial subset of the possible theories also in this case . using the general expression ( [ adiabaticphase ] ) , and the above description of possible phase - factors
, we may compute the adiabatic phase induced by a process where a fluxon @xmath218 winds once around another fluxon @xmath219 .
the results are listed in the last column of table [ finaltable ] , and they are valid for _ both _ parities of @xmath11 .
[ c]|c|c|c|c||c|c|c|@xmath218 & @xmath219 & @xmath220 & @xmath221 & @xmath297 & @xmath298 & @xmath234 + @xmath252 & @xmath299 & @xmath252 & @xmath299 & @xmath300 & @xmath301 & @xmath302 + @xmath289 & @xmath299 & @xmath303 & @xmath304 & @xmath305 & @xmath306 & @xmath307 + @xmath252 & @xmath308 & @xmath309 & @xmath310 & @xmath300 & @xmath311 & @xmath312 + @xmath289 & @xmath308 & @xmath313 & @xmath314 & @xmath305 & @xmath305 & @xmath311 +
generally , in order to appear as a low energy sector of some physical hamiltonian , the chern - simons gauge theory has to involve gauge transformations that depend only on a local flux configuration .
furthermore , to be interesting from the view point of quantum computation , the theory should allow for a local gauge invariant electric field operator that moves a flux or fuses two fluxes together .
here we have analyzed non - abelian gauge models that satisfy these general conditions ; our main result is the equation ( [ generalpaircond ] ) for the phase factor @xmath247 associated with a gauge transformation .
furthermore , we have computed the flux braiding properties for a given phase factor that satisfies these conditions . finally , we have applied our general results to the simplest class of non - abelian groups , dihedral groups @xmath0 .
the fluxon braiding properties in these groups are summarized in table i. inspection of the table
i shows that even for the smallest groups , chern - simons term modifies braiding properties in a very non - trivial manner and gives phases to the braiding that were trivial in the absence of the chern - simons term . in the scheme@xcite where the pair of two compensating fluxes @xmath315 are used to encode one bit ,
the transformations allowed in the absence of chern - simons term are limited to conjugation . in the presence of chern - simons term the braiding of such bit with the controlled flux results in a richer set of transformations that involve both conjugation by group elements and phase factors
( see table i ) but does not change the state of the flux as it should .
we hope that this will make it possible to construct the universal quantum computation with the simplest group @xmath12 that does not involve operations that are difficult to protect ( such as charge doublets ) .
the implementation of the microscopic hamiltonians discussed in this paper in a realistic physical system is a challenging problem .
the straightforward implementation would mean that the dominant term in the microscopic hamiltonian is @xmath316 so that all low energy states are gauge invariant .
this is not easy to realize in a physical system because operator @xmath317 involves a significant number of surrounding bonds .
we hope , however , this can be achieved by a mapping to the appropriate spin model as was the case for abelian chern - simons theories ; this is the subject of the future research .
* acknowledgments * we are thankful to m. grbig and r. moessner for useful discussions .
li is thankful to lpthe , jussieu for their hospitality while bd has enjoyed the hospitality of the physics department at rutgers university .
this work was made possible by support from nato clg grant 979979 , nsf dmr 0210575 .
( shown as black dot ) can be expressed via the phase factors @xmath213 and @xmath318 characterizing the transition between reference states ( bottom and top ) .
this involves gauge transformations on all grey ( blue ) sites .
the gauge invariance implies that the total phase factor accumulated when moving around the diagram is unity which implies the condition on the ratio @xmath319 . as in text ,
@xmath320 stands for a triple @xmath321 , and @xmath322 is obtained by conjugating the components of @xmath320 by @xmath109 .
, width=288 ] in the absence of additional constraints on the gauge transformation phase factors , @xmath100 , the definition of the electric field operator implies that it is gauge invariant under gauge transformations on all sites except @xmath83 while the the gauge transformation on the latter gives @xmath323 with @xmath324 here the notations are those of fig .
[ electricfieldconstruction ] , where state @xmath325 corresponds to the initial state on the figure .
the computation leading to this equation is depicted on fig [ electricfieldgaugeinvariancegeneral ] .
the question we have to address now is whether it is possible to choose amplitudes @xmath160 so that the function @xmath326 is equal to unity . to simplify notations ,
let us denote triples by a single label @xmath320 .
the group @xmath3 acts on these triples according to : @xmath327 let us also introduce the function @xmath328 defined as : @xmath329 the condition @xmath330 now reads : @xmath331 now we show that solutions @xmath213 exist if and only if the function @xmath328 satisfies the following conditions : @xmath332 note that the second condition is very similar to the fundamental property ( [ defgaugegroup ] ) which enforces the existence of a local gauge symmetry group .
in fact , using ( [ defgaugegroup ] ) , we can check that property ( [ grouprel ] ) always holds .
property ( [ condstab ] ) is a new and quite strong condition imposed on gauge phase - factors @xmath100 ; it is equivalent to the condition ( [ phasefactorcondition ] ) which in the main text has been derived as _ necessary_. we now show that this condition is also _
sufficient_. let us consider the equation : @xmath333 where @xmath334 and @xmath213 are complex numbers with unit modulus , @xmath320 belongs to a set @xmath335 on which the group @xmath3 acts , and @xmath336 . here
, the function @xmath334 is supposed to be known , and we are looking for `` amplitudes '' @xmath213 .
let us denote by @xmath337 the stabilizor of @xmath320 , namely it is composed of all the group elements @xmath109 such that from eq .
( [ ampleqap ] ) , we get the two conditions : @xmath338 the second equation is obtained from : @xmath339 note that these two conditions imply that : @xmath340 let us now show that when these two conditions are satisfied , we can always reconstruct a system of amplitudes solving eq .
( [ ampleqap ] ) . from the form of this equation
, we see that the various orbits in @xmath335 for the action of @xmath3 remain uncoupled .
let us then concentrate on one orbit generated by a fixed given element @xmath320
. elements of this orbit are in one to one correspondence with the left cosets @xmath341 of @xmath3 , since @xmath342 when @xmath343 .
let us choose in each coset a representative element @xmath111 .
our problem is to find amplitudes @xmath344 knowing @xmath213 .
let us define : @xmath345 from eq .
( [ condc ] ) , we have for any @xmath63 in @xmath3 , since @xmath63 may always be written as @xmath346 with @xmath343 . now eq .
( [ ampleqap ] ) is an immediate consequence of the condition ( [ condb ] ) . to conclude , when conditions ( [ conda ] ) and ( [ condb ] ) are satisfied , solutions of eq .
( [ ampleqap ] ) can be constructed independently on each orbit of @xmath335 for the action of @xmath3 . on a given orbit ,
the solution is unique , up to the choice of @xmath213 for one particular element of this orbit .
the distinguishing feature of a torus is the appearance of non - trivial closed loops that are classified according to their winding numbers associated with two fundamental cycles @xmath347 and @xmath348 , chosen with a common origin @xmath349 . as a consequence ,
even if local fluxes vanish on all plaquettes , the fluxes @xmath350 and @xmath351 associated to @xmath347 and @xmath348 may not vanish .
these global degrees of freedom are the origin of the topological degeneracies exhibited by a large class of gauge - invariant models on a closed space with non - trivial topology @xcite .
how are these degeneracies affected by the presence of a chern - simons term
? precise formulas for the hilbert - space dimension of the _ pure _ chern - simons theory with a finite gauge group @xmath3 have been given @xcite in terms of the action via functions in @xmath4 .
in this paper we consider a much larger hilbert space induced by all allowed classical gauge configurations , @xmath352 .
we may expect to recover the pure chern - simons hilbert - space ( which dimension is independent of the system size ) by projecting the full hilbert - space onto the gauge - invariant ground - state of some local gauge - invariant hamiltonian .
this goes beyond the scope of the present paper because our goal here is classification and basic properties of such local gauge invariant hamiltonian .
however , it is instructive to understand how , in the present formalism , a chern - simons term affects the gauge dynamics in the topological fluxless sector on a torus . and @xmath348 are depicted as black dashed lines .
the vertical string of horizontal blue lines represents links carrying the group element @xmath350 .
similarly , the horizontal string of vertical red lines represents links carrying @xmath351 . on all other links @xmath73 on the square lattice @xmath353 , so we have not represented them not to overcrowd the figure . , width=192 ] starting from the trivial configuration @xmath353 , we may induce a new fluxless state in the bulk but with a non - trivial @xmath350 along @xmath354 by creating a string of parallel horizontal links carrying each the group element @xmath350 as shown on fig .
[ figtorustemplate ] .
this string may be viewed as the result of a succession of elementary processes discussed in section [ sectionprocesses ] : first the creation of a @xmath350 , @xmath355 fluxon pair , then the motion of one of them along @xmath348 and finally annihilation of the two fluxons .
as we have seen , each of these processes may be described by a gauge - invariant operator .
let us denote by @xmath356 the product of all these operators involved in the creation of the string .
similarly , we define @xmath357 .
more generally , we may start from a state already characterized by a pair of fluxes @xmath358 .
since @xmath347 and @xmath348 commute up to homotopy , we require that @xmath350 and @xmath351 commute in @xmath3 .
the string picture allows us to construct operators @xmath359 and @xmath360 such that : @xmath361 denoting by @xmath362 any state with the pair of fluxes @xmath358 , we show below that @xmath359 and @xmath360 do not commute , but their actions on a state @xmath362 are related by phase - factors : @xmath363 note that in order for all four states , , , to be defined , the following constraints have to be imposed : @xmath364 this relation ( [ eqphasestrings ] ) is reminiscent of the phase factors involved in permuting electrical field operators attached to adjacent links for abelian models with a chern - simons term @xcite .
let us now briefly explain how to derive the relations ( [ defininglambda ] ) and ( [ eqphasestrings ] ) expressing the commutation rules between @xmath359 and @xmath360 .
the first step is to choose template states for a pair of fluxes @xmath358 along elementary cycles @xmath347 and @xmath348 .
these are depicted on fig .
[ figtorustemplate ] .
let us now compare the action of and on this template .
the situation is illustrated on fig .
[ figstringcreation ] .
note that we do not need explicitely the values of the fluxon hopping amplitudes @xmath160 , but simply the fact they are local and that fluxon moving operators commute with gauge transformation .
the value of @xmath365 obtained in this way ( see eq . ( [ eqphasestrings ] ) ) is valid not only for our template state @xmath362 but for any state deduced from it by a gauge transformation . | we construct the local hamiltonian description of the chern - simons theory with discrete non - abelian gauge group on a lattice .
we show that the theory is fully determined by the phase factors associated with gauge transformations and classify all possible non - equivalent phase factors .
we also construct the gauge invariant electric field operators that move fluxons around and create / anihilate them .
we compute the resulting braiding properties of the fluxons .
we apply our general results to the simplest class of non - abelian groups , dihedral groups @xmath0 . |
the association of powerful radio galaxies with luminous extended emission line nebulae is well established ( e.g. , heckman _ et al . _
1982 , meisenheimer & hippelein 1992 , baum & mccarthy 2000 ) .
there has been great debate over whether nuclear photo - ionization or shock heating is the dominant source of ionization .
however , it is becoming clear that both mechanisms are present in the extended nebulae , e.g. , @xcite . in principle
it should be possible to use the properties of the shocked gas to study the nature of the jet - cloud interaction and constrain the physics of the propagation of the radio source . here
we present a high resolution hst / wfpc2 imaging study of three powerful radio galaxies in which previous ground based imaging and spectroscopy has revealed the presence of shocked gas in the extended nebula ( clark _ et al . _ 1997 , 1998
; villar - martn _ et al . _ 1999
; tadhunter _ et al . _ 2000
; solorzano - iarrea & tadhunter 2003 ) .
we detail the data reduction and analysis techniques in 2 .
the properties of the sources are summarized in tables 1 & 2 . in
3 we present the results of the emission line imaging and a comparison with the radio source morphology . the results are summarized in 4 . throughout the paper ,
we have used h@xmath1=75 km s@xmath2mpc@xmath2and q@xmath1=0.0 .
we observed the redshifted [ oiii]@xmath05007 and [ oii]@xmath03727 emission lines using the hst / wfpc2 linear ramp filters ( lrf ) .
the lrf are divided into four parallel strips each with the central wavelength varying across each strip by about 6 percent @xcite . in general for a given wavelength , the effective field of view is about 13 .
the object was placed in that part of the chip which corresponds to the desired wavelength .
in addition to the narrow - band emission line images taken with the ramp filters , intermediate - band continuum images were taken using the f467 m and f547 m filters , in order to facilitate accurate continuum subtraction and allow examination of the detailed continuum structures .
the central wavelengths and bandwidths of the medium band filters were chosen to avoid bright emission lines .
the parameters of the hst observations are given in table 3 .
the data were calibrated with standard software in stsdas . due to the fine vignetting of the lrf filters , flats obtained from nearby narrow band filters
were used for flat fielding ( biretta , baggett & noll 1996 ) .
the drizzle algorithm @xcite was used for combining dithered images .
the algorithm is also very efficient at rejecting cosmic rays .
these dithered images were used in the subsequent analysis .
the tasks in the synphot package were used for flux calibration . adopting an input `` average '' emission line profile , we used calcphot to estimate the number of counts expected for a gaussian signal
. the characteristics of this input were then used to convert the image from counts to flux units .
the photometric calibration of the lrf data should be accurate to @xmath3 ( biretta _ et al . _
1996 ) .
it was necessary to register the lrf and continuum images using wregister , since the source is not located at the same pixels in these images .
this task performs a spatial transformation so that the wcs coordinates of the input image match the coordinates of the reference image at the same logical pixel coordinates .
this first shift is good to a few pixels .
we then manually shifted by a few pixels to register the nucleus on the images .
our images do not show significant dust , so the location of the nucleus was obvious on the continuum and line images .
we used calcphot to estimate the flux expected through each of the filters for a standard elliptical galaxy template .
we used the ratios of the estimated fluxes to scale the medium band image to subtract the continuum contribution from the lrf images . over much of the extent of the nebulae , the continuum is weak or undetected and the stellar continuum correction to the lrf images ranges from a few percent to less than one percent of the counts in the lrf image . in a few locations , e.g. , the nuclei of 3c277.3 and pks2250 - 41 , the correction to the lrf image is as large as 10 - 20% . in principle
, there could be a contribution to the continuum from e.g. , scattered nuclear light or a young stellar population ( e.g. , chambers & miley 1990 ) . if this were the case , the result could be an increase in the correction to the lrf images , but over most of the nebula the correction is still a small fraction of the counts in the lrf image .
radio observations images for all three sources were obtained with the vla in the a configuration .
in addition a 36 hour observation on 3c 277.3 was obtained using merlin2 ( table 4 ) .
calibration and data reduction was achieved using standard techniques in the nrao aips software .
observations of a bright point source calibrator were used to determine the antenna gains and phases and to place the observations in the radio reference frame with an accuracy of @xmath4 .
the flux density scale was determined using observations of 3c286 .
after calibration , the data were deconvolved using the clean algorithm , and the instrumental gains and phases were further corrected using self - calibration .
the vla and merlin2 data for 3c277.3 were combined to produce a final image .
the average signal for the background for each of the emission line images was obtained .
this background level was subtracted from the corresponding images .
the emission line images were smoothed with a gaussian with a fwhm of 2 pixels in order to increase the surface brightness sensitivity .
the smoothing also mitigates the consequences of any small errors in the image registration .
the pixels with values less than 3@xmath5 were masked .
the masked [ oiii ] image was divided by the masked [ oii ] , to obtain the ratio image . in pixels where the [ oii ] flux was masked , but [ oiii ] was not , we divided the [ oiii ] flux by the 3@xmath5 value for the unmasked [ oii ] image so as to obtain a lower limit on the ratio in those regions .
the optical images were rotated so that they were aligned with the radio images and then the optical and radio nuclei ( which are visible on all images ) were registered and the images overlayed .
we estimate that the radio and optical images are registered to within 1 pixel ( @xmath6 ) .
the radio and optical images and various overlays are presented in figures 1 - 12 .
a small but non - negligible redenning correction @xcite , was applied for all three galaxies ( table 1 ) . for 3c 171 and pks 2250 - 41 ,
the reddening correction includes galactic as well as internal contributions and was determined using observed h@xmath7/h@xmath8 ratios @xcite . for 3c 277.3
the h@xmath7/h@xmath8 and h@xmath9/h@xmath8 measured by solorzano - iarrea & tadhunter ( 2003 ) , are consistent with case b suggesting that the redenning is negligible .
3c 171 is a well studied fr ii radio source at redshift of 0.2384 ( e.g. , heckman _
et al . _
1984 , baum _ et al . _
1988 , blundell 1996 , clark _ et al . _
1998 , tadhunter _ et al . _ 2000 , solorzano - iarrea & tadhunter 2003 ) .
the radio image shows a faint nucleus and two hot - spots and considerable extended diffuse emission which includes large ` plumes ' which extend roughly perpendicular to the radio lobes @xcite .
et al . _ ( 1984 ) and baum _ et al . _ ( 1988 ) found that the inner structure of the powerful radio source ( within the hotspots ) 3c 171 is co - spatial with an extended emission line nebula .
et al . _ ( 2000 ) obtained deep h@xmath7 images which revealed that there is faint diffuse gas extending beyond the radio source and that the morphology of the emission line gas is closely related to the radio morphology , but does not exhibit the ionization cone " morphology expected for simple agn photo - ionization .
long slit @xcite and integral field spectroscopy @xcite revealed ( 1 ) complex kinematics in the gas including line splitting , and ( 2 ) lower ionization gas associated with the hot - spots , and ( 3 ) an anti - correlation between line width and ionization state in the gas .
hardcastle ( 2003 ) suggested that the depolarization of the radio emission is produced by a warm shocked medium .
taken together , these observations are consistent with the emission line gas having been shocked by the expanding radio source , e.g. clark _ et al . _ ( 1998 ) .
we present our wfpc2 images and radio images in figures 1 - 4 .
values of the [ oiii]/[oii ] line ratio are given in table 6 for selected regions .
the medium band continuum image shows some very faint diffuse emission aligned along the radio axis .
comparison of the hst and radio images ( figures [ 171_oii_radio ] , [ 171_oiii_radio ] ) shows that the high surface brightness emission line gas tends to follow the central ridge line of the radio source . however , there are some asymmetries in the source .
we detect [ oiii ] but not [ oii ] above a level of @xmath10 ergs s@xmath2@xmath11 arcsec@xmath12 associated with the western hot spot ( the [ oii ] emission line gas begins just interior to the hot - spot ) .
however , the lower resolution ground based observations by solorzanno - iarrea & tadhunter ( 2003 ) show traces of [ oii ] , [ oiii ] and h@xmath8 line emission in this region . on the eastern side
there is [ oii ] in a band across the hot spot , while the [ oiii ] extends mainly south of the lobe .
we find low ratios of [ oiii]/[oii ] @xmath130.5 associated with the gas near the hot - spots , and higher values @xmath14 about midway between the lobes and the nucleus .
the extended structures in this object show a marked similarity with the aligned structures in high redshift radio galaxies .
the lack of any line emission for the western hot - spot is surprising , given that the radio morphology suggests that the jet has hit something there @xcite .
one explanation might be that the shock driven through the warm / cool gas at the western hot - spot may be so strong / fast that the gas is heated to high enough temperatures that the cooling time becomes long ( i.e. the gas has not yet cooled to @xmath151 - 2@xmath1610@xmath17 k , so that the optical emission lines become visible ) . the anti - correlation between line width and ionization noted by clark _
et al . _ ( 1998 ) suggests that the compressed , high velocity post - shock gas has a low ionization state , whereas the kinematically quiescent gas has a high ionization state .
it is likely that the high surface brightness regions of [ oii ] emission map the kinematically disturbed post - shock gas , whereas the [ oiii ] emission is a combination of mostly precursor and some post - shock gas .
the striking differences between the emission line structures of the two lines may therefore reflect differences in the spatial distributions of precursor and cooling post - shock gas ; in this context it is not surprising that the [ oii ] is stronger in the eastern lobe than the [ oiii ] , and that [ oiii ] dominates in the agn - photoionized nuclear regions .
coma a is classified as intermediate between fri and frii radio galaxies , displaying very broad and diffuse bubble - like lobes .
it is relatively nearby with a redshift of 0.0857 .
et al . _ ( 1981 ) first detected optical continuum and emission lines associated with the extended radio lobes .
there is an extensive network of emission line filaments which seem to wrap around the radio lobes @xcite .
van breugel _
_ ( 1985 ) suggested that the southern jet is deflected by a massive cloud at the location of knot k1 .
et al . _ ( 2002 ) detect the 21 cm line of hi in absorption against the extended radio lobes of coma a. they suggest that the source is expanding into a gaseous disk , presumably acquired in a merger which triggered the radio activity .
the kinematics of the emission line gas are consistent with interaction with the radio source @xcite .
the hst and radio images are given in figures 5 - 8 .
values of the [ oiii]/[oii ] line ratio are given in table 6 for selected regions .
the medium band continuum image shows a very prominent nucleus and the faint continuum emission associated with the northern knot k1(fig .
[ 277_cont_radio ] ) .
the peak of the continuum emission is shifted just to the east of the radio peak of k1 . in our narrow band lrf images
we detect the emission line gas associated with the nucleus and the filament which runs mainly n - s and to the east of the two radio knots ( figs .
[ 277_oiii_radio ] , [ 277_oii_radio ] ) .
we do not detect any line emission directly associated with knot k2 .
we see a peak in the line emission just to the east of knot k1 .
the line emission in this region exceeds that from the nucleus . the observed [ oiii]/[oii ] ratio also peaks just east of k1 with a relatively high value of @xmath18 consistent with the suggestion of solorzano - iarrea & tadhunter ( 2003 ) that the emission lines in this feature are dominated by photoionized precursor emission .
the linear structures visible in the high resolution radio images are very interesting .
they are suggestive of shocks that have been driven back into the jets as a consequence of the strong interaction with the emission line cloud .
the fact that they are not exactly perpendicular to the jets , but misaligned by @xmath1530 degrees to perpendicular , may be significant , since oblique shocks have been proposed as a mechanism for deflecting jets , and the jet is indeed significantly deflected in 3c277.3 .
the modeling work of tingay ( 1997 ) demonstrates that it is possible to produce a significant deflection for an obliquity of 30 degrees , although as he notes , it may be difficult to produce a deflection as large as that seen in 3c277.3 , unless the jet is inclined towards the observer , so that the real jet deflection is less than measured .
the nature of the jet - cloud interaction appears somewhat different to those associated with the other two sources . whereas in 3c171 and pks2250
- 41 the _ lowest _ ionization features are closely associated with the detailed radio features ( knots , hot - spots ) , in 3c277.3 the highest ionization feature _ within the cloud _ is closely associated with the radio knot . while this close association within 0.2 arcseconds in our hst observations
suggests that the jet is ionizing the cloud , it is significant that , unlike the other two sources and also the extended radio lobe in 3c277.3 itself @xcite , the line emission associated with the jet - knots in 3c277.3 shows no sign of kinematic disturbance .
all this is consistent with the idea that , in this case , we are observing pure shock photoionized precursor emission ; the shock driven through the warm ism may be so fast that the shock gas has not yet cooled sufficiently to radiate optical emission lines , or the post shock cloudlets may have been shredded / dispersed by hydrodynamic interaction with the hot post - shock wind before they had a chance to cool ( e.g. , klein , mckee & colella 1994 ) .
this may be testable with x - ray observations .
however , we can not entirely rule out the idea the precursor gas in the cloud is photoionized by a narrow blazar beam from the agn which points exactly in the jet direction ; the ionization gradients across the cloud are consistent with this idea provided that the lorentz factor of the jets close to the nucleus is fairly high ( in order to give a narrow beam ) .
the optical continuum emission associated with the warm emission line cloud is also interesting .
_ ( 1981 ) detected significant optical polarization ( @xmath19 with a position angle of @xmath20 ) that is aligned perpendicular to the radio axis .
on the basis of the optical / radio spectral index and polarization they interpreted the optical continuum as synchrotron emission from the same knots that emit the radio emission .
however , in this case it is surprising that the continuum emission visible in our hst images appears to be significantly displaced to the east of peak of the high frequency radio emission . on the other hand ,
the polarization properties of the continuum emission , and its morphological similarity with the [ oiii ] emission , are consistent with the idea that it represents light from the blazar beam of the agn scattered into our line of sight ( very similar to pks2152 - 69 : see tadhunter _
et al . _
1988 , di serego alighieri _
et al . _
this would also support the idea that the cloud is photoionized by the blazar beam .
pks 2250 - 41 is a frii source at a redshift of 0.308 .
the first detailed observations by @xcite showed large scale emission line arcs or filaments associated with the radio lobes .
both the kinematics and the ionization of the gas appear consistent with a scenario in which the radio source is interacting with the emission line gas @xcite .
et al . _ ( 1997 ) find that the radio emission from the western lobe has very low polarization consistent with depolarization by surrounding / mixed gas .
our 15 ghz vla observations resolve structure in the lobes and reveal evidence for secondary hot - spots ( e.g. , fig .
[ pks_oii_radio ] ) .
we also ( just ) detect the nucleus in the radio which allows improved registration of the optical and radio images .
the hst and radio images are given in figures 9 - 12 .
values of the [ oiii]/[oii ] line ratio are given in table 6 for selected regions .
note that unfortunately the location of the redshifted wavelength of the [ oii ] line on the lrf lies close to the edge of the chip .
thus , the coverage of the [ oii ] line does not include the eastern lobe
. this may not be significant as ground based imaging by clark _
et al . _ ( 1997 ) does not show any emission line gas associated with this lobe .
our lrf images show line emission from the nuclear region and from the bright filament which points toward the companion galaxy to the north - east .
there is also a hint of much more diffuse line emission which is not well detected in these observations . as noted by tadhunter _
et al . _ ( 1994 ) , the emission line gas in the western lobe takes the form of an arc which wraps around the outer boundary of the lobe ( fig .
[ pks_oiii_radio ] ) .
the [ oiii ] extends in a roughly 180@xmath21 semi circle around the front side of the lobe .
the [ oii ] is somewhat smaller in extent and is concentrated closer to the source axis and appears nested " within the [ oiii ] arc .
the [ oii ] emission from this nested region exceeds that from the nucleus .
the [ oiii]/[oii ] ratio in the lobe is low ( @xmath22 ) and is somewhat higher at the leading edge of the lobe and declines slightly with distance inwards towards the nucleus .
the differences between the detailed [ oiii ] and [ oii ] emission line morphologies in the lobe can be explained in terms of different proportional contributions of precursor , agn - photoionized , and lower ionization , kinematically - disturbed post - shock gas , with agn photoionization dominating close to the nucleus , shock ( photo)ionization of the precursor emission dominating in the extended arc , and cooling post - shock gas dominating the [ oii ] emission in the lobe .
this is consistent with the interpretation of clark _
_ ( 1997 ) and villar - martn _ et al . _ ( 1999 ) .
the medium band continuum image shows a possible companion galaxy to the north ( clark _ et al .
_ 1997 ) and emission associated with the secondary hot - spot in the western lobe .
a spur of the continuum emission appears to point toward the companion while the companion itself is elongated in the direction of the nucleus .
the position angles of the spur in the continuum , appears to be aligned with the major axis of the companion to within 6 degrees .
the alignment is even better between the pa for the major axis of the companion and the filament of pks2250 - 41 .
table 5 summarizes the position angles of each of the distinctive features of the companion and pks2250 - 41 .
recent vlt observations of the putative companion galaxy show balmer absorption lines at the same redshift as pks2250 - 41 indicating that it is indeed a companion in the same group and that it has had recent star formation ( tadhunter , in preparation ) .
the spatial coincidence between the optical continuum emission and secondary hot - spot in the western lobe is striking .
one possibility is that the secondary hot - spot in fact represents compact radio emission from the nucleus of a companion galaxy that the jet is colliding with ( consistent with the jet - galaxy collision idea presented in clark _ et al . _
1997 ) . in this case
, the optical light represents the stellar light of the bulge of this companion galaxy to the west .
alternatively , both the optical and radio emission might be explained in terms of synchrotron emission from the secondary hot - spot .
however this possibility is ruled out as a dominant source of emission , due to the lack of optical polarization ( @xmath23 , see shaw _
et al . _
1995 ) .
the highly collimated nature of the emission line structure in the nuclear region is reminiscent of the structures aligned along the radio axis in high redshift radio galaxies ( best _ et al .
_ 2000 ) , yet in this case the structure is misaligned from the radio axis by 25 degrees . while radio and optical structures in high z radio galaxies are rarely in perfect alignment , a deviation of this size is seen in less than 25% of powerful radio sources studied by chambers , miley & van breugel ( 1987 ) .
moreover , in contrast to the high - z objects , the emission line kinematics of the linear structure are quiescent .
however , the fact that this structure points towards the north - eastern companion ( and the companion points towards it ! ) , suggests that the gas has been captured in an interaction with the companion , and that the linear structure is a result of this interaction ( perhaps a disk of captured gas seen edge on ) .
in contrast , on the western side of the nucleus in pks2250 - 41 the jets are likely to be directly interacting with the debris of the other companion galaxy
. therefore one can interpret pks2250 - 41 as a complex system in which the dominant galaxy is interacting with two companions , and the jet is ploughing into the debris of the companion on the west side of the nucleus .
the hst imaging results presented in this paper demonstrate the full diversity of jet - cloud interaction phenomenon , with structures ranging from the bow - shock - like structure that cocoons the western radio lobe in pks2250 - 41 , to the linear structures closely aligned along the radio axis in 3c171 , to the bright emission line cloud associated with the knot along the radio jet in 3c277.3 . amongst low / intermediate redshift radio galaxies pks2250 - 41 , 3c277.3 and 3c171 are unusual in the strengths of their jet - cloud interactions .
in addition , pks2250 - 41 and 3c171 have close companions aligned along their radio jets while 3c277.3 shows evidence of a recent merger in which the disk of the debris is observed face - on .
it is likely that these objects represent the few rare cases in which the jets are ploughing into the debris left over from the merger events that triggered the activity .
significant differences are seen between the [ oii ] and [ oiii ] structures in all three objects .
all of these differences can be explained in terms of varying contributions of cooling post - shock gas , shock photoionized precursor gas , and agn photoionized gas .
the correlations between the emission line morphology and the [ oiii]/[oii ] line ratios with the radio emission seen in ground based observations are clarified by the hst imaging .
we confirm that the radio lobes and hot - spots are preferentially associated with lower ionization gas .
3c 171 exhibits high surface brightness emission line gas mainly along the radio source axis .
the lowest ionization gas is seen at the eastern hot spot . in 3c 277.3
there is bright high ionization gas ( and continuum ) offset just to the east of the radio knot k1 .
our observations are consistent with the suggestion of @xcite that this emission is produced by precursor gas ionized by the shock being driven into the cloud by the deflected radio jet . in pks 2250 - 41
we resolve the emission line arc which wraps around the outer rim of the western lobe .
the lower ionization [ oii ] emission is nested just interior to the higher ionization [ oiii ] emission , suggesting that we have resolved the cooling region behind the bow shock .
we also detect possible continuum emission from the secondary hot - spot .
thus , our observations support the hypothesis that in these sources , the interaction between the expanding radio source and the ambient gas strongly influences the morphology , kinematics , and ionization of the gas .
support for program 6657 ( pi c. tadhunter ) was provided by nasa through a grant from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas 5 - 26555 .
the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc .
merlin is operated as a national facility by the university of manchester on behalf of the uk particle physics and astronomy research council .
this research made use of ( 1 ) the nasa / ipac extragalactic database(ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration ; and ( 2 ) nasa s astrophysics data system abstract service .
we thank shireen gonzaga for help with the wfpc2 observations and peter thomasson and tom muxlow for assistance with the merlin2 observations for 3c277.3 .
karen wills acknowledges financial support from pparc and the royal society .
we thank the anonymous referee for very detailed and helpful comments .
baum , s. a. , heckman , t. m. , bridle , a. , van breugel , w. j. m. , miley , g. k. , 1988 , apjs , 68 , 643 baum , s. a. , & mccarthy , p. j. , 2000 , , 119 , 2634 biretta , j. , baggett , s. , noll , k. , 1996 , photometric calibration of wfpc2 linear ramp filter data in synphot , wfpc2 instrument science report 96 - 06 .
biretta , lubin . , _
et al . _
2002 , wfpc2 instrument handbook , version 7.0 ( baltimore : stsci ) blundell , k. m. 1996 , mnras , 283 , 538 chambers , k. c. , miley , g. k. , van breugel , w. , 1987 , nature , 329 , 604 chambers , k. c. , & miley , g. k. , 1990 , in asp conf .
10 , the evolution of the universe of galaxies , ed .
r. g. kron ( san francisco : asp ) , 373 clark , n. e. , tadhunter , c. n. , morganti r. , killeen , n. e. b. , fosbury , r. a. e. , hook , r. n. , seibert , j. , shaw , m. a. , 1997 , mnras , 286 , 558 clark , n. e. , axon , d. j. , tadhunter , c. n. , robinson , a. , obrien , p. , 1998 ,
apj , 494 , 546 de young , d. s. , 2002 , the physics of extragalactic radio sources ( university of chicago press , chicago il ) di serego alighieri , s. , courvoisier , t. j .-
l . , fosbury , r. a. e. , tadhunter , c. n. , binette , l. , 1988 , nature , 334 , 591 fruchter , a. s. , hook , r. n. , 2002 , pasp , 114 , 144 hardcastle , m. j. , 2003 , mnras , 339 , 360 heckman , t. m. , miley , g. k. , balick , b. , van breugel , w. j. m. , & butcher , h. r. , 1982 , , 262 , 529 heckman , t . m. , van breugel , w. j. m. , miley , g. k. , 1984 , apj , 286 , 509 hunstead , r.,w . , 1971 , mnras , 152 , 277 klein , r. i. , mckee , c. f. , colella , p. , 1994 ,
apj , 420 , 213 koekemoer , a. , _ et al . _ `` hst dither handbook '' , version 2.0 ( baltimore : stsci ) mccarthy , p. j. , van breugel , w. j. m. , spinrad , h. , djorgovski , s. , 1987 , apj , 321 , l29 meisenheimer , k. & hippelein , h. , 1992 , , 264 , 455 miley , g. k. , heckman , t. m. , butcher , h. r. , van breugel , w. j. m. , 1981 , apj , 247 , l5 morganti , r. , killeen , n. e. b. , tadhunter , c. n. , 1993 , mnras , 263 , 1023 morganti , r. , oosterloo , t. a. , tinti , s. , tadhunter , c. n. , wills , k. a. , van moorsel , g. , 2002 , a&a , 387 , 830 odea , c. p. , de vries , w. h. , koekemoer , a. m. , baum , s. a. , morganti , r. , fanti , r. , capetti , a. , tadhunter , c. n. , barthel , p. d. , axon , d. j. , gelderman , r. , 2002 , aj , 123 , 2333 osterbrock , d. e. , 1989 , astrophysics of gaseous nebulae and active galactic nuclei , university science books , mill valley ca raymond , j. c. , 1979 , apjs , 39 , 1 shaw , m. , tadhunter , c. n. , dickson , r. , morganti , r. , 1995 , , 275 , 703 shull , j. m. , mckee , c. f. , 1979 , apj , 227 , 131 solorzano - iarrea , c. , tadhunter , c. n. , 2003 , mnras , 348 , 705 spinrad , h. , marr , j. , aguilar , l. , djorgovski , s. , 1985 , pasp , 97 , 932 tadhunter , c. n. , fosbury , r. a. e. , di serego alighieri , s. , bland , j. , danziger , i. j. , goss , w. m. , mcadam , w. b. , snijders , m. a. j. , 1988 , , 235 , 403 tadhunter , c. n. , shaw , m. , clark , n. e. , morganti , r. , 1994 , a&a , 288l , 21 tadhunter , c. n. , villar - martn , m. , morganti , r. , bland - hawthorn , j. , axon , d. , 2000 , mnras , 314 , 849 tingay , s. j. , 1997 , a&a 327 , 550 van breugel , w. j. m. , fillipenko , a. v. , heckman , t. m. , miley , g. k. , 1985 , apj , 293 , 83 van breugel . w. , miley , g. , heckman , t. , butcher , h. , bridle , a. , 1985 , apj , 290 , 496 veilleux , s. , osterbrock , d. e. , 1987 , apjs , 63 , 295 villar - martn , m. , tadhunter , c. n. , morganti , r. , axon , d. , koekemoer , a. , 1999 , mnras , 307 , 24 crrrrrr 3c171 & 06 51 14.8 & @xmath2454 09 00 & 0.2384 & 18.89 & 0.40 & 0.87 + 3c277.3 & 12 54 11.7 & @xmath2427 37 33 & 0.0857 & 15.94 & 0.012 & 0.99 + pks2250 - 41 & 22 53 03.1 & @xmath2540 57 46 & 0.3100 & 20.0 & 0.14 & 0.95 + 3c171 & 5.18@xmath1610@xmath26 & 1.39@xmath1610@xmath27 & 33 & 111.5 & 3.38 & -0.94 + 3c277.3 & 4.56@xmath1610@xmath28 & 1.14@xmath1610@xmath29 & 40 & 58.8 & 1.47 & -0.70 + pks2250 - 41 & 2.75@xmath1610@xmath26 & 2.06@xmath1610@xmath27 & 77 & 21.4 & 4.05 & -0.99 + crrrrrrr & [ oii ] & 3 & fr418n & 02/10/1999 & 2000 & 4615 & 60 + 3c171 & [ oiii ] & 2 & fr680p15 & 02/10/1999 & 3000 & 6195 & 81 + & continuum & 2 & f547 m & 02/10/1999 & 1000 & 5484 & 484 + & [ oii ] & 3 & fr418n18 & 08/01/1997 & 2000 & 4044 & 53 + 3c277.3 & [ oiii ] & 4 & fr533n & 08/01/1997 & 1000 & 5436 & 71 + & continuum & 2 & f467 m & 08/01/1997 & 1000 & 4670 & 178 + & [ oii ] & 3 & fr533n & 09/01/1999 & 2000 & 4862 & 63 + pks2250 - 41 & [ oiii ] & 3 & fr680n & 09/01/1999 & 3000 & 6550 & 85 + & continuum & 2 & f547 m & 09/01/1999 & 1000 & 5484 & 484 + crrrrrrr 3c171 & vla & 08/06/1995 & 3550s & 8 ghz & 50 mhz & 0.044 & @xmath30 pa= @xmath31 + 3c277.3 & vla & 06/10/1982 & 680s & 5 ghz & 50 mhz & + 3c277.3 & merlin2 & 03/04/2000 & 36 hrs & 5 ghz & 15 mhz & 0.043 & @xmath32 + pks2250 - 41 & vla & 09/17/1995 & 5210s & 15 ghz & 50 mhz & 0.090 & @xmath33 pa= @xmath34 + crrrrrrrr 3c171 & nucleus & [email protected] + & eastern lobe & [email protected] + & eastern hotspot & @xmath36 + & western lobe & [email protected] + & western hotspot & @xmath37 + 3c277.3 & nucleus & [email protected] + & knot k@xmath38 & [email protected] + & @xmath39 e of knot k@xmath40 & [email protected] + pks2250 - 41 & nucleus & [email protected] + & filament & [email protected] + & western hotspot & [email protected] + | we present the results of hst / wfpc2 medium and narrow band imaging and vla and merlin2 radio imaging of three powerful radio galaxies : 3c 171 , 3c 277.3 , and pks 2250 - 41 .
we obtained images of the rest frame [ oiii]@xmath05007 and [ oii]@xmath03727 line emission using the linear ramp filters on wfpc2 . the correlations between the emission line morphology and the [ oiii]/[oii ] line ratios with the radio emission seen in ground based observations are clarified by the hst imaging .
we confirm that the radio lobes and hot - spots are preferentially associated with lower ionization gas .
3c 171 exhibits high surface brightness emission line gas mainly along the radio source axis .
the lowest ionization gas is seen at the eastern hot spot . in 3c 277.3
there is bright high ionization gas ( and continuum ) offset just to the east of the radio knot k1 .
our observations are consistent with previous work suggesting that this emission is produced by precursor gas ionized by the shock being driven into the cloud by the deflected radio jet . in pks 2250 - 41 we resolve the emission line arc which wraps around the outer rim of the western lobe . the lower ionization [ oii ] emission is nested just interior to the higher ionization [ oiii ] emission suggesting that we have resolved the cooling region behind the bow shock .
we also detect possible continuum emission from the secondary hot - spot .
thus , our observations support the hypothesis that in these sources , the interaction between the expanding radio source and the ambient gas strongly influences the morphology , kinematics , and ionization of the gas . |
experimental results from heavy - ion collisions at the large hadron collider ( lhc ) and the relativistic heavy - ion collider ( rhic ) give strong indications for the occurrence of a quark - gluon plasma ( qgp ) .
the successful description of the qgp , even at early times , by relativistic viscous hydrodynamics suggests a rather fast thermalization within a few fm @xcite . starting from the highly anisotropic scenario of two colliding ultrarelativistic nuclei , it is still an open question which processes are responsible for this fast thermalization .
one candidate are color instabilities @xcite such as the chromo - weibel instability @xcite , which is the analogue of the weibel instability in electromagnetic plasmas @xcite . in the limit of weak coupling , which is assumed for the present energy scales at lhc and rhic ,
the two nuclei can be described in the color glass condensate ( cgc ) framework @xcite .
it provides an expansion of inclusive quantities , such as the expectation value of the energy - momentum tensor @xmath2 , in powers of the strong coupling @xmath3 , where the leading order is obtained by solving the yang - mills equations .
we want to approach this problem from first principles by applying real - time lattice qcd techniques , approximating qcd by yang - mills theory in the classical limit .
the classical approximation is justified by the high occupation of gluons in the non - equilibrium medium produced after the heavy - ion collision .
in contrast to other works using bjorken ( expanding ) coordinates @xcite , our exploratory studies focus on finite , static box simulations .
the classical yang - mills equations for the gauge fields and their conjugate momenta can be solved using the framework of lattice gauge theory @xcite , where the continuum gauge fields @xmath4 are represented in terms of lattice gauge fields @xmath5 .
we follow the mclerran - venugopalan model @xcite to construct the initial conditions for the lattice gauge fields @xmath6 and the chromo - electric fields @xmath7 within the cgc framework .
we set the longitudinal component of the collective gauge field @xmath8 and construct the transverse components by merging the color fields from the two colliding nuclei , denoted by @xmath9 and @xmath10 , by solving @xcite @xmath11 \right\ } = 0\end{aligned}\ ] ] for given @xmath9 and @xmath10 , with @xmath12 the generators of @xmath1 . to get the color fields @xmath9 and @xmath10 , one solves the poisson equation @xmath13 for nuclear color sources with the following gaussian distribution : @xmath14 the color fields
are then constructed from the obtained `` color potentials '' @xmath15 via @xmath16 , with @xmath17 and @xmath18 .
the transverse chromo - electric fields are set to zero and the longitudinal part is constructed out of the gauge fields @xmath6 via @xmath19\big[u_i^{(2)\dagger}(x ) - u_i^{(1)\dagger}(x)\big ] - \text{h.c.}\\ & \hspace{0.45 cm } + \big[u_i^\dagger(x-\hat{\imath } ) - 1\big]\big[u_i^{(2)\dagger}(x-\hat{\imath } ) - u_i^{(1)\dagger}(x-\hat{\imath})\big ] - \text{h.c . }
\bigg\}\bigg)\;. \end{split}\end{aligned}\ ] ] in order to reach isotropization , fluctuations need to be added to the initial fields @xcite .
this is done by a small random variation of the idealized @xmath20-function - like distribution of the initial chromo - electric fields . to this end , we add a fluctuation term @xmath21 .
we will discuss the effect of the seed @xmath22 in .
to study the thermalization time , we observe the ratio of the longitudinal to the transverse pressure @xmath23 @xcite .
the pressure components are related to the spatial diagonal elements of the energy - momentum tensor @xmath24 @xcite and the chromo - electric and chromo - magnetic longitudinal and transverse energy densities as : @xmath25 = \epsilon_{e_l}(t ) + \epsilon_{b_l}(t)\end{aligned}\ ] ] the lattice expressions for the components of the chromo - electric and chromo - magnetic energy densities @xmath26 , with @xmath27 , are @xmath28\\ h^e_i(t , x ) & = \frac{1}{g^2 } \text{re}\text{tr } \big [ e_i(x)e_i(x ) \big ] \label{eq : hei}\end{aligned}\ ] ] from which we get the energy densities @xmath29 by averaging over the transverse lattice plane @xmath30 or the longitudinal axis @xmath31 , respectively .
our results are based on calculations on a @xmath32-lattice averaged over 30 independent measurements .
we set the coupling constant @xmath33 and fix @xmath34 in accordance with the literature and conforming to the expectations of rhic physics @xcite . the value for the seed @xmath22 , which controls the fluctuations , is purely artificial .
the higher it is , the faster the system thermalizes , but there is an upper bound : since @xmath22 only enters the electric part of the initial energy density , we have to monitor its influence on the initial total energy density , which should not significantly change , otherwise the classical approximation is no longer tenable . in our setup
we chose @xmath35 , which leads to a change of the initial value of the total energy density , which stays constant during the simulation , by less than @xmath36 ( cf .
( left ) ) . , electric energy density @xmath37 and total energy density @xmath38 .
( right ) pressure ratio @xmath23 and hydrodynamization line at @xmath39 . ] , electric energy density @xmath37 and total energy density @xmath38 .
( right ) pressure ratio @xmath23 and hydrodynamization line at @xmath39 . ] evidence for thermalization is provided by the pressure ratio @xmath23 approaching @xmath40 .
we are especially interested in the so - called hydrodynamization time , which is the time where hydrodynamic models should start to be applicable .
( right ) shows the value of the pressure ratio , which slowly approaches the hydrodynamization level , which is commonly set to @xmath41 .
shows the energy density @xmath42 ( cf . ) in the @xmath43-plane , where we averaged over the @xmath44-direction .
since we are only interested in the qualitative structure , we performed only one measurement here .
this explains the vertical lines , which are a result of the initial fluctuations . at @xmath45 ,
two filaments start emerging , which are clearly visible at @xmath46 .
later , at @xmath47 the filamentous structure is almost completely dissolved , which suggests an imminent thermalization .
this characteristic filamentation is a strong indication for the presence of a chromo - weibel instability . averaged over the @xmath44-direction is plotted in the @xmath43-plane for different times .
one can easily see the arising filamentation . ]
averaged over the @xmath44-direction is plotted in the @xmath43-plane for different times .
one can easily see the arising filamentation . ]
averaged over the @xmath44-direction is plotted in the @xmath43-plane for different times .
one can easily see the arising filamentation . ]
we have provided evidence of the thermalization ( hydrodynamization ) of a color glass condensate such as the one created at relativistic heavy - ion collisions within classical @xmath1 yang - mills theory . to this end , we studied observables such as the energy density and its components and the longitudinal to transverse pressure ratio . by looking at components in a transverse - longitudinal plane , we were able to visualize the occurrence of chromo - weibel instabilities represented by a clear filamentation of the energy density .
the next steps will be an analysis of finite volume and discretization effects as well as studies of an expanding system .
the authors acknowledge fruitful discussions with b. schenke and r. venugopalan .
m. a. is supported by the marie skodowska - curie individual fellowship 658574 fastth .
o. p and b. w. are supported by the helmholtz international center for fair within the loewe program of the state of hesse .
s. z. is supported by the humboldt foundation .
calculations were done on the loewe - csc at geothe - university , the authors thank its administrative staff and the hpc - hessen for programming support .
99 u. w. heinz and p. f. kolb , nucl .
a * 702 * , 269 ( 2002 ) doi:10.1016/s0375 - 9474(02)00714 - 5 [ hep - ph/0111075 ] .
p. romatschke and u. romatschke , phys .
lett . * 99 * , 172301 ( 2007 ) doi:10.1103/physrevlett.99.172301 [ arxiv:0706.1522 [ nucl - th ] ] .
s. mrowczynski , b. schenke and m. strickland , arxiv:1603.08946 [ hep - ph ] .
m. attems , a. rebhan and m. strickland , phys .
d * 87 * , no . 2 , 025010 ( 2013 ) doi:10.1103/physrevd.87.025010 [ arxiv:1207.5795 [ hep - ph ] ] .
b. arnold , j. lenaghan , g. d. moore and l. g. yaffe , phys .
lett . * 94 * , 072302 ( 2005 ) doi:10.1103/physrevlett.94.072302 [ nucl - th/0409068 ] .
d. bodeker , jhep * 0510 * , 092 ( 2005 ) doi:10.1088/1126 - 6708/2005/10/092 [ hep - ph/0508223 ] .
a. h. mueller , a. i. shoshi and s. m. h. wong , phys .
b * 632 * , 257 ( 2006 ) doi:10.1016/j.physletb.2005.10.047 [ hep - ph/0505164 ] .
s. mrowczynski , phys .
b * 393 * , 26 ( 1997 ) doi:10.1016/s0370 - 2693(96)01621 - 8 [ hep - ph/9606442 ] .
e. s. weibel , phys .
* 2 * , 83 ( 1959 ) .
doi:10.1103/physrevlett.2.83 e. iancu and r. venugopalan , in * hwa , r.c .
et al . : quark gluon plasma * 249 - 3363 [ hep - ph/0303204 ] .
f. gelis , e. iancu , j. jalilian - marian and r. venugopalan , ann .
nucl . part .
sci . * 60 * , 463 ( 2010 ) doi:10.1146/annurev.nucl.010909.083629 [ arxiv:1002.0333 [ hep - ph ] ] .
f. gelis , int .
j. mod .
a * 28 * , 1330001 ( 2013 ) doi:10.1142/s0217751x13300019 [ arxiv:1211.3327 [ hep - ph ] ] . k. fukushima and f. gelis , nucl .
a * 874 * , 108 ( 2012 ) doi:10.1016/j.nuclphysa.2011.11.003 [ arxiv:1106.1396 [ hep - ph ] ] .
j. ambjorn , t. askgaard , h. porter and m. e. shaposhnikov , phys .
b * 244 * , 479 ( 1990 ) .
doi:10.1016/0370 - 2693(90)90350-f l. d. mclerran and r. venugopalan , phys . rev .
d * 49 * , 3352 ( 1994 ) doi:10.1103/physrevd.49.3352 [ hep - ph/9311205 ] . l. d. mclerran and r. venugopalan , phys .
d * 49 * , 2233 ( 1994 ) doi:10.1103/physrevd.49.2233 [ hep - ph/9309289 ] .
l. d. mclerran and r. venugopalan , phys . rev .
d * 50 * , 2225 ( 1994 ) doi:10.1103/physrevd.50.2225 [ hep - ph/9402335 ] .
a. krasnitz and r. venugopalan , nucl .
b * 557 * , 237 ( 1999 ) doi:10.1016/s0550 - 3213(99)00366 - 1 [ hep - ph/9809433 ] .
p. romatschke and r. venugopalan , phys .
lett . * 96 * , 062302 ( 2006 ) doi:10.1103/physrevlett.96.062302 [ hep - ph/0510121 ] .
p. romatschke and r. venugopalan , phys .
d * 74 * , 045011 ( 2006 ) doi:10.1103/physrevd.74.045011 [ hep - ph/0605045 ] . t. epelbaum and f. gelis , phys .
lett . * 111 * , 232301 ( 2013 ) doi:10.1103/physrevlett.111.232301 [ arxiv:1307.2214 [ hep - ph ] ] . | to achieve an understanding of the thermalization of a quark - gluon plasma , starting from qcd without using model assumptions , is a formidable task .
we study the early stage dynamics of a relativistic heavy ion collision in the framework of real - time simulations of classical yang - mills theory in a static box with the color glass condensate as initial condition .
our study generalizes a previous one by fukushima and gelis from @xmath0 to the realistic case of @xmath1 .
we calculate the chromo - electric and chromo - magnetic energy densities as well as the ratio of longitudinal and transverse pressure as a function of time as probes for thermalization .
our preliminary results on coarse lattices show the occurrence of weibel instabilities prior to thermalization . |
layered nitride @xmath0-@xmath3ncl@xcite ( @xmath4hf , zr ) doped with carriers is one of the most interesting group of superconductors .
the mother compound @xmath0-@xmath3ncl is composed of alternate stacking of honeycomb ( hc ) @xmath3n bilayer and cl bilayer.@xcite this is a band insulator and becomes a superconductor upon doping electrons by na or li intercalation .
they exhibit relatively high @xmath2 up to @xmath5k for @xmath4hf and @xmath6k for @xmath4zr .
the bilayer hc lattice structure consisting of @xmath3 and n is considered to be playing the main role in the occurrence of superconductivity , and the two dimensional nature of the superconductivity has been revealed by nuclear magnetic resonance ( nmr)@xcite and muon spin relaxation ( @xmath7sr ) studies.@xcite despite the relatively high @xmath2 , experimental as well as theoretical studies indicate extremely low density of states ( dos ) at the fermi level.@xcite in fact , they have the highest @xmath2 among materials with the specific heat coefficient @xmath8 as small as @xmath9 1mj / molk@xmath10 .
the electron phonon coupling is also estimated to be weak,@xcite and the isotope effect is found to be small.@xcite these experiments suggest that some kind of unconventional pairing mechanism may be at work , but on the other hand , nmr knight shift measurement suggests spin singlet pairing,@xcite and the tunneling spectroscopy@xcite and specific heat@xcite experiments find a fully open , seemingly @xmath11-wave like gap .
however , regarding this fully open gap , recent experiments show that the anisotropy of the gap increases with doping .
namely , when the doping concentration is small , the specific heat coefficient increases linearly as a function of the magnetic field , suggesting an isotropic gap,@xcite while a steep increase of the coefficient at low magnetic field is observed for higher doping.@xcite also , @xmath12 determined from the specific heat measurement ranges from @xmath13 in the lightly doped regime to less than @xmath14 in the heavily doped regime,@xcite which may be an indication that the maximum and minimum values of the gap have a substantial difference when heavily doped , and the specific heat is mainly governed by the minimum value .
moreover , a recent superfluid density measurement by @xmath7sr shows that the gap is nearly isotropic when the doping concentration is small , while the anisotropy increases for large doping.@xcite very recently , absence of coherence peak in the spin lattice relaxation rate has been found in an nmr experiment,@xcite again suggesting unconventional pairing , most probably with some kind of sign change in the superconducting gap .
furthermore , for li@xmath15hfncl , an intercalation of organic molecules tetrahydrofuran ( thf ) between the layers is found to enhance @xmath2 .
@xcite given these experimental circumstances , here we study two types of models for the doped @xmath0-mncl , a two band model on a hc lattice , and a single band model where the nitrogen site is effectively integrated out . while the latter model does not seem to suffice for explaining the relatively high @xmath2 in this series of material , the former one provides a possibility of spin fluctuation mediated @xmath16-wave superconductivity in the doped @xmath0-mncl.@xcite spin fluctuation mediated pairing in a doped _ band insulator _ may sound surprising and odd at first sight since one can not expect strong spin fluctuations .
nevertheless , we show that the relatively high @xmath2 despite the low dos and the temperature independent spin susceptibility is a characteristic feature of the spin fluctuation mediated superconductivity of a doped band insulator on the hc lattice .
the most probable pairing state below @xmath2 is the @xmath1 state , whose gap anisotropy on the fermi surface is found to be strongly enhanced upon increasing the carrier concentration .
we also study the effect of the three dimensionality , where the inter layer hopping suppresses superconductivity .
we first obtain the effective model for @xmath0-mncl .
although the material consists of bilayer hc lattice , here we consider models on a single layer lattice , and try to reproduce the bands that lie close to the fermi level in the first principles band calculations .
the simplest one is a single band model on a triangular lattice , in which the nitrogen site degrees of freedom is effectively integrated out , so that there is only one site per unit cell .
this is a model that corresponds to the single band hubbard model for the cuprates , where the oxygen @xmath17 , @xmath18 degrees of freedom is integrated out from the original three band @xmath16-@xmath19 model . by considering hopping integrals up to fourth nearest neighbors ( fig.[fig1 ] , upper panel )
, we can reproduce the single band that crosses the fermi level in the first principles band calculations.@xcite the band dispersion of the single band model with @xmath20 ( ev ) , @xmath21 , @xmath22 , @xmath23 is shown by the green dash - dotted line in fig.[fig1 ] .
( color online ) upper : the single band model on a triangular lattice .
middle : the two band model on a honeycomb lattice .
lower : the first principles band calculation taken from ref.@xcite and the band dispersion of the single band model ( dash - dotted green ) and the two band model(dashed red).,width=226 ] to go a step further , we also consider a model on a single layer hc lattice consisting of alternating m and
n orbitals with a level offset , which roughly reproduces the conduction and valence bands closest to the fermi level , as shown by the dashed lines in the middle panel of fig.[fig1 ] . here
we take the nearest neighbor ( m - n ) hopping @xmath24ev , the level offset @xmath25 , and the next nearest neighbor ( m - m ) hopping @xmath26 . the on - site interaction @xmath27 is considered in the single band model , and the on - site interaction @xmath27 is introduced on both m and n orbitals in the two orbital model@xcite .
the band filling @xmath28 is defined as the number of electrons / number of sites . for the single band model , @xmath29 , where @xmath30 is the li or na content ,
since there are two m sites per unit cell , which results in bonding and antibonding bands , and the electrons are doped only in the bonding band .
for the two band model , @xmath31 corresponds to the non - doped case , and @xmath30 and @xmath28 is related by @xmath32 .
we use fluctuation exchange ( flex ) method @xcite , which is kind of a self - consistent random phase approximation , to obtain the green s function .
then we solve the linearized eliashberg equation , @xmath33 here , @xmath34 is the green s function matrix ( with @xmath35etc ... labeling sites in a unit cell ) obtained by flex .
@xmath36 is the spin singlet pairing interaction matrix given by @xmath37 with the spin and charge susceptibility matrices @xmath38^{-1}$ ] , where @xmath39 is the irreducible susceptibility matrix @xmath40 ( @xmath41 is the number of @xmath42-point meshes ) . in the following
, the maximum eigenvalue of the spin susceptibility matrix will be referred to as the spin susceptibility .
we take up to @xmath43 @xmath42-point meshes and up to 16384 matsubara frequencies .
the eigenvalue of the eliashberg equation @xmath44 increases upon lowering the temperature and reaches unity at @xmath45 .
we first present calculation results for the single band model . in fig.[fig2 ] , the eigenvalue of the eliashberg equation is shown as functions of temperature for spin singlet and triplet pairings for @xmath46 .
it can be seen that the triplet pairing strongly dominates over singlet pairing .
the gap of the spin triplet pairing has an @xmath47-wave form , where the nodes of the gap do not intersect the fermi surface ( the ridge of the green s function squared ) as shown in fig.[fig3 ] . on the other hand
, the singlet pairing has a @xmath16-wave gap ( not shown ) .
the possibility of this type of spin triplet @xmath47-wave pairing was in fact proposed on a canonical triangular lattice in ref.@xcite .
the triplet pairing dominates over singlet pairing because ferromagnetic spin fluctuations arise due to the dilute band filling as shown in fig.[fig3](b ) , and also because the nodes of the @xmath47-wave gap do not intersect the fermi surface .
( color online ) the contour plots of the flex result at the lowest matsubara frequency for the single band model in the hexagonal brillouin zone with @xmath46 and @xmath48 ( a ) the green s function of the upper band squared , ( b ) the spin susceptibility , ( c ) the superconducting gap.,width=302 ] ( color online ) the contour plots of the flex result at the lowest matsubara frequency for the single band model in the hexagonal brillouin zone with @xmath46 and @xmath48 ( a ) the green s function of the upper band squared , ( b ) the spin susceptibility , ( c ) the superconducting gap.,width=302 ] nonetheless , as can be seen from fig.[fig2 ] , the temperature at which the @xmath47-wave eigenvalue reaches unity , if any , seems to be very low .
we have also tried a model that considers the nearest neighbor off - site interaction @xmath36 in addition to @xmath49 . in this case ,
strong charge fluctuations arise , which makes the triplet vs. singlet competition more subtle , but in any case , the eigenvalue of the eliashberg equation remained small even when the magnitude of @xmath36 is close to the point where the charge susceptibility diverges . considering the experimental fact that the @xmath2 is relatively high in the doped @xmath0-mncl , and also that the pairing occurs in the spin singlet channel@xcite , we believe that the single band model does not suffice as an effective model for @xmath0-mncl .
we now move on to the two band model . in this model
, we find that spin singlet pairing strongly dominates over triplet pairing . in fig.[fig4 ] , we plot @xmath2 of the singlet pairing superconductivity as a function of the band filling .
it can be seen that @xmath50k is obtained , which can be considered as relatively high noting that @xmath51k ( assuming @xmath52ev appropriate for the cuprates ) is obtained by the same method for the hubbard model on the square lattice.@xcite the difference between the single and two band models is dramatic in that not only the @xmath2 but even the leading pairing symmetry is different .
this is in contrast to the case of the cuprates , where the single band hubbard and three band @xmath16-@xmath19 models is expected to give roughly the same conclusions .
@xmath2 plotted as a function of the band filling for the two band model on the hc lattice.,width=264 ] ( color online ) ( a ) the maximum value of the spin susceptibility ( blue , left axis ) and the eigenvalue of the eliashberg equation ( red , right ) as functions of temperature for the square lattice with @xmath53 ( dashed ) or @xmath54 ( dash dotted ) and for the honeycomb lattice ( solid ) for @xmath55 .
@xmath56 in all cases .
hc and sq stand for the honeycomb ( with @xmath57 ) and the square lattices , respectively . the temperature
at which @xmath58 is the @xmath2 .
( b ) @xmath2 as functions of the level offset @xmath59 obtained by flex+eliashberg equation for the square lattice with @xmath56 , @xmath53 ( dashed ) or for the hc lattice with @xmath56 and @xmath55 ( solid).,width=302 ] ( color online ) the contour plots of the flex result at the lowest matsubara frequency for the honeycomb lattice ( @xmath60 ) in the hexagonal brillouin zone with @xmath55 and @xmath61 ( a ) the green s function of the upper band squared , ( b ) the spin susceptibility , ( c ) the superconducting gap of the upper band , ( d ) the fermi surface ( the two circles ) and the sign of the gap function are schematically shown in the extended zone scheme .
the dashed arrows represent the wave vectors at which the spin fluctuations develop.,width=302 ] the high @xmath2 obtained in this model can be understood as a combination of several features of this model .
let us start with a canonical hc lattice , where @xmath57 and @xmath62 .
we show in fig.[fig5](a ) the flex result of the maximum value of the spin susceptibility as a function of temperature for the band filling of @xmath55 .
the spin susceptibility is nearly independent of @xmath63,@xcite which is in sharp contrast with the hubbard model on the square lattice .
for example , for the square lattice with @xmath53 , we have a strong enhancement of the spin susceptibility upon lowering the temperature .
with further hole doping to @xmath54 , the spin susceptibility is suppressed , but even there , the spin susceptibility moderately increases upon lowering the temperature .
also in fig.[fig5](a ) , we show the eigenvalue @xmath44 of the linearized eliashberg equation as functions of temperature for the hc and square lattices .
@xmath2 is the temperature where @xmath44 reaches unity .
the dos at @xmath64 is nearly the same for the square lattice with @xmath54 and the hc lattice with @xmath55 , and also the spin susceptibility has similar values at low temperature , but still , the hc lattice has higher @xmath2.@xcite
so what is the origin of this high @xmath2 ?
fig.[fig6](a ) shows the contour plot of the green s function squared , whose ridge corresponds to the fermi surface , which consists of disconnected 2d pockets centered around k and k points .
the spin susceptibility in fig.[fig6](b ) is maximized at wave vectors that bridge the opposite sides of each pieces of the fermi surface . as can be seen in fig.[fig6](c ) ( and more clearly in ( d ) ) the gap has a @xmath16-wave form , where the gap changes sign across the wave vector at which the spin susceptibility is maximized .
in contrast to the case of the square lattice , one of the nodes of the @xmath16-wave gap does not intersect the fermi surface because of its disconnectivity , i.e. , the gap is like `` @xmath19-wave '' if we focus only on one of the fermi surfaces .
since smaller number of gap nodes on the fermi surface is favorable for superconductivity,@xcite this can be a reason for high @xmath2 despite the low dos and the weak spin fluctuations .
we now introduce the m - n level offset @xmath59 as in the model for @xmath0-mncl . as shown in fig.[fig5](b ) , we find that superconductivity is relatively robust against the introduction of @xmath59 .
this again is in sharp contrast with the square lattice , where finite @xmath59 strongly suppresses superconductivity .
the weak effect of @xmath59 may be because the dos of the hc lattice is small near the band center , so that the introduction of @xmath59 , which opens up a gap at the center of the band , is less effective than in the case of the square lattice , where the dos diverges at the band center for @xmath62 .
let us now comment on the origin of the difference between the single and two band models , namely , low @xmath2 triplet pairing in the former model and the high @xmath2 singlet pairing in the latter .
the difference mainly comes from the difference in the band filling , i.e. , for the single band model , the band filling is dilute , favoring ferromagnetic spin fluctuations , while in the two band model , it is near half filling in favor of antiferromagnetic spin fluctuations .
of course , if the level offset @xmath59 is large enough , the two band case should tend to the single band case , as have been shown in ref.@xcite , but in the present case , the magnitude of @xmath59 is small enough to retain the nature of the @xmath62 situation . apart from the @xmath2 , another important issue is the form of the superconducting gap . by symmetry , there are two degenerate @xmath16-wave gaps , and the most probable form of the gap below @xmath2 is the form @xmath1 , where the two @xmath16-wave gaps mix with a phase shift of @xmath65.@xcite since the two @xmath16-wave gaps have nodal lines at different positions , this kind of mixture leads to a gap that has a finite absolute value on the entire fermi surface .
the @xmath1 form of the gap can be constructed from the @xmath16-wave gap obtained by solving the linearized eliashberg equation . in the upper panels of fig.[fig7 ] , we show the contour plot of the @xmath1 gap for the case of @xmath66 and @xmath67 for the model for @xmath0-mncl .
( color online ) upper two panels : the contour plots of the @xmath1 gap for @xmath66 ( top ) and @xmath67 ( middle ) . the model for @xmath0-mncl
is used . the yellow ( light ) solid line ( closed around the brillouin zone edge )
represents the fermi surface .
lower panel : the normalized @xmath1 gap along the fermi surface .
@xmath68 is defined in the top panel .
, width=245 ] to clearly see the anisotropy on the fermi surface , we plot the magnitude of the gap ( normalized by the maximum value ) in the lower panel of fig.[fig7 ] .
it can be seen that the @xmath1 gap is nearly isotropic for low doping concentration , while it becomes more and more anisotropic for higher doping .
the @xmath1 pairing scenario is consistent with the experimental findings that the superconductivity occurs in the spin singlet@xcite , @xmath11-wave like channel,@xcite while the coherence peak in the spin lattice relaxation rate is absent.@xcite it is also consistent with the experimental finding that the gap anisotropy increases upon doping.@xcite since the @xmath1 state breaks time reversal symmetry , it is interesting to experimentally investigate such a possibility . finally , we consider the effect of dimensionality by adopting a three dimensional model where the two band model on the hc lattice is connected by vertical ( m - m and
n - n ) hopping integrals @xmath69 ( inset of fig.[fig8 ] ) . here , we take up to @xmath70 @xmath42-point meshes and up to 8192 matsubara frequencies . in fig.[fig8 ] , we plot @xmath2 as a function of @xmath69 for the band filling of @xmath55 .
as can be seen , the three dimensionality suppresses @xmath2 , which is in agreement at least qualitatively with the experimental finding that the intercalation of thf molecules between the layers enhances @xmath2.@xcite in fact , the suppression of @xmath2 with increasing three dimensionality is a general trend for spin fluctuation mediated pairing as has been studied in ref.@xcite .
namely , since the pairing interaction is large around a certain wave vector in spin fluctuation mediated pairing , the fraction of the volume where the pairing interaction is large in the brillouin zone becomes smaller as the three dimensionality increases .
@xmath2 plotted as a function of @xmath69 for the model shown in the inset .
@xmath24ev , @xmath26 , @xmath25 , @xmath27 , @xmath55.,width=302 ]
in the present study , we have studied two kinds of model for the the doped @xmath0-mncl .
we have found that the single band model does not suffice as a minimal effective model of the material . on the other hand , from the study on the two band model
, we have proposed a possibility of spin fluctuation mediated @xmath1-wave superconductivity in the doped @xmath0-mncl .
we have shown that the relatively high @xmath2 is obtained as a characteristic feature of the spin fluctuation mediated superconductivity in a doped band insulator on the hc lattice .
we have also found that the gap anisotropy on the fermi surface strongly increases upon increasing the doping concentration , and the inter layer hopping suppresses superconductivity .
these results are in qualitative agreement with the experimental findings .
one remaining issue is the curious doping dependence of the superconductivity . for li@xmath15zrncl
, @xmath2 shows an increase upon lowering the carrier concentration until a sudden disappearance of the @xmath2 and a superconductor - insulator transition is observed.@xcite recently , this increase of @xmath2 has been shown to be correlated with the increase of the uniform spin susceptibility , which can be considered as a support for the present spin fluctuation scenario.@xcite in fact , although the spin fluctuation is nearly _ temperature _ independent in our model ( so that the development of the spin fluctuation upon lowering the temperature need not be observed experimentally ) , it is strongly _ doping _ dependent and is enhanced upon lowering the doping rate.@xcite nonetheless , the origin of the _ sudden _ drop of @xmath2 at a certain doping rate can not be explained within the present two band model ( see fig.[fig4 ] ) and remains as an open problem . also , the almost constant behavior of @xmath2 in the heavily doped regime@xcite is puzzling since for such a large doping , the fermi level hits the second band and a large difference in the dos should be detected ( which is not the case experimentally )
. a rigid band picture may not be valid in the heavily doped regime , as suggested in some experiments@xcite and band calculations.@xcite the author acknowledges y. taguchi , y. iwasa , y. kasahara , h. tou , r. kadono , m. hiraishi , s. yamanaka , t. ekino , a. sugimoto , h. fukuyama , t. oguchi , m. onoue , and h. aoki for fruitful discussions .
numerical calculation has been done at the computer center , issp , university of tokyo .
this study has been supported by grants - in aid from mext of japan and jsps .
9 s. yamanaka _
et al . _ ,
nature * 392 * , 580 ( 1992 ) .
s. shamoto _
et al . _ , physica ( amsterdam ) * 402c * , 283 ( 2004 ) . h. tou _
et al . _ ,
b * 63 * , 020508 ( 2000 )
. y. j. uemura _
et al . _ , physica ( amsterdam ) * 289 - 290b * , 389 ( 2000 ) .
et al . _ ,
b * 69 * , 134522 ( 2004 ) .
et al . _ ,
lett . * 86 * , 5775 ( 2001 ) .
y. taguchi , m. hisakabe , and y. iwasa , phys .
lett . * 94 * , 217002 ( 2005 ) .
r. weht _ et al .
_ , europhys . lett . * 48 * , 320 ( 1999 )
. a. kitora , y. taguchi , and y. iwasa , j. phys .
jpn . * 76 * , 023706 ( 2007 ) .
r.heid and k.p.bohnen , phys .
b * 72 * , 134527 ( 2005 ) .
et al . _ ,
b * 67 * , 100509 ( 2003 ) .
y. taguchi _
et al . _ ,
b * 76 * , 064508 ( 2007 )
. h. tou , y. maniwa , and s. yamanaka , phys .
b * 67 * , 100509 ( 2003 ) .
et al . _ , physica c * 388 - 389 * , 573 ( 2003 ) ; physica b * 328 * , 23 ( 2003 ) .
y. kasahara _
et al . _ ,
lett . * 103 * , 077004 ( 2009 ) .
m. hiraishi _
_ , arxiv : 0910.5557 . h. tou , private communications . t. takano , t. kishiume , y. taguchi , and y. iwasa , phys . rev . lett . * 100 * , 247005 ( 2008 ) .
theoretically , a possibility of unconventional pairing mediated by charge fluctuations has been proposed in a. bill _ et al .
b * 66 * 100501(r ) ( 2002 )
. i. hase and y. nishihara , phys . rev .
b * 60 * , 1573 ( 1999 ) .
h. sugimoto and t. oguchi , j. phys .
. jpn . * 73 * , 2771 ( 2004 ) .
generally , the on - site repulsion on the m site and n site can be different , but here we assume them to be equal for simplicity . n.e .
bickers , d.j .
scalapino , and s.r .
white , phys .
lett . * 62 * , 961 ( 1989 ) .
t. dahm and l. tewordt , phys .
. lett . * 74 * , 793 ( 1995 ) .
s. grabowski , m. langer , j. schmalian , k.h .
bennemann , europhys . lett . * 34 * , 219 ( 1996 ) .
k. kuroki and r. arita , phys .
b * 63 * , 174507 ( 2001 ) .
k. kuroki , in proc .
workshop on superconductivity in diamond and related materials , sci .
technol . adv .
* 9 * , 044202 ( 2008 ) .
possibility of high @xmath2 superconductivity on the hc lattice has also been proposed in the context of rvb theory of graphene by pathak _
et al . _ , arxiv : 0809.0244 . k. kuroki and r. arita , phys . rev .
b * 64 * , 024501 ( 2001 ) .
r. arita , k. kuroki , and h. aoki , phys .
b * 60 * , 14585 ( 1999 ) .
p. monthoux and g.g .
lonzarich , phys .
b * 63 * , 054529 ( 2001 ) .
g. baskaran , phys .
* 91 * , 097001 ( 2003 ) .
b. kumar and b.s .
shastry , phys .
b * 68 * , 104508 ( 2003 ) .
m. ogata , j. phys .
soc . jpn . * 72 * , 1839 ( 2003 ) .
y. taguchi , a. kitora , and y. iwasa , phys .
lett . * 97 * ( 2006 ) 107001
. t. yokoya _
et al . _ ,
b * 70 * , 193103 ( 2004 ) .
m. onoue and t. oguchi , private communications | we study two types of models for the superconducting layered nitride @xmath0-mncl(m = hf , zr ) ; a single band model on a triangular lattice , and a two band model on a honeycomb lattice .
we find that the former model does not suffice as an effecitive model , while the latter one can be a good candidate .
we propose from the study on the two band model a possibility of spin fluctuation mediated @xmath1-wave superconductivity in the doped @xmath0-mncl .
we show that the relatively high @xmath2 obtained in the doped band insulator is a characteristic feature of the spin fluctuation mediated superconductivity on a honeycomb lattice .
we also find that the gap anisotropy on the fermi surface strongly increases upon increasing the doping concentration , and the inter layer hopping suppresses superconductivity .
these results are in qualitative agreement with the experimental findings . |
the bright 5.54 s x - ray pulsar is the second example of a transient anomalous x - ray pulsar ( taxp ) and the first one confirmed by measuring a rapid spin - down rate .
all of its observed and derived physical parameters are consistent with classification as an anomalous x - ray pulsar ( axp ) , one that had an impulsive outburst sometime between 2002 november and 2003 january , when it was discovered serendipitously by @xcite using the _ rossi x - ray timing explorer _ ( _ rxte _ ) .
its flux was observed to be declining with an exponential time constant of @xmath6 days from a maximum of @xmath7 ergs @xmath8 s@xmath9 .
the source was then localized precisely using two target of opportunity ( too ) observations with the _ chandra _ x - ray observatory by ( * ? ? ?
* hereafter paper i ) and @xcite . in comparison ,
archival detections by several x - ray satellites indicate a long - lived quiescent baseline flux of @xmath10 ergs @xmath8 s@xmath9 lasting at least 13 years and possibly for 23 years prior to the outburst ( paper i ) .
fading of an ir source within the _ chandra _ error circle , similar to ones associated with other axps , confirmed its identification with @xcite .
the x - ray spectra and pulse profiles from three observations obtained with _ xmm - newton _ during the decline of the outburst were studied by ( * ? ? ?
* hereafter paper ii ) .
the short duty cycle of activity of suggests the existence of a significant population of as - yet unrecognized , although not necessarily undetected , young neutron stars ( nss ) .
in this paper we present a new _ xmm - newton _ observation of .
this data set , combined with previous _ xmm - newton _ observations acquired over the last year , allows us to characterize the spectral evolution of a transient axp .
we show that the decay rates of the two fitted x - ray spectral components are distinct , and can be extrapolated in time to approach the previous quiescent spectrum .
for consistency with papers i and ii , we express results in terms of a maximum distance of @xmath11 kpc to the pulsar . however , there is reason to believe that is significantly closer than this , and we take into account a more realistic estimate of 2.5 kpc in our discussion of proposed physical models for the x - ray emission and outburst mechanism .
a fourth _ xmm - newton _ observation of was obtained on 2004 september 18 .
the previous three observations are described in paper ii .
we use the data collected with the european photon imaging camera ( epic , @xcite ) which consist of three ccd imagers , the epic pn and the two epic moss , each sensitive to x - rays in the @xmath12 kev energy range . in the following we concentrate on data taken with the epic pn detector , which provided a timing resolution of 48 ms in `` large window '' mode , for ease of comparison with the earlier data sets .
the fast readout of this instrument ensures that its spectrum is not affected by photon pileup .
data collected with the two epic mos sensors used the `` small window '' mode and `` timing '' mode , providing a time resolution of 0.3 s and 1.5 ms , respectively . for the new data set , we followed the reduction and analysis procedures used for the previous _ xmm - newton _ observations of , as outlined in paper ii . the new observation is mostly uncontaminated by flare events and the filtered data set resulted in a total of 26.5 ks of good epic pn exposure time ( 24.4 ks live time ) .
we checked for timing anomalies that were evident in some previous epic pn data sets and found none .
photon arrival times were converted to the solar system barycenter using the _ chandra _ derived source coordinates r.a .
@xmath13 , decl .
@xmath14 ( j2000.0 ) given in paper i. the barycentric pulse period of measured at each _ xmm - newton_observing epoch is given in table [ summary ] .
these are derived from photons obtained with both the epic pn and mos cameras , with the exception of the short 2003 october 12 observation for which only epic pn data of sufficient time resolution is available .
the errors are the 95% confidence level determined from the @xmath15 test . as shown in figure [ timing ] , the four period measurements can be fitted to yield a mean spin - down rate of @xmath16 s s@xmath9 over the year - long interval .
this implies a characteristic age @xmath17 yr , surface magnetic field @xmath18 g , and spin - down power @xmath19 erg s@xmath9 , comparable to the earlier values ( * ? ? ?
* and paper ii ) . deviations from a constant @xmath20 are evident , however , since @xcite fitted values in the range @xmath21 s s@xmath9 in the first 9 months of the outburst .
as discussed in paper ii ( 2.2 ) , timing glitches in axps can result in large increases in their period derivatives [ @xmath22 ; e.g. , 1e 2259 + 586 @xcite and 1rxs j170849.0400910 @xcite ] .
accordingly , it is possible that experienced a glitch and an increase in @xmath20 at the time of its outburst , and that @xmath20 is now relaxing to its long - term value .
but since there is no prior ephemeris for in its quiescent state , we do not know if its outburst was triggered by a glitch .
alternatively , the spin - down torque could have been enhanced in the early stages of the outburst by an increase over the dipole value of the magnetic field strength at the speed - of - light cylinder @xcite , or by a particle wind and alfvn waves @xcite , effects that are expected to decline after the first few months .
_ xmm - newton _ observations of have shown that its spectrum is equally well fitted by a power - law plus blackbody model , as commonly quoted for axps , or a two - temperature blackbody model . in paper
ii we argued that the two - temperature model is more physically motivated , while the power - law plus blackbody model suffers from physical inconsistencies . as we shall show , the new data bolster these arguments , so we concentrate mainly on the double blackbody model in this work .
table [ summary ] presents a summary of spectral results from all four _ xmm - newton _ observations of .
lcccc @xmath23 ( @xmath24 @xmath8 ) & @xmath25 & @xmath25 & @xmath25 & @xmath26 ( fixed ) + @xmath27 ( kev ) & @xmath28 & @xmath29 & @xmath30 & @xmath31 + @xmath32 ( kev ) & @xmath33 & @xmath34 & @xmath35 & @xmath36 + area @xmath37 ( @xmath38 ) & @xmath39 & @xmath40 & @xmath41 & @xmath42 + area @xmath43
( @xmath38 ) & @xmath44 & @xmath45 & @xmath46 & @xmath47 + bb1 flux & @xmath48 & @xmath49 & @xmath50 & @xmath51 + bb2 flux & @xmath52 & @xmath53 & @xmath54 & @xmath55 + total flux & @xmath56 & @xmath57 & @xmath58 & @xmath59 + @xmath60(bol ) ( ergs s@xmath9 ) & @xmath61 & @xmath62 & @xmath63 & @xmath64 + @xmath65(bol ) ( ergs s@xmath9 ) & @xmath66 & @xmath67 & @xmath68 & @xmath69 + @xmath70(dof ) & 1.1(187 ) & 1.1(84 ) & 1.1(194 ) & 1.2(188 ) + + + epic pn exposure ( ks ) & 11.5 & 6.9 & 17.0 & 26.5 + epic pn live time ( ks ) & 8.1 & 6.2 & 15.8 & 24.4 + off - axis angle ( arcmin ) & 0.0 & 8.8 & 0.0 & 0.0 + count rate ( s@xmath9 ) & 10.6 & 4.8 & 5.8 & 3.4 + epoch ( mjd / tdb ) & 52890.5642044 & 52924.0000320 & 53075.4999960 & 53266.4999776 + period ( s ) & 5.539344(19 ) & 5.53943(8 ) & 5.539425(16 ) & 5.539597(12 ) + [ summary ] as with the earlier data , the 2004 september source spectrum was accumulated in a @xmath71 radius aperture which encloses @xmath72 of the energy . background was taken from a circle of the same size displaced @xmath73 along the readout direction .
the spectra were grouped into bins containing a minimum of 400 counts ( including background ) and fitted using the xspec package . for this analysis
the column density is held fixed at @xmath74 @xmath8 , the value determined from previous fits , which are all consistent .
the best fit to the two - temperature blackbody model yields temperatures of @xmath75 kev and @xmath76 kev with a fit statistic of @xmath77 for 188 degrees of freedom ( fig .
[ specfig ] ) .
each of these two temperatures , which we refer to as `` warm '' and `` hot '' , respectively , remained essentially the same , nominally to within @xmath78 of that reported for the first _ xmm - newton _ observation ( see table [ summary ] ) .
however , the warm and hot blackbody luminosities declined by 33% and 70% , respectively , in one year . with the temperature of the hot component remaining essentially constant , its decline in luminosity is attributable to a decrease in its emitting area , @xmath79
the most recent value , @xmath80 @xmath38 , is @xmath81 of the ns surface area . in the case of the warm component ,
the decay in flux is modest , so that it is not yet possible to decide within the errors whether the temperature or the area is the primary variable .
the most recent area measurement of the warm component , @xmath82 @xmath38 , is @xmath83 of the ns surface area .
0.15 in -0.05 in phase - resolved spectroscopy using the new _ xmm - newton _ data set is compared here to that reported in paper ii . as before , we fit for the intensity normalization in each of 10 phase bins with the two temperatures and column density fixed at the phase - averaged value listed in table [ summary ] .
this is equivalent to determining the relative projected area of emission as a function of rotation phase .
the fits are again found to be each statistically acceptable as a set , so no significant test can be made for variations in additional parameters such as the temperatures or column density .
the results of the phase - resolved spectral fits are shown in figure [ phase ] .
comparison between the data sets taken a year apart shows that the modulation with phase of each blackbody component has remained steady to within the measurement errors .
the phase alignment of the two temperature components has remained the same , and the pulses peak at the same phase .
this is consistent with the picture of a small , hot region surrounded by a warm , concentric annulus that occupies @xmath84 the surface area of the star .
neither component disappears at any rotation phase .
in particular , if the hot component were completely eclipsed , the spectral decomposition in figure [ specfig ] indicates that the light curves at @xmath85 kev should dip to zero ; clearly they do not .
with the set of _ xmm - newton _ measurements spanning a year , a more accurate characterization of the x - ray flux decay of is possible than that reported in the initial _ rxte _ study of @xcite .
the bolometric flux over time , shown in figure [ history ] , reveals a new and complex behavior . while the _ rxte _ data were consistent with an exponential decay of time constant @xmath86 days , the subsequent _ xmm - newton _ measurements show that the inferred bolometric luminosity of the two spectral components are not declining at the same rate . from the data presented in table [ summary ] we find that the luminosity of the hotter temperature component falls exponentially with @xmath87 days , while the warm component decreases with @xmath2 days , which , although less reliably characterized , is clearly longer than @xmath88 .
the shorter time constant is very close to the one describing the _ rxte _ data .
furthermore , the _ rxte _ spectrum itself is fitted best by a single temperature blackbody of @xmath89 kev @xcite , without a power - law contribution , which is consistent with our interpretation of the _ xmm - newton _ spectrum . since _
rxte_is sensitive only at energies @xmath90 kev where the hotter blackbody component dominates , the overall x - ray decay from the beginning of the outburst to the present appears consistent with separate exponential time constants corresponding to the two distinct thermal - spectrum components .
we also note that it is not possible to fit an alternative power - law temporal decay to the hot blackbody flux ; such a model would require a decay index that steepens with time . as shown in figure [ history ] , we expect that the x - ray flux of will return to its historic quiescent level by the year 2007 .
if the quiescent spectrum is to match that observed before the current outburst , then it would likely resemble the _ rosat _ observation of 1992 march 7 .
although of relatively poor quality , the _ rosat_spectrum can be reasonably well fitted with a single blackbody of @xmath91 kev covering @xmath92 @xmath38 and @xmath93 ergs s@xmath9 ( paper i ) .
this blackbody is significantly cooler and larger than the outburst warm component of @xmath94 kev and @xmath95 @xmath38 , so it should be present even now , although masked by the fading outburst emission .
we estimate that the @xmath96 kev blackbody spectrum measured with _ rosat
_ would contribute 30% of the latest ( 2004 september ) _ xmm - newton _
measured flux at 0.5 kev .
therefore , we expect that such a cooler component will begin to dominate the soft x - ray spectrum in late 2006 , consistent with the projection of the flux in figure [ history ] .
the energy - resolved folded light curves from the 2003 september 8 _ xmm - newton _ observation ( fig .
[ pulse]a ) show that the pulse peak , in general , is somewhat narrower than a sinusoid , an effect that is more pronounced at higher energy .
examination of the light curves measured one year later ( 2004 september 18 ; fig .
[ pulse]d ) shows a clear energy - dependent change .
this is easily seen by forming the ratio of the curves taken at the two epochs , normalized by their count - weighted mean .
these ratio curves , presented in fig
. [ ratio ] , show changes in amplitude of up to 20% .
specifically , the pulse shape of the lower - energy light curves has evolved to a smaller amplitude , more sinusoidal profile .
this change is highly significant as determined by the reduced @xmath97 ( 24 degrees - of - freedom ) for the null hypothesis of a flat line ( see labels on fig . [ ratio ] ) .
in particular , the most pronounced variation is found in the 1.01.5 kev band with @xmath98 corresponding to an infinitesimal probability of the pulse shape being constant . in contrast , the pulse shape of the higher energy bins remained essentially unchanged , as evidence by their @xmath97 values near unity .
this trend is reflected in the two intervening observations ( fig .
[ pulse]c and [ pulse]d ) as well .
0.1 in the pulse profile must be made up of some combination of pulsed emission intrinsic to each of the two measured spectral components . in any given energy band ,
the changing relative contribution of the soft and hard spectral components forces the combined pulse shapes to evolve in time .
one way of quantifying this change is to define the pulsed fraction @xmath99 , the fraction of flux above the minimum in the light curve .
this is found to increase smoothly with energy from 34% at less than 1 kev to @xmath100 above 5 kev at the earlier epoch ( see table [ pulsefit ] and fig .
[ pulse ] ) . at the latest epoch reported herein , the pulsed fraction in the lower energy band decreased to 26% while it remained essentially unchanged at the higher energy .
the simplest hypothesis for modeling the light curve evolution is to assume that the intrinsic light curve of each spectral component is steady in time .
if so , the pulsed fraction of the light curve at a given energy and epoch @xmath101 can be expressed as a linear combination of two light curves of fixed pulsed fractions @xmath102 and @xmath103 weighted by the ratio of counts from the respective spectral component @xmath104 and @xmath105 , @xmath106 + f_{bb2}\,r(e , t ) } } , \label{pf}\end{aligned}\ ] ] where @xmath107 $ ] is the normalized flux ratio .
this expression is only valid in the case that the minima of the two light curves coincide , as is evident for the profiles .
for an exponential decay of the component luminosities , the time - dependent model for the pulsed fraction is simply , @xmath108 where @xmath109 is the flux ratio at time @xmath110 and @xmath111 d is the differential time constant .
we can test this hypothesis directly by fitting for @xmath102 and @xmath103 using a joint least - squares fit to the set of 24 ( six energy bands at four epochs ) measured pulsed fractions @xmath101 given in table [ pulsefit ] and the ratio @xmath112 for the double blackbody spectral model tabulated in table [ specratio ] . treating the ratio @xmath112 as the independent variable , the best fit yields @xmath113 for the hotter component and @xmath114 for the cooler one .
the fit statistic is @xmath115 for 22 degrees - of - freedom , corresponding to a probability @xmath116 .
the lower panel of fig . [ history ] shows the best fit model for the decay of the the 1.01.5 kev pulsed fraction over time and its extrapolation to an epoch dominated by the warm spectral component . on the other hand ,
if we apply the blackbody plus power - law model , the corresponding values of @xmath112 , also listed in table [ specratio ] , are completely different and the result is @xmath117 for 22 degrees - of - freedom , corresponding to a probability @xmath118 , 13 times less likely relative to the double blackbody model , for the assumption of fixed intrinsic pulsed fractions .
evidently the pulsed fraction of the hotter component is much higher than for the cooler component .
this difference would account for the gradual shift from a sharper , more triangular pulse shape found at higher energies , to a rounder and more symmetric sinusoidal light curve seen at lower energies .
as was hypothesized above , the decrease in pulsed fraction with time at low energies follows from the fact that the hot spectral component is decaying more rapidly than the cooler one , and so its contribution to the pulsed fraction in the low - energy bin is declining . to the extent that we are not able to discern differences greater than @xmath119 , which represents the measurement uncertainty , the pulsed fractions @xmath102 and @xmath103 are individually independent of energy .
however , we also can not rule out a model in which @xmath102 itself increases with increasing energy .
0.15 in 0.15 in 0.05 in 0.15 in 0.15 in fig .
_ continued _ energy - dependent pulse profiles of obtained with the _ xmm - newton _ epic pn detector for ( _ c _ ) 2004 march 11 and ( _ d _ ) 2004 september 18 .
the epochs of phase zero are given in table [ summary ] .
background has been subtracted .
pulsed fraction at low x - ray energies has decreased in time , while it has remained essentially unchanged at high energy .
0.05 in it is useful to model the shape of the light curves of in detail to further quantify their evolution .
overall , the pulse profiles suggest some linear combination of sinusoidal and triangular functions , for the soft and hard x - rays , respectively . in the following , we model the photon counts as a function of phase , @xmath120 , at a given energy @xmath121 and epoch @xmath122 by the two - component model , @xmath123 where @xmath124 + \gamma_s(e , t)\ ] ] and @xmath125 + \gamma_t(e , t ) , \cr \hfill \quad if |\phi-\phi_t| < \delta/2\cr \gamma_t(e , t ) , \hfill if |\phi-\phi_t| \geq \delta/2 \cr } \ ] ] here
, @xmath126 is the amplitude of the pulsed signal and @xmath127 represent the `` unpulsed , '' or minimum level , for the sinusoidal component , @xmath128 and @xmath129 are the analogous parameters for the triangular component , and @xmath130 is the duty cycle ( full width ) of the triangular pulse .
initial fits to the light curves indicate that the model can be constrained by fixing the triangular width @xmath130 to a single value for all observations .
the same proves true for the relative phase @xmath131 between the peak fluxes of the two components .
accordingly , we used a bootstrap approach to determine these shape parameters . after finding a global fit by iteration
, the triangular width is set at @xmath132 cycles .
we then determined @xmath131 , which is found to be consistent with zero .
finally , the absolute phase @xmath133 at each epoch was fixed .
we therefore conclude that there is no need to allow energy or time dependence for the individual component pulse shapes .
lcccc 5 + @xmath134 & @xmath135 & @xmath136 & @xmath137 & @xmath138 + @xmath139 & @xmath140 & @xmath141 & @xmath142 & @xmath143 + @xmath144 & @xmath145 & @xmath146 & @xmath147 & @xmath148 + @xmath149 & @xmath150 & @xmath151 & @xmath152 & @xmath153 + @xmath154 & @xmath155 & @xmath156 & @xmath157 & @xmath158 + @xmath159 & @xmath160 & @xmath161 & @xmath162 & @xmath163 + @xmath134 & @xmath164 & @xmath165 & @xmath166 & @xmath167 + @xmath139 & @xmath168 & @xmath169 & @xmath170 & @xmath171 + @xmath144 & @xmath172 & @xmath173 & @xmath174 & @xmath172 + @xmath149 & @xmath168 & @xmath175 & @xmath176 & @xmath177 + @xmath154 & @xmath175 & @xmath178 & @xmath179 & @xmath180 + @xmath159 & @xmath181 & @xmath182 & @xmath183 & @xmath184 + @xmath134 & @xmath185 & @xmath186 & @xmath187 & @xmath188 + @xmath139 & @xmath189 & @xmath190 & @xmath191 & @xmath192 + @xmath144 & @xmath193 & @xmath194 & @xmath195 & @xmath192 + @xmath149 & @xmath196 & @xmath197 & @xmath198 & @xmath199 + @xmath154 & @xmath200 & @xmath201 & @xmath202 & @xmath203 + @xmath159 & @xmath204 & @xmath205 & @xmath206 & @xmath207 + @xmath134 & @xmath208 & @xmath209 & @xmath210 & @xmath211 + @xmath139 & @xmath212 & @xmath213 & @xmath214 & @xmath215 + @xmath144 & @xmath216 & @xmath216 & @xmath217 & @xmath218 + @xmath149 & @xmath219 & @xmath219 & @xmath220 & @xmath221 + @xmath154 & @xmath222 & @xmath223 & @xmath219 & @xmath220 + @xmath159 & @xmath224 & @xmath225 & @xmath226 & @xmath227 + [ pulsefit ] lcccc 5 + @xmath134 & 0.389 & 0.332 & 0.285 & 0.218 + @xmath139 & 0.591 & 0.499 & 0.466 & 0.399 + @xmath144 & 0.800 & 0.702 & 0.696 & 0.661 + @xmath149 & 0.941 & 0.886 & 0.896 & 0.893 + @xmath154 & 0.996 & 0.988 & 0.991 & 0.992 + @xmath159 & 1.000 & 1.000 & 1.000 & 1.000 + @xmath134 & 0.928 & 0.911 & 0.925 & 0.952 + @xmath139 & 0.755 & 0.718 & 0.753 & 0.807 + @xmath144 & 0.533 & 0.480 & 0.524 & 0.567 + @xmath149 & 0.347 & 0.291 & 0.331 & 0.340 + @xmath154 & 0.247 & 0.191 & 0.223 & 0.200 + @xmath159 & 0.376 & 0.278 & 0.325 & 0.261 + [ specratio ] with the shape of the pulse components fixed , we fit the restricted model with just @xmath228 , and @xmath229 ( @xmath230 , since the @xmath229 s are indistinguishable in this fit ) as free parameters .
this model fitted to all 24 individual pulse profiles yields a reduced @xmath70 ranging from @xmath231 with a typical value of @xmath232 , corresponding to a probability of @xmath233 for 22 degrees - of - freedom .
the best - fit model parameters , derived pulsed fractions , and fit statistic for each light curve are presented in table [ pulsefit ] . a clear trend is seen as the pulsed fraction decreases with time .
the best - fit models are shown in figure [ pulse ] overlaid on plots of the energy - resolved pulsed profiles for the four epochs .
we checked that a single component model , either a pure sinusoidal or triangular pulse , does not adequately characterize the data based on an unacceptably large @xmath70 for many of the fits .
it is possible to further reduce the number of free parameters in the fit by constraining the coefficients @xmath228 , and @xmath229 so that the phase - averaged photon count ratios @xmath234 are identical with the ratios derived from the two - temperature blackbody spectral fits in each energy interval at each epoch .
that is , we test the hypothesis that the warm blackbody is responsible for only the sinusoidal pulse and the hot blackbody for only the triangular pulse by requiring @xmath235 at all epochs .
we find that , while reasonable fits to the light curves are possible for the lower and highest energy bands , fits in the @xmath236 kev range proved unacceptable .
this argues against a simple one - to - one correspondence between the spectral and the chosen light curve component decomposition .
it is possible that we have not started with the correct spectral or light curve basis pairs or that the basic pulse profiles are each energy dependent .
it is also possible that there may be a third , unfitted softer component associated with the quiescent flux , as seen by rosat ( see 2.3 ) .
continuing observation of the spectrum as the source fades should clarify the relationship between spectral and light curve components .
while the @xmath237 kev x - ray spectra of axps ( including ) clearly require a two - component fit , it is not possible to prove from spectra alone that either of two very different models , namely a power - law plus blackbody , or a two - temperature blackbody , is correct . for , either model yields an acceptable chi - square when fitted to the x - ray spectrum .
however , in paper ii we presented physical arguments against the properties of the particular power law that results from the power - law plus blackbody fit . unlike some rotation - powered pulsars
whose spectra are fitted by hard power laws plus soft blackbodies , in axps the roles of these two components are reversed .
for , most of the x - rays belonging to the @xmath238 power - law component are _ lower _ in energy than those fitted by the blackbody ( see figure 5a of paper ii ) .
this steep power law must turn down sharply just below the x - ray band in order not to exceed the faint , unrelated ir fluxes . as shown in paper ii
, this would be difficult to achieve in a synchrotron model . and in a comptonization model , where are the seed photons ?
even if such a sharp cutoff existed at the low end of the epic energy band , a model in which the power - law component is due to compton upscattering of thermal x - rays from the surface does not explain why most of the power - law photons have lower energy than the observed blackbody photons unless there is a larger source of seed photons at energies below 0.5 kev .
however , such a seed source would have to be thermal emission from the neutron star . since the two - temperature model uses a large fraction of the surface area for the @xmath239 kev photons ,
we prefer to regard the x - rays in this band as plausible seeds for , rather than as the product of , inverse compton scattering . supporting this assumption in the specific case of
are the shapes of its pulse profiles .
the smooth increase in pulsed fraction with energy , and the strict phase alignment at all energies , are unnatural under the hypothesis of two different emission mechanisms and locations .
the evolving spectral shape and pulse profiles of during its decline from outburst offer new and complementary evidence concerning the appropriate spectral decomposition .
we find that the the pulsed fraction is declining with time at low energies , while remaining essentially constant at high energies .
this fact is consistent with the changing contributions of fluxes in the two - temperature fit as the two components decline at different rates .
the warm and hot components contribute significantly to the soft x - rays , whereas the hard x - rays are supplied entirely by the hot component .
this is quantified in table [ specratio ] , where it is seen that the hot blackbody component accounts for 3060% of the flux below 1.5 kev , and @xmath240 of the flux above 3 kev at all times . under this spectral decomposition , the pulsed fraction at energies above 3 kev is large and unchanging because one and only one spectral component is present at those energies , and it has the larger pulsed fraction of the two spectral components . in contrast , the pulsed fraction at low energies ( @xmath241 kev ) is understood to decrease in time because the hot component makes only a fractional contribution to its light curve , and its fraction decreases faster than the warm spectral component , which has the intrinsically smaller pulsed fraction .
this effect is also illustrated in figure [ history ] , where the modeled pulse fraction is compared to the data .
0.15 in 0.15 in 0.05 in in contrast , consider how the spectral components would contribute to the various energy bands in the power - law plus blackbody spectral model .
table 3 shows that , unlike the two blackbody model , a fitted power - law component accounts for 79% of the soft ( @xmath241 kev ) x - rays at _ all _ epochs , while decreasing its contribution to the hard ( @xmath242 kev ) x - rays over time . in this model ,
the pulsed fraction should remain constant at low energies because only the power law is present there , while at high energies the pulsed fraction should vary because both the blackbody and the power - law components are present , in varying proportions .
this spectral decomposition would drive an evolution of the pulse shapes that is opposite of what is observed .
thus , we find that the detailed evolution of the x - ray emission from further supports the assumption of a purely thermal surface emission model , and leaves no evidence of a steep power law in the @xmath243 kev band , as is commonly fitted to individual observations of axps .
our decomposition of the pulsed light curves is also consistent with the absence of modulation in the _ rosat _ observation of 1992 march 7 , to an observed upper limit of 24% in the @xmath244 kev pspc energy band .
since the _ rosat _ spectrum is fitted by an even cooler blackbody of @xmath96 kev , it can be expected that the pulsed fraction in quiescence will fall below the 25% value fitted to the warm blackbody in _ xmm - newton _ data once that component fades away . for consistency with papers i and ii
, we have parameterized results here in terms of a distance of 5 kpc .
however , this was considered an upper limit based on an absorption kinematic distance to the neighboring supernova remnant [email protected] @xcite and its x - ray measured column density @xmath246 @xmath8 @xcite . in comparison , the column density fitted to a power - law plus blackbody model for was @xmath247 @xmath8 . since on physical grounds we strongly prefer the double blackbody model , which requires @xmath248 @xmath8 , it seems appropriate to reduce the distance estimate to one that is more compatible with the smaller @xmath23 value .
we now assume that the distance to is 2.5 kpc , and explore the consequences of this revision .
the effect is to reduce the inferred x - ray luminosities and blackbody surface areas . for @xmath249 kpc ,
the surface area emitting the warm blackbody component is @xmath250 @xmath8 at all times , which is less than 1/6 the area of a neutron star .
this makes it easier to understand how the pulsed fraction of the softest x - rays can be as large as @xmath251 .
we can also revise the estimate of the total energy emitted in the outburst . at a distance of 2.5 kpc ,
the bolometric luminosity quoted in paper ii is @xmath252 ergs s@xmath9 with respect to the initial time @xmath253 of the outburst observed by _
rxte_@xcite , which corresponds to an extrapolated energy of @xmath254 ergs .
since _ rxte _ is only sensitive to the hotter blackbody component , we should add another contribution from the _ xmm - newton _ measured warm component , which can be estimated as @xmath255 ergs s@xmath9 .
the extrapolation required to integrate the contribution of the warm component is less certain than for the hot component , but it corresponds to an energy of only @xmath256 ergs .
the total estimated energy is then @xmath257 ergs , which is comparable to the amount of heat assumed to be deposited in the crust during a deep heating event @xcite , or the extra energy stored in an azimuthally twisted external magnetic field @xcite . in the following section , we discuss how the observations might probe the actual mechanism of the outburst .
two mechanisms have been invoked to explain the late - time `` afterglow '' of an outburst of a magnetar .
the first involves surface heating by long - lived currents flowing on closed but azimuthally twisted external magnetic field lines , and comptonization of the resulting surface x - ray emission by the same particles @xcite .
the originating event could be a sudden fracture in the crust in response to twisting of the internal magnetic field , or a plastic deformation of the crust that gradually transfers internal magnetic twist to the external field .
it is thought that the major outbursts of sgrs are magnetically trapped fireballs triggered by a sudden fracture , but this is not known to apply to because no bursting behavior was seen , and no prior rotational ephemeris exists to test for a glitch . in any case , the extra energy above that of a pure dipole , @xmath258 , stored in a twisted magnetic field external to the star is , @xmath259 where @xmath260 rad is the azimuthal twist from north to south hemisphere @xcite .
this is enough to account for the extrapolated @xmath261 ergs radiated if @xmath262 is a small fraction of a radian . in this model ,
the x - ray spectrum is a combination of surface thermal emission , and compton scattering and cyclotron resonant scattering of the thermal x - rays .
compton upscattered flux may result in power - law emission that dominates at high energies such as that reported above 10 kev for the axp in kes 73 @xcite .
however , the flat power - law index ( @xmath263 ) does not contribute to the softer spectrum , in particular below 2 kev .
so we consider that the potential power - law tail in the fainter may no be detectable yet .
the decay timescale is determined by the rate at which power is consumed by the magnetospheric currents , which is comparable to that needed to power the observed x - ray luminosities of axps in general and in outburst .
furthermore , it is the evolution of surface heating that basically determines the instantaneous x - ray luminosity and its decay , but more specific predictions of decay curves have not been made . in the second picture , a deep crustal heating event , heat deposited suddenly by rearrangement of the magnetic field
is gradually transferred by diffusion and radiation , and the radiated x - rays are purely thermal . while the originating event might have been a fracture of the crust as in the first scenario , an assumption that the heat is evenly distributed throughout the crust results in a particular decay curve from one - dimensional models of heat transfer @xcite . in this model , the heating is virtually instantaneous , within @xmath264 s , while it is the rate of conduction cooling that determines the observed decay curve .
@xcite assumed a deposition of @xmath265 ergs @xmath266 to a depth of 500 m , which is @xmath267 of the magnetic energy density at the surface . if occurring over the entire crust ,
this is @xmath268 ergs , comparable to the total x - ray energy emitted during the decay of .
however , @xcite showed that the resulting cooling luminosity would follow a @xmath269 power law , which is rather slow compared to the observed exponential decay of the hot component of , while 80% of the energy should be conducted into the neutron star core rather than radiated from the surface .
it is not obvious if the behavior of supports or excludes either of these models .
one interesting possibility is that the slowly decaying warm component is the radiation from a deep heating event that affected a large fraction of the crust , while the hot component of the spectrum is powered by external surface heating at the foot - points of twisted magnetic field lines by magnetospheric currents that are decaying more rapidly .
it is not yet clear if the flux decay of the warm component is primarily an effect of decreasing temperature or decreasing area , because the decay has so far been modest .
it is possible that the decay of the warm component _ is _ consistent with a @xmath269 power law rather than an exponential of @xmath2 days .
fits to @xmath270 as a function of time allows a power - law decay index in the range @xmath271 to @xmath272 .
in contrast , because of its more rapid decay , it is evident that the hotter component is declining exponentially , in area rather than in temperature .
this shrinking in area might be easier to understand as a decay of the currents or a rearrangement of the magnetic field lines that are channeling the heating on the surface .
but there is no good evidence yet of an inverse compton scattered component from the particles responsible for that current , at least not at energies @xmath273 kev .
perhaps the solid angle subtended by the twisted field lines in the magnetosphere is too small to scatter most of the thermal photons .
each of the models is challenged in some aspect by the observations , but each may still have some applicability . observing the decay through to quiescence should help to clarify the situation .
the pulsed light curves offer an additional diagnostic of the emission and viewing geometry . if we adopt the hypothesis that virtually all of the @xmath243 kev flux is surface thermal emission , then the symmetry of the light curves and their strict pulse - phase alignment as a function of energy argue for a concentric geometry in which a small hot spot is surrounded by a larger , warmer region . the large observed pulsed fractions are achievable in a realistically modeled ns atmosphere that accounts for the different opacities of the normal modes of polarization in a strong magnetic field @xcite .
many geometries were modeled by @xcite , but only total pulsed fractions were reported rather than detailed light curves .
it is important to model the pulse profiles completely , since the number of peaks and their phase relationship depend in detail on the surface and viewing geometry .
in addition , the relative effects of fan versus pencil beaming , as regulated by the anisotropic opacity in a strong magnetic field , allow the number of peaks in the light curve to differ from the number of hot spots on the surface . whereas the pulsed fraction of the hot component in is 53% , this is not high enough to require a single spot .
it is possible that the flat interpulse region of the triangular , hard x - ray pulse is actually emission from an antipodal spot at large viewing angle .
@xcite showed that for large magnetic fields , most of the flux would be in a broad fan beam , while @xcite argue that cyclotron resonant scattering in the magnetosphere can significantly reduce this effect .
detailed physical modeling of the pulse profiles of is needed before definite conclusions about the emission and viewing geometries can be drawn .
four sets of _ xmm - newton _ observations of spanning a year , clarify several new behaviors that might apply to other axps and axp - like objects : * a two - temperature blackbody model is arguably preferred over the standard blackbody+power law model , * the components of the compound spectrum are decaying at rates that differ significantly , and * the component temperatures of the blackbody emission are nearly constant , implying that the area of emission is steadily decreasing .
several outstanding questions remain that can be addressed by continuing x - ray observations of over the next few years .
will the current spectrum evolve into the prior quiescent one ? is the cool quiescent spectrum still present so that it will begin to dominate over the warm temperature component ? to what degree is the quiescent emission pulsed ? why do the temperatures hardly change with time , requiring the apparent decrease in area ?
what is the emission geometry on the ns and what can we learn from it ?
future observations will test our predictions for the flux decay and pulsed fractions and allow a better understanding of the relationship between the spectral components and the pulse profile .
ultimately , the goal of this study is to disentangle the processes of thermal and non - thermal emission to which magnetars convert their energy , during outburst and quiescence .
we thank fred jansen and norbert schartel for providing the three _ xmm - newton _ target of opportunity observations of .
_ xmm - newton _ is an esa science mission with instruments and contributions directly funded by esa member states and nasa .
this research is supported by nasa grants nng05gc43 g .
roberts , m. s. e. , ransom , s. m. , gavriil , f. , kaspi , v. m. , woods , p. , ibrahim , a. , markwardt , c. , & swank , j. h. 2004 , in x - ray timing 2003 : rossi and beyond , ed .
p. kaaret , f. k. lamb , & j. h. swank ( melville , ny : american institute of physics ) , 306 | we present a multi - epoch spectral study of the transient anomalous x - ray pulsar obtained with the _ newton x - ray multi - mirror mission _ ( _ xmm - newton _ ) .
four observations taken over the course of a year reveal strong spectral evolution as the source fades from outburst .
the origin of this is traced to the individual decay rates of the pulsar s spectral components .
a two - temperature fit at each epoch requires that the temperatures remains nearly constant at @xmath0 kev and @xmath1 kev while the luminosities of these components decrease exponentially with @xmath2 days and @xmath3 days , respectively .
the integrated outburst energy is @xmath4 ergs and @xmath5 ergs for the two spectral components , respectively .
one possible interpretation of the _ xmm - newton _ observations is that the slowly decaying cooler component is the radiation from a deep heating event that affected a large fraction of the crust , while the hotter component is powered by external surface heating at the foot - points of twisted magnetic field lines , by magnetospheric currents that are decaying more rapidly .
the energy - dependent pulse profile of is well modeled at all epochs by the sum of a broad pulse that dominates in the soft x - rays and a narrower one at higher energies .
these profiles peak at the same phase , suggesting a concentric emission geometry on the neutron star surface .
the spectral and pulse evolution together argue against the presence of a significant `` power - law '' contribution to the x - ray spectrum below 8 kev . the extrapolated flux is projected to return to the historic quiescent level , characterized by an even cooler blackbody spectrum , by the year 2007 . |
the emergence of schemes that utilize trapped ions or atoms for quantum information , and the interest in quantum statistics of ultra cold atoms , have provided renewed interest for laser cooling techniques . starting with the case of a single atom
, the present goal is now to laser cool larger numbers of atoms and to prepare them for applications e.g. as a quantum register .
the cooling of a large number of particles using lasers is a prerequisite for coherent control of atomic systems .
trapped ions in linear paul traps are currently considered to be promising candidates for a scalable implementation of quantum computation @xcite
. long lived internal states serve to hold the quantum information ( qubits ) and are manipulated coherently ( single - bit gates ) by laser beams focused on the ions @xcite .
the excitation of the ion s common vibrational motion ( gate mode ) provides the coupling between qubits which is necessary for two - bit quantum gate between ions .
the cirac - zoller proposal for a two - bit gate @xcite requires that initially the gate mode is optically cooled to the ground state .
later mlmer and srensen proposed a two - bit gate which releases this condition and only requires cooling into the lamb - dicke regime , typically with a mean vibrational quantum number @xmath11 @xcite . in general ,
the motion of a string of @xmath2 cold trapped ions in a linear paul trap is described by @xmath3 harmonic modes of vibration . cooling _ all modes _ close to the vibrational ground state is a prerequisite to realizing quantum gates with high fidelity . in typical experimental situations
this requires cooling techniques which lead to lower mean vibrational state than simple doppler cooling .
sideband cooling has been shown to achieve sub - doppler temperatures @xcite but proves increasingly inefficient for cooling many modes .
the recently proposed method of eit - cooling provides parallel multi - mode cooling while its experimental effort is significantly reduced . the first part of this paper ( sec . [ eitprinc ] ) describes the principle of the eit - cooling method and illustrates its results with several realistic examples , the cooling of a single hg@xmath0 ion held in a paul trap , and of neutral rb atoms confined in a far - detuned optical dipole trap ( sect .
[ hgrb ] ) . in section [ expt ]
we report the first experimental demonstration of the eit - method , achieving ground state cooling of a single trapped ca@xmath0 ion . the detailed description of all necessary experimental ingredients and measurements is meant to provide a recipe for implementing eit - cooling in various experimental situations .
we conclude by a proposal for cooling a linear string of ions ( sec . [ linear ] ) , and discuss the relevance for quantum information processing .
the theory of laser cooling of atoms in a harmonic trap has been extensively discussed in the literature , see @xcite for a comprehensive review . in many experimental situations a simple , intuitive rate equation picture @xcite applies which can also be derived from the general theory in the appropriate limits @xcite . in this section
we recall those intuitive considerations , a detailed theoretical derivation being beyond the scope of this paper .
consider a trapped 2-level atom with states @xmath4 and @xmath5 , excited below saturation by a laser detuned from resonance by @xmath6 .
the scattering rate , i.e. steady - state excited state population times decay rate , is denoted by @xmath7 .
the atom is coupled to one vibrational degree of freedom , a harmonic oscillator with frequency @xmath8 and quantum states @xmath9 , such that the internal transition @xmath10 where the spatial extension of the motional wave packet @xmath11 is smaller than the laser wavelength @xmath12 . in this
limit the relevant transitions are @xmath13 ( carrier ) and @xmath14 ( sidebands ) .
see fig .
[ basics]a for an illustration .
the probability that scattering of a photon takes the atom from @xmath15 to @xmath16 has two contributions , corresponding to the two possible intermediate states @xmath17 or @xmath18 .
the rate coefficients @xmath19 for cooling and heating transitions ( see fig . [
basics]b ) are therefore given by @xmath20 summing up all processes in lowest order @xmath21 , a rate equation for the populations @xmath22 of the states @xmath9 is obtained which leads to a simple dynamical equation for the mean vibrational excitation @xmath23 , @xmath24 in the more general case that the laser @xmath25 vektor is at an angle @xmath26 to the direction of vibration , and that the spontaneous recoils are spatially distributed with an average projection @xmath27 on the vibrational axes @xcite , eq .
( [ aplusminus ] ) changes into @xmath28 using eqs .
( [ evolution],[aplusminus1 ] ) , the cooling limit @xmath29 and the cooling time constant , @xmath30 can be easily evaluated .
one finds the well - known cooling limits @xmath31 for doppler cooling where @xmath32 and @xmath33 , and @xmath34 for sideband cooling where @xmath35 and @xmath36 @xcite .
it is useful to look at the transitions displayed in fig .
[ basics]a from a different viewpoint : absorption on one of the sidebands causes a drift of the motional energy while carrier absorption , being followed by either cooling or heating emission , causes a diffusion of the energy which in average contributes to heating .
therefore for optimum cooling , absorption on the cooling sideband should be maximized while carrier and heating - sideband absorption should be minimized .
this leads to the principle of eit - cooling .
electromagnetically induced transparency ( eit ) , like coherent population trapping or `` dark resonance '' , is a manifestation of quantum interference between atomic transition amplitudes .
a review can be found in @xcite .
eit arises in three- ( or multi- ) level systems and consists in the cancellation of the absorption on one transition induced by simultaneous coherent driving of another transition .
consider a 3-level atom as shown in fig . 1
, with ground state @xmath4 , stable or metastable state @xmath37 and excited state @xmath5 .
state @xmath5 has linewidth @xmath38 and is coupled to @xmath4 and @xmath37 by dipole transitions which are laser - excited .
eit arises when the detunings of the two lasers from state @xmath5 are equal : the system evolves into a coherent superposition of @xmath4 and @xmath37 , and light scattering seizes . in order to use this situation for cooling , the transition @xmath39 ) at detuning @xmath40 above resonance .
then the absorption spectrum on the transition @xmath41 seen by the other ( `` cooling '' ) laser is described by a fano - like profile @xcite which has three characteristic features , see fig .
1a , b : the broad resonance at @xmath42 , the dark resonance ( eit ) at @xmath43 , and the narrow resonance at @xmath44 , where @xmath45 is the ac - stark shift created by the coupling laser . for eit - cooling ,
the laser frequencies are set to the dark resonance condition , @xmath43 .
then , taking into account the harmonic motion , all @xmath46 transitions are cancelled .
furthermore , by choosing a suitable rabi frequency @xmath47 , the spectrum is designed such that the @xmath48 ( red ) sideband corresponds to the maximum of the narrow resonance , whereas the blue sideband falls into the region of the spectrum of small excitation probability , as shown in fig .
1b . the conditions for enhancing the red - sideband absorption while eliminating the carrier is therefore : @xmath49 in summary , due to the eit condition no absorption happens unless the ion is moving , and by adjusting the coupling rabi frequency such that the bright resonance matches the red sideband , absorption accompanied by the loss of one phonon is made much more probable than absorption accompanied by gaining one phonon .
the fact that the diffusion normally caused by @xmath50 becomes independent of the projection of the vibrational mode on the laser direction @xcite .
a calculation starting from the full quantum mechanical master equation @xcite shows that in the lamb - dicke regime and below saturation of the cooling transition , the eit - cooling process can be approximated by the same rate equation discussed above , only that now the appropriate coefficients @xmath19 contain the quantum interference around @xmath51 .
the explicit form of the coefficients is given in @xcite .
[ 3levelsim ] shows the cooling dynamics calculated from a monte - carlo simulation of the master equation and from the rate equation approximation .
eit - cooling shows to be a robust technique even if the system involved deviates from the ideal 3-level scheme .
for example , in the experiments described below , due to non - ideal polarizations of the laser beams , photons were scattered from a fourth level which introduced extra heating . the theoretical cooling limit for this
particular case increases by about a factor of 3 , and the experiment still yielded cooling close the ground state @xcite , see sec .
[ expt ] .
another interesting situation is when the excited state decays through an additional channel into a fourth level from which it has to be repumped with an additional laser .
we studied the example of a single trapped hg@xmath0 ion where the states @xmath52 for eit - cooling are taken to be the @xmath53 and @xmath54 levels , respectively .
the @xmath55 level into which @xmath5 decays with 1/3 probability is pumped out resonantly back to @xmath5 .
our model calculation is shown in fig .
[ hgsim ] .
a cooling limit of @xmath56 is achieved which is significantly below the doppler cooling limit of about 20 and which could help to improve the hg@xmath0 single ion frequency standard @xcite .
finally , it would certainly be also interesting to apply eit - cooling in cold - atom experiments , e.g. to rb or cs .
in fact , the outer @xmath57-states of any @xmath58 transition offer a suitable level systems for its implementation .
specifically , we give an estimate for the application of eit - cooling to @xmath59rb atoms in a co@xmath60-laser standing wave dipole trap @xcite .
one main difference to an ion trap are that the trap frequency is generally lower and varies along the laser beam waist .
we assumed a distribution of trap frequencies between 25 and 50 khz , corresponding to the intensity variation over one rayleigh length .
we applied the rate equation model to the @xmath61 3-level system and included resonant repumping from @xmath62 .
the detailed parameters and the calculated results are displayed in fig .
[ dipole ] . according to that estimate
it should be possible to cool to @xmath63 , much below the doppler cooling limit , on a time scale of ms .
it should be remarked that in this case the motional state before cooling may be outside the lamb - dicke regime , such that initially the considerations of sec .
[ cbasics ] do not apply . instead , higher order sidebands , i.e. transitions with @xmath64 have to be taken into account . in the example of fig .
[ dipole ] one finds that the absorption probability for cooling transitions is larger than that for heating transitions for many higher - order sidebands ( over several 100 khz ) , therefore starting outside the lamb - dicke regime does not seem to restrict the application of eit - cooling .
the theoretical background of the method described above has to be modified only slightly to apply it to our experiment with a single trapped ca@xmath0 ion .
we implemented the eit - scheme on the @xmath65 transition , whose zeeman sublevels @xmath66 constitute a four - level system .
we denote these levels by @xmath67 and @xmath68 , see fig . [ levels ] .
three of the levels , @xmath67 and @xmath69 , together with the @xmath70- and @xmath71-polarized laser beams , form an effective three - level system of the kind considered above @xcite .
the modifications due to the fourth level @xmath72 are discussed in sec .
[ procedure ] .
the ca@xmath0 system seems ideal for the demonstration of the eit - cooling method since in previous experiments we have found very small heating rates @xcite and therefore the cooling results can be measured with high precision .
we will briefly introduce the experimental setup , for details we refer to @xcite .
first we describe all light sources which are necessary for the experiment : for the s@xmath73p@xmath74 dipole transition at 397 nm , light from a ti : sapphire laser near 793 nm is frequency doubled in a lbo crystal inside a build - up cavity .
up to 5 mw of light near 397 nm are sent to the trap setup through a single mode uv fiber . as displayed in fig .
[ setup ] , we use the beam deflected into @xmath75 order from an acousto - optical modulator ( ao 1 ) driven at 80 mhz for doppler cooling . typically , this radiation is red - detuned from the resonance by 20 mhz .
the two beams for eit - cooling , labeled @xmath70 and @xmath71 , are derived from the uv fiber output as the @xmath76 deflection orders of two further acousto - opical modulators ao 2 and ao 3 driven at 86 mhz and 92 mhz , respectively .
we chose a detuning of @xmath77 75 mhz ( natural linewidth of the transition is 20 mhz ) .
the difference in the ao 2 and 3 drive frequencies compensates for the 12 mhz zeeman splitting of the @xmath67 states in the quantization b - field of 4.4 gauss .
the three output beams are focused onto the single ion in the paul trap from different directions corresponding to their respective polarisations and the orientation of the b - field . for a reproducible and fine tuning of the light intensity (
typically few 10 @xmath78w in a waist of @xmath79 60 @xmath78 m ) the rf - drive power of all ao s is computer - controlled .
the rf - power , and therefore the light beams , can also be switched on and off within a few @xmath78s .
light from a second ti : sapphire laser near 729 nm is used for excitation on the quadrupole transition from the s@xmath74 ground state to the metastable d@xmath80 state ( spontaneous lifetime 1 s ) .
the laser frequency is stabilized to a bandwidth of @xmath81 100 hz using a stable high - finesse optical reference cavity and a pound - drever - hall servo circuit .
the laser frequency can be tuned by an acousto - optical modulator , its intensity is controlled through the rf - drive power . with this laser
we can selectively excite vibrational sidebands of the s@xmath82d@xmath80 transition at a few mhz detuning from the carrier . the relative excitation probability on the first red and the first blue sideband reveals information about the ion s vibrational quantum state as described in sect . [ t ] .
the light of grating stabilized laser diodes near 866 nm and 854 nm serves to pump the ion out of both metastable d levels .
while the light of the laser at 866 nm is turned on continuously troughout the experiment , the laser at 854 nm is also switched on and off by a computer - controlled acousto - optical modulator .
our ion trap is a 3d - quadrupole paul trap : a ring with inner diameter of 1.4 mm is formed from a mb - wire of 0.2 mm diameter , and two endcaps are made out of sharpened tips of the same material at 1.2 mm distance @xcite .
the ring electrode is driven by @xmath79 650 v radio frequency at 20.9 mhz , while the tips are held near ground potential .
for the experiments described below , a single @xmath83ca@xmath0 ion is generated by electron bombardment of a weak thermal calcium beam .
its oscillation frequencies in the trap potential ( @xmath84 , @xmath85 , @xmath86 ) are ( 1.69 , 1.62 , 3.32 ) mhz . additional compensation electrodes , together with a differential voltage applied to the tips , can be used to compensate for stray static electric fields and to shift the ion into the center of the rf - potential .
the average motional quantum number @xmath87 can be measured with high precision using laser excitation of vibrational sidebands , which are spectrally well resolved since the linewidth of the transition , here s@xmath88d@xmath80 , is well below the vibrational frequencies of the trapped ion .
the situation is schematically depicted in fig . [ dressed ] , where the level scheme obtained for a two - level atom in a harmonic trapping potential is shown together with the carrier ( @xmath89 ) and the first sideband ( @xmath90 ) transitions .
the excitation probabilities @xmath91 and @xmath92 on the @xmath93 to @xmath94 transition ( red sideband ) and the @xmath93 to @xmath95 transition ( blue sideband ) are proportional to @xmath96 and @xmath97 , respectively .
@xmath91 drops to zero for @xmath98 , see fig . [
dressed ] . for a thermal phonon distribution with mean phonon number @xmath87 , the probability for finding @xmath96 phonons is @xmath99 .
thus @xmath87 is directly related to the measured ratio @xmath100 .
it can be shown that this result holds for incoherent _ and _ coherent excitation if the phonon distribution @xmath101 is thermal .
our cooling results will be specified either as @xmath87 or by the zero - phonon occupation @xmath102 for the three trap vibrational modes . one way to determine the cooling result is to first excite a vibrational sideband at 729 nm and then probe the s@xmath82p@xmath74 transition at 397 nm ( shelving method ) @xcite .
if sideband excitation to the d@xmath80 state was successful , no fluorescence is emitted on the s@xmath103d@xmath80 transition , whereas in the other case we observe many scattered photons .
this procedure is repeated 100 times on both the red and the blue sideband , and the difference in their excitation probabilities yields @xmath87 and @xmath102 .
the second method we used is driving coherent rabi oscillations on the blue sideband @xcite .
we record the excitation probability @xmath104 as a function of 729 nm excitation time @xmath105 .
theoretically , @xmath106 , where @xmath107 denotes the s@xmath82d@xmath80 rabi frequency and @xmath21 the lamb - dicke factor @xcite .
the analysis of a time evolution @xmath104 reveals the mean phonon number @xmath87 or can even be used to measure all coefficients of the phonon distribution @xmath101 . applying the eit - cooling principle to our system
, we find that the intensity of the eit-@xmath70 beam should be such that the stark shift equals the frequency of the vibrational mode to be cooled .
if the waist sizes of the beams are known , their intensities can be calculated .
there are also a few possible methods to experimentally determine the ac - stark shift or the corresponding rabi frequencies : + a ) a rough estimate may be gained from the fluorescence rate as a function of light power .
finding first the approximate saturation power level , one can then deduce the necessary laser power for both eit - beams .
+ b ) in the @xmath108 three - level system excited by the 866 nm and 397 nm lasers , one can measure the fluorescence rate if one of the laser frequencies is tuned over the resonance and the other is kept fix .
optical bloch equations are then used to fit the excitation spectrum and determine the relevant rabi frequencies @xcite .
+ c ) the time constant of optical pumping between the @xmath109 states may be used . to probe their population ,
one of the two states must be selectively excited to the d@xmath80 level .
+ d ) one can measure the broadening and ac - stark shift of the narrow s@xmath74 to d@xmath80 transition when the ion is simultaneously illuminated with a pulse of 729 nm and one of the eit - beams .
we used the last method to determine the ac - stark shift @xmath110 as a function of power in the eit - beams . from this , we could extrapolate to the power level where @xmath111 . in more detail ,
the ion was excited on the @xmath112 transition with a 3 @xmath78s pulse at 729 nm and _ simultaneously _ with a eit-@xmath70 pulse . although the method suffered from asymetric excitation line shapes and from rapid optical pumping out of the s@xmath113 level , it lead to a good estimate for the right power level . for a final optimization of the eit-@xmath70 beam power we used the cooling results for a specific vibrational mode @xcite .
we investigated eit cooling of the @xmath114 and the @xmath115 oscillation at 1.62 and 3.32 mhz , respectively .
the experiment proceeded in three steps : doppler precooling , eit - cooling and finally determination of the vibrational quantum state , see fig .
[ pulse ] : \(i ) we doppler precooled the ion on the s@xmath74 to p@xmath74 transition at 397 nm ( natural linewidth @xmath116 20 mhz ) .
a detuning of approximately -20 mhz with respect to the @xmath117 transition line was chosen . to avoid optical pumping into the d@xmath118 states , we used the repumping beam near 866 nm @xcite .
doppler cooling was applied for 1.5 ms . the theoretical doppler cooling limit on this transition of 0.5 mk corresponds to mean vibrational quantum numbers of @xmath119 3 and @xmath120 6 .
the cooling limits reached in our experiment were higher , due to the fact that the simple assumption of a two level system in the determination of the doppler limit does not hold in our case and because the 12 mhz zeeman splitting of the s@xmath74 state does not permit optimum detuning for all transitions between the substates .
we experimentally determined the mean excitation numbers after doppler cooling to be @xmath121 and @xmath122 .
\(ii ) after doppler cooling , we applied the two eit - beams , 397nm-@xmath70 and 397nm-@xmath71 for up to 7.9 ms .
a 1.8 ms pulse duration was found to be sufficient , longer cooling time did not lower the mean phonon number ( see sec .
[ dyn ] where we will discuss the cooling time ) .
the power levels were set as discussed above in sec .
[ ac ] . for our setup with laser beam waists of @xmath79 50 @xmath78 m ,
a laser power of @xmath123 50 @xmath78w was used for the @xmath70-beam .
+ the @xmath25-vectors of the two eit - cooling beams enclosed an angle of @xmath124 and illuminated the ion in such a way that their difference @xmath125 had a component along all trap axes ( @xmath126 , where @xmath127 denotes the angle between @xmath125 and the respective trap axis ) . as a result , _
all _ vibrational directions were cooled .
since the vacuum recipient did not allow for two orthogonal beams , we could not realize the ideal constellation of fig .
[ levels ] : while the @xmath70-beam was parallel to the quantization axis as desired , the @xmath71-beam was not orthogonal to it and therefore had a @xmath128-component .
the extra excitation due to this non - ideal constellation is indicated in fig .
[ levels ] by a dashed line .
it results in a slightly higher final mean phonon number .
a detailed discussion is found in ref .
@xcite .
\(iii ) we analyzed the vibrational state after eit - cooling by spectroscopy on the @xmath129 quadrupole transition at 729 nm , as described in sec .
finally a diode laser at 854 nm served to repump the ion from the d@xmath80 to the s@xmath74 level .
with the ac - stark shift @xmath110 set to the frequency of the radial y - mode , we monitored the vibrational state after eit - cooling by exciting the blue sideband of the @xmath130 d@xmath131 transition with a 729 nm pulse and then measuring the @xmath132 level occupation as a function of the pulse length @xmath105 @xcite .
the observed rabi - oscillations were subsequently fitted to determine the mean vibrational occupation number @xmath133 @xcite , see fig .
[ singlemode ] . the lowest mean vibrational number @xmath134 observed corresponds to a 84@xmath135 ground state probability .
we repeated this experiment on the @xmath115-mode at 3.3 mhz after having increased the intensity of the @xmath70-beam to adjust @xmath110 .
for this mode , a minimum mean vibrational number of @xmath136 was obtained , corresponding to a 90@xmath135 ground state probability .
2 we found the cooling results largely independent of the intensity of the @xmath71-beam as long as it is much smaller than the @xmath70 intensity . in our experiment the intensity ratio was @xmath137 and we varied the intensity of the @xmath71-beam by a factor of 4 , with no observable effect on the final @xmath87 . by determining the dependence of the mean vibrational quantum number on the eit cooling time , we measured the cooling time constant for the y - mode to be 250 @xmath78s , as shown in fig .
[ coolingdyn ] . to show that the eit method is suitable to simultaneously cool several vibrational modes with significantly different frequencies of oscillation
, we chose the axial @xmath115-mode at 3.3 mhz , and the radial @xmath114-mode at 1.62 mhz .
the intensity of the @xmath70-beam was set such that the ac - stark shift was roughly halfway between the two mode frequencies .
again we applied the eit cooling beams for 7.9 ms after doppler cooling .
this time we determined the final @xmath138 by comparing the excitation probability on the red and the blue sidebands of the @xmath139 transition .
the result is shown in fig .
[ twomodes ] .
we find both modes cooled deeply inside the lamb - dicke regime ( @xmath140 ) , with @xmath141 and @xmath142 ground state probability .
since eit - cooling allows simultaneous cooling of several modes at different frequencies , it seems to be particularly suited for ion strings in linear traps .
the frequencies of the axial vibrational eigenmodes of a linear string have been calculated @xcite and measured @xcite . for a 10-ion string trapped in a linear trap with a center - of - mass axial frequency of 0.7 mhz ,
the closest inter - ion spacing is found to be 3.0 @xmath78 m ( 4.5 @xmath78 m for n=5 ) .
the axial vibration frequencies are 0.7 mhz , 1.22 mhz , .. 4.6 mhz .
the radial trap frequencies in a linear trap must be made sufficiently high in order to prevent a transition from the linear configuration of @xmath2 ions to a zig - zag configuration .
it was estimated that this transition occurs at @xmath143 , thus the radial trap frequency must exceed 3.7 mhz for 10 ions ( 2 mhz for @xmath2=5 ) @xcite . typically , the radial frequency is chosen higher .
the linear ion trap experiment at innsbruck uses @xmath144 4 mhz , the be@xmath0 experiments at nist @xcite have @xmath144 20 mhz . apart from the purely axial modes there exist @xmath145 radial modes .
their frequencies arrange in a band below @xmath146 which overlaps with the axial mode frequencies .
we now estimate the performance of eit - cooling for a 10-ion string .
the result is displayed in fig .
[ eitlin ] : for a 10-ion string indeed _ all @xmath3 _
vibrational modes are cooled to a mean phonon number @xmath87 below one .
this is promising for the application of cold ion strings for quantum information processing @xcite .
as discussed in the introduction , it is required that all modes which couple to the laser light ( spectator modes ) must be cooled well into the lamb - dicke regime .
the reason for that is that thermally excited spectator modes cause a blurring of the rabi frequency @xmath147 which disturbs the precision of quantum operations . for the case of @xmath148 spectator modes
@xmath149 with @xmath150 and lamb - dicke factors @xmath151 , the initial state is a mixed state .
each time the experiment will happen with slightly different initial conditions and the rabi frequency for carrier or sideband transitions will differ .
the relative blurring reads like @xmath152 ( see equ . 126 in ref .
@xcite ) which we find as small as @xmath153 for our specific example . the maximum number of rabi oscillations
will then be @xmath154 2000 .
in summary we have presented eit - cooling , a novel ground state cooling method which uses electromagnetically induced transparency on two coupled dipole transitions .
we have demonstrated its successful application to a single ca@xmath0 ion in a paul trap .
we have also shown several estimates for its implementation in various trapped - ion and trapped - atom experiments , taking into account different parameter regimes as well as deviations from the ideal case such as extra levels and transitions .
the eit - cooling method , according to these estimates and considering its experimental simplicity , bears the potential to become a standard tool for the preparation of ultracold atoms and ions
. a particularly important application will be cooling of an ion string in a linear trap , as a step in preparing such a string for coherent manipulations in quantum information processing .
this work is supported by the austrian fonds zur frderung der wissenschaftlichen forschung ( sfb15 and start - grant y147-phy ) , by the european commission ( tmr networks quantum information ( erb - frmx - ct96 - 0087 ) and quantum structures ( erb - fmrx - ct96 - 0077 ) ) , and by the `` institut fr quanteninformation gmbh '' .
g. janik , w. nagourney , h. dehmelt , j. opt .
b * 2 * , 1251 ( 1985 ) ; i. siemers , m. schubert , r. blatt , w. neuhauser , p. e. toschek , europhys .
lett . * 18 * , 139 ( 1992 ) ; m. schubert , i. siemers , r. blatt , w. neuhauser , p. e. toschek , phys .
lett * 68 * , 3016 ( 1992 ) ; phys .
a * 52 * , 2994 ( 1995 ) .
q. a. turchette , d. kielpinski , b. e. king , d. leibfried , d.m .
meekhof , c. j. myatt , m.a .
rove , c. a. sackett , c. s. wood , w. m. itano , c. monroe , and d. j. wineland , phys .
rev . * a 61 * , 063418 ( 2000 ) . | a novel method of ground state laser cooling of trapped atoms utilizes the absorption profile of a three ( or multi- ) level system which is tailored by a quantum interference . with cooling rates comparable to conventional sideband cooling , lower final temperatures may be achieved .
the method was experimentally implemented to cool a single ca@xmath0 ion to its vibrational ground state . since a broad band of vibrational frequencies
can be cooled simultaneously , the technique will be particularly useful for the cooling of larger ion strings , thereby being of great practical importance for initializing a quantum register based on trapped ions .
we also discuss its application to different level schemes and for ground state cooling of neutral atoms trapped by a far detuned standing wave laser field . 2 |
after popper , it is a commonplace opinion in philosophy of science that the ` value ' of definitions besides mathematics is generally low .
nevertheless , for many more practical needs , from jurisprudence to health planning , the definitions are necessary to impose theoretical boundaries of a subject despite of their incompleteness and temporariness .
this highly applies to definitions of drugs and drug use . following the standard definitions @xcite , a _ drug _ is a ` chemical that influences biological function ( other than by providing nutrition or hydration ) ' .
a _ psychoactive drug _ is a ` drug whose influence is in a part on mental functions ' .
abusable psychoactive drug _ is a ` drug whose mental effects are sufficiently pleasant or interesting or helpful that some people choose to take it for a reason other than to relieve a specific malady ' . in our study
we use the term ` drug ' for abusable psychoactive drug regardless of whether it s illicit or not .
drug use is a risk behaviour that does not happen in isolation ; it constitutes an important factor for increasing risk of poor health , along with earlier mortality and morbidity , and has significant consequences for society @xcite .
drug consumption and addiction constitutes a serious problem globally .
it includes numerous risk factors , which are defined as any attribute , characteristic , or event in the life of an individual that increases the probability of drug consumption .
a number of factors are correlated with initial drug use including psychological , social , individual , environmental , and economic factors @xcite .
these factors are likewise associated with a number of personality traits @xcite .
while legal drugs such as sugar , alcohol and tobacco are probably responsible for far more premature death than illegal recreational drugs @xcite , the social and personal consequences of recreational drug use can be highly problematic @xcite .
psychologists have largely agreed that the personality traits of the five factor model ( ffm ) are the most comprehensive and adaptable system for understanding human individual differences @xcite .
the ffm comprises neuroticism ( n ) , extraversion ( e ) , openness to experience ( o ) , agreeableness ( a ) , and conscientiousness ( c ) .
a number of studies have illustrated that personality traits are associated to drug consumption .
roncero et al @xcite highlighted the importance of the relationship between high n and the presence of psychotic symptoms following cocaine - induced drug consumption .
vollrath & torgersen @xcite observed that the personality traits of n , e , and c are highly correlated with hazardous health behaviours .
a low score of c , and high score of e or high score of n correlate strongly with multiple risky health behaviours .
flory et al @xcite found alcohol use to be associated with lower a and c , and higher e. they found also that lower a and c , and higher o are associated with marijuana use .
sutina et al @xcite demonstrated that the relationship between low c and drug consumption is moderated by poverty ; low c is a stronger risk factor for illicit drug usage among those with relatively higher socioeconomic status .
they found that high n , and low a and c are associated with higher risk of drug use ( including cocaine , crack , morphine , codeine , and heroin ) .
it should be mentioned that high n is positively associated with many other addictions like internet addiction , exercise addiction , compulsive buying , and study addiction @xcite .
an individual s personality profile plays a role in becoming a drug user .
terracciano et al @xcite demonstrated that the personality profiles for the users and non - users of nicotine , cannabis , cocaine , and heroin are associated with a ffm of personality samples from different communities .
they also highlight the links between the consumption of these drugs and low c. turiano et al @xcite found a positive correlation between n and o , and drug use , while , increasing scores for c and a decreases risk of drug use .
previous studies demonstrated that participants who use drugs including alcohol and nicotine have a strong positive correlation between a and c and a strong negative correlation for each of these factors with n @xcite .
three high - order personality traits are proposed as endophenotypes for substance use disorders : positive emotionality , negative emotionality , and constraint @xcite .
the statistical characteristics of groups of drug users and non - users have been studied by many authors ( see , for example , terracciano et al @xcite ) .
they found that the personality profile for the users and non - users of tobacco , marijuana , cocaine , and heroin are associated with a higher score on n and a very low score for c. sensation seeking is also higher for users of recreational drugs @xcite .
the problem of risk evaluation for individuals is much more complex .
this was explored very recently by yasnitskiy et al @xcite , valeroa et al @xcite and bulut & bucak @xcite .
both individual and environmental factors predict substance use and different patterns of interaction among these factors may have different implications @xcite .
age is a very important attribute for diagnosis and prognosis of substance use disorders .
in particular , early adolescent onset of substance use is a robust predictor of future substance use disorders @xcite .
valeroa et al @xcite evaluated the individual risk of drug consumption for alcohol , cocaine , opiates , cannabis , ecstasy and amphetamines .
input data were collected using the spanish version of the zuckerman - kuhlman personality questionnaire ( zkpq ) .
two samples were used in this study .
the first one consisted of 336 drug dependent psychiatric patients of one hospital .
the second sample included 486 control individuals .
the authors used a decision tree as a tool to identify the most informative attributes .
the sensitivity of 40% and the specificity of 94% were achieved for the training set .
the main purpose of this research was to test if predicting drug consumption was possible and to identify the most informative attributes using data mining methods .
the decision tree methods were applied to explore the differential role of personality profiles in drug consumer and control individuals .
the two personality factors , neuroticism and anxiety and the zkpq s impulsivity , were found to be most relevant for drug consumption prediction .
low sensitivity ( 40% ) does not provide application of this decision tree to real life problems .
in our study we tested the associations with personality traits for different types of drugs separately using the revised neo five - factor inventory ( neo - ffi - r ) @xcite , the barratt impulsiveness scale version 11 ( bis-11 ) @xcite , and the impulsivity sensation - seeking scale ( impss ) @xcite to assess impulsivity and sensation - seeking respectively .
bulut & bucak @xcite detected a risk rate for teenagers in terms of percentage who are at high risk without focusing on specific addictions .
the attributes were collected by an original questionnaire , which included 25 questions .
the form was filled in by 671 students .
the first 20 questions asked about the teenagers financial situation , temperament type , family and social relations , and cultural preferences .
the last five questions were completed by their teachers and concerned the grade point average of the student for the previous semester according to a 5-point grading system , whether the student had been given any disciplinary punishment so far , if the student had alcohol problems , if the student smoked cigarettes or used tobacco products , and whether the student misused substances . in bulut
et al s study there are five risk classes as outputs .
the authors diagnosed teenagers risk to be a drug abuser using seven types of classification algorithms : @xmath0-nearest neighbor , id3 and c4.5 decision tree based algorithms , nave bayes classifier , nave bayes / decision trees hybrid approach , one - attribute - rule , and projective adaptive resonance theory .
the classification accuracy of the best classifier was reported as 98% .
yasnitskiy et al @xcite , attempted to evaluate the individual s risk of illicit drug consumption and to recommend the most efficient changes in the individual s social environment to reduce this risk .
the input and output features were collected by an original questionnaire .
the attributes consisted of : level of education , having friends who use drugs , temperament type , number of children in the family , financial situation , alcohol drinking and smoking , family relations ( cases of physical , emotional and psychological abuse , level of trust and happiness in the family ) .
there were 72 participants .
a neural network model was used to evaluate the importance of attributes for diagnosis of the tendency to drug addiction .
a series of virtual experiments was performed for several test patients ( drug users ) to evaluate how it is possible to control the propensity for drug addiction .
the most effective change of social environment features was predicted for each patient .
the recommended changes depended on the personal profile , and significantly varied for different patients .
this approach produced individual bespoke advice for decreasing drug dependence . in our study , the database was collected by an anonymous online survey methodology by elaine fehrman yielding 2051 respondents .
the database is available online @xcite .
twelve attributes are known for each respondent : personality measurements which include n , e , o , a , and c scores from neo - ffi - r , impulsivity ( imp . ) from ( bis-11 ) , sensation seeking ( ss ) from ( impss ) , level of education ( edu . ) , age , gender , country of residence , and ethnicity .
the data set contains information on the consumption of 18 central nervous system psychoactive drugs including alcohol , amphetamines , amyl nitrite , benzodiazepines , cannabis , chocolate , cocaine , caffeine , crack , ecstasy , heroin , ketamine , legal highs , lsd , methadone , magic mushrooms ( mmushrooms ) , nicotine , and volatile substance abuse ( vsa ) , and one fictitious drug ( semeron ) which was introduced to identify over - claimers .
participants selected for each drug either they never used this drug , used it over a decade ago , or in the last decade , year , month , week , or day .
participants were asked about substances , which were classified as central nervous system depressants , stimulants , or hallucinogens .
the depressant drugs comprised alcohol , amyl nitrite , benzodiazepines , tranquilizers , gamma - hydroxybutyrate solvents and inhalants , and opiates such as heroin and methadone / prescribed opiates . the stimulants consisted of amphetamines , nicotine , cocaine powder , crack cocaine , caffeine , and chocolate . although chocolate contains caffeine , data for chocolate
was measured separately , given that it may induce parallel psychopharmacological and behavioural effects in individuals congruent to other addictive substances @xcite .
the hallucinogens included cannabis , ecstasy , ketamine , lsd , and magic mushrooms .
legal highs such as mephedrone , salvia , and various legal smoking mixtures were also measured .
we use four different definitions of ` drug users ' based on the recency of use .
firstly , two isolated categories ( ` never used ' and ` used over a decade ago ' ) are placed into the class of non - users , and all other categories are merged to form the class of users . secondly , we merge the categories ` used in last decade ' , ` used over a decade ago ' and ` never used ' into the group of non - users and place four other categories ( ` used in last year - month - week - day ' ) into group of users .
this classification is called ` year - based ' one .
also ` month - based ' and ` week - based ' user / non - user separations are considered .
the objective of the study was to assess the potential effect of big five personality traits , impulsivity , sensation - seeking , and demographic data on drug consumption for different drugs , groups of drugs and for different definitions of drug users .
the study had two purposes : ( i ) to identify the association of personality profiles ( i.e. neo - ffi - r ) with drug consumption and ( ii ) to predict the risk of drug consumption for each individual according to their personality profiles .
part of the results was presented in the preprint @xcite .
the sample was created by an anonymous online survey .
it was found to be biased when compared with the general population , which was indicated from comparison to the data published by egan , et al @xcite and costa jr & mccrae @xcite .
such a bias is usual for clinical cohorts @xcite .
our study reveals that the personality profiles are strongly associated with belonging to groups of the users and non - users of the 18 drugs . for analysis , we use the following subdivision of the sample t-@xmath1 : the interval 44 - 49 indicates a moderately low score , @xmath2 , the interval 49 - 51 indicates a neutral score @xmath3 , and the interval 51 - 56 indicates a moderately high @xmath4 score .
we found that the n and o scores of drug users of all 18 drugs are moderately high @xmath4 or neutral @xmath3 , except for crack usage for the week - based classification , for which the o score is moderately low @xmath2 .
the a and c scores are moderately low @xmath2 or neutral @xmath3 for all groups of drug users and all user / non - user separations . for most groups of illicit drug users the a and c scores are moderately low @xmath2 except two exclusions :
the a score is neutral @xmath3 in the year - based classification for lsd users and in the week - based classification for lsd and magic mushrooms users .
the a and c scores for groups of legal drugs users ( i.e. alcohol , chocolate , caffeine , and nicotine ) are neutral @xmath3 , apart from nicotine users , whose c score is moderately low @xmath2 for all bases of user / non - user separation .
the impact of the e score is drug specific .
for example , for the week - based user / non - user separation the e scores are : * the e score of users is moderately low @xmath2 for amphetamines , amyl nitrite , benzodiazepines , heroin , ketamine , legal highs , methadone , and crack ; * the e score of users is moderately high @xmath4 for cocaine , ecstasy , lsd , magic mushrooms , and vsa ; * the e score of users is neutral @xmath3 for alcohol , caffeine , chocolate , cannabis , and nicotine . for more details
see section ` comparison of personality traits means for drug users and non - users ' in ` results ' .
usage of some drugs are significantly correlated .
the structure of these correlations is analysed in section ` correlation between usage of different drugs ' .
two correlation measures are utilised : the pearson correlation coefficient ( pcc ) and the relative information gain ( rig ) .
we found three groups of drugs with highly correlated use .
the central element is clearly identified for each group .
these centres are : _ heroin , ecstasy , and benzodiazepines_. it means that the drug consumption has a ` modular structure ' .
the modular structure has clear reflection in the correlation graph .
the idea to merge the correlated attributes into ` modules ' called as _ correlation pleiades _ is popular in biology @xcite .
the concept of correlation pleiades was introduced in biostatistics in 1931 @xcite .
they were used for identification of the modular structure in evolutionary physiology @xcite . according to berg @xcite ,
correlation pleiades are clusters of correlated traits . in our approach
, we distinguish the core and the peripheral elements of correlation pleiades and allow different pleiads to have small intersections in their periphery . ` soft ' clustering algorithms
relax the restriction that each data object is assigned to only one cluster ( like probabilistic @xcite or fuzzy @xcite clustering ) .
see the book of r. xu and d. wunsch @xcite for the modern review of hard and soft clustering .
we refer to @xcite for a discussion of clustering in graphs with intersections .
the three groups of correlated drugs centered around heroin , ecstasy , and benzodiazepines are defined for the decade- , year- , month- , and week - based classifications : * the heroin pleiad includes crack , cocaine , methadone , and heroin ; * the ecstasy pleiad consists of amphetamines , cannabis , cocaine , ketamine , lsd , magic mushrooms , legal highs , and ecstasy ; * the benzodiazepines pleiad contains methadone , amphetamines , cocaine , and benzodiazepines .
analysis of the intersections between correlation pleiads of drugs can generate important question and hypotheses : * why is cocaine a peripherical member of all pleiads ? * why does methadone belong to the periphery of both the heroin and benzodiazepines pleiades ? *
do these intersections reflect the structure of individual drug consumption or the structure of the groups of drug consumers ?
correlation analysis of the decade - based classification problems demonstrates that the consumption of legal drugs ( i.e. alcohol , chocolate and caffeine ) is not correlated with consumption of other drugs .
the consumptions of seven illicit drugs ( i.e. amphetamines , cannabis , cocaine , ecstasy , legal highs , lsd , and mushrooms ) are symmetrically correlated ( when the correlations are measured by relative information gain , which is not symmetric a priori ) .
there are also many strongly asymmetric correlations .
for example , knowledge of amphetamines , cocaine , ecstasy , legal highs , lsd , and magic mushroom consumption is useful for the evaluation of ketamine consumption , but on the other hand , knowledge of ketamine consumption is significantly less useful for the evaluation of usage of the drugs listed above . in this study , we evaluated the individual drug consumption risk separately , for each drug and pleiad of drugs .
we also analyzed interrelations between the individual drug consumption risks for different drugs .
we applied several data mining approaches : decision tree , random forest , @xmath0-nearest neighbors , linear discriminant analysis , gaussian mixture , probability density function estimation , logistic regression and nave bayes .
the quality of classification was surprisingly high .
we tested all the classifiers by _ leave - one - out cross validation_. the best results with sensitivity and specificity being greater than 75% were achieved for cannabis , crack , ecstasy , legal highs , lsd , and vsa .
sensitivity and specificity greater than 70% were achieved for the following drugs : amphetamines , amyl nitrite , benzodiazepines , chocolate , caffeine , heroin , ketamine , methadone and nicotine .
the poorest result was obtained for prediction of alcohol consumption .
an exhaustive search was performed to select the most effective subset of input features , and data mining methods to classify users and non - users for each drug .
users are defined for each correlation pleiad of drugs as users of any of the drug from the pleiade .
we consider the classification problem for drug pleiades for the decade- , year- , month- , and week - based user / non - user separations .
these problems are much better balanced for short periods ( the week - based user definition ) than the classification problems for separate drugs .
for example , there are 184 users for the heroin pleiad but only 29 heroin users in the database for the week - based definition of users .
the quality of classification is high .
for example , for the month - based user / non - user separation of the heroin pleiad consumption , the best classifier is a decision tree with five features and sensitivity 74.18% and specificity 74.11% .
a decision tree with seven attributes is the best classifier for the year - based classification problem of the ecstasy pleiad users / non - users and has sensitivity 80.65% and specificity 80.72% . in the week - based separation of the benzodiazepines
pleiad users / non - users , the best classifier is a decision tree with five features , sensitivity 75.10% , and specificity 75.76% . the creation of classifiers provided the capability to evaluate the risk of drug consumption in relation to individuals .
the risk map is a useful tool for data visualisation and for the generation of hypotheses for further study ( see section ` risk evaluation for the decade - based user / non - user separation ' ) .
the main results of the work are : * presentation and descriptive analysis of a database with information of 1885 respondents and usage of 18 drugs . *
demonstration that the personality traits ( five factor model , impulsivity , and sensation seeking ) together with simple demographic data give the possibility of predicting the risk of consumption of individual drugs with sensitivity and specificity above 70% for most drugs . *
the best classifiers and most significant predictors are found for each individual drug in question .
* revelation of three correlation pleiads of drugs , that are the clusters of drugs with correlated consumption centered around heroin , ecstasy , and benzodiazepines .
* the best robust classifiers and most significant predictors are found for use of pleiads of drugs . *
the risk map technology is developed for the visualization of the probability of drug consumption .
the database was collected by elaine fehrman between march 2011 and march 2012 . in january 2011
, the research proposal was approved by the university of leicester s forensic psychology ethical advisory group , and subsequently received favourable opinion from the university of leicester school of psychology s research ethics committee ( prec ) .
the data are available online @xcite .
an online survey tool from survey gizmo was employed to gather data which maximised anonymity , this being particularly relevant to canvassing respondents views , given the sensitive nature of drug use .
all participants were required to declare themselves at least 18 years of age prior to informed consent being given .
the study recruited 2051 participants over a 12-month recruitment period .
of these persons , 166 did not respond correctly to a validity check built into the middle of the scale , so were presumed to being inattentive to the questions being asked .
nine of these persons were found to also have endorsed using a fictitious recreational drug , and which was included precisely to identify respondents who over - claim , as have other studies of this kind @xcite .
this led a useable sample of 1885 participants ( male / female = 943/942 ) .
the snowball sampling methodology recruited a primarily ( 93.7% ) native english - speaking sample , with participants from the uk ( 1044 ; 55.4% ) , the usa ( 557 ; 29.5% ) , canada ( 87 ; 4.6% ) , australia ( 54 ; 2.9% ) , new zealand ( 5 ; 0.3% ) and ireland ( 20 ; 1.1% ) .
a total of 118 ( 6.3% ) came from a diversity of other countries , none of whom individually met 1% of the sample or did not declare the country of location . further optimizing anonymity , persons reported their age band , rather than their exact age ; 18 - 24 years ( 643 ; 34.1% ) , 25 - 34 years ( 481 ; 25.5% ) , 35 - 44 years ( 356 ; 18.9% ) , 45 - 54 years ( 294 ; 15.6% ) , 55 - 64 ( 93 ; 4.9% ) , and over 65 ( 18 ; 1% ) .
this indicates that although the largest age cohort band were in the 18 to 24 range , some 40% of the cohort was 35 or above , which is an age range often missed in studies of this kind .
the sample recruited was highly educated , with just under two thirds ( 59.5% ) educated to , at least , degree or professional certificate level : 14.4% ( 271 ) reported holding a professional certificate or diploma , 25.5% ( 481 ) an undergraduate degree , 15% ( 284 ) a master s degree , and 4.7% ( 89 ) a doctorate .
approximately 26.8% ( 506 ) of the sample had received some college or university tuition although they did not hold any certificates ; lastly , 13.6% ( 257 ) had left school at the age of 18 or younger .
participants were asked to indicate which racial category was broadly representative of their cultural background . an overwhelming majority ( 91.2% ; 1720 )
reported being white , ( 1.8% ; 33 ) stated they were black , and ( 1.4% ; 26 ) asian .
the remainder of the sample ( 5.6% ; 106 ) described themselves as ` other ' or ` mixed ' categories .
this small number of persons belonging to specific non - white ethnicities precludes any analyses involving racial categories . in order to assess personality traits of the sample , the revised neo five - factor inventory ( neo - ffi - r ) questionnaire
was employed @xcite .
the neo - ffi - r is a highly reliable measure of basic personality domains ; internal consistencies are 0.84 ( n ) ; 0.78 ( e ) ; 0.78 ( o ) ; 0.77 ( a ) , and 0.75 ( c ) egan @xcite .
the scale is a 60-item inventory comprised of five personality domains or factors .
the neo - ffi - r is a shortened version of the revised neo - personality inventory ( neo - pi - r ) @xcite .
the five factors are : n , e , o , a , and c with 12 items per domain .
the five traits can be summarized as : 1 .
_ neuroticism _ ( * n * ) is a long - term tendency to experience negative emotions such as nervousness , tension , anxiety and depression ; 2 .
_ extraversion _ ( * e * ) is manifested in outgoing , warm , active , assertive , talkative , cheerful , and in search of stimulation characteristics ; 3 .
_ openness to experience _
( * o * ) is a general appreciation for art , unusual ideas , and imaginative , creative , unconventional , and wide interests , 4 .
_ agreeableness _ ( * a * ) is a dimension of interpersonal relations , characterized by altruism , trust , modesty , kindness , compassion and cooperativeness ; 5 .
_ conscientiousness _ ( * c * ) is a tendency to be organized and dependable , strong - willed , persistent , reliable , and efficient .
all of these domains are hierarchically defined by specific facets @xcite .
egan et al @xcite observe that the score o and e domains of the neo - ffi instrument are less reliable than n , a , and c. participants were asked to read the 60 neo - ffi - r statements and indicate on a five - point likert scale how much a given item applied to them ( i.e. 0 = ` strongly disagree ' , 1 = ` disagree ' , 2 = ` neutral ' , 3 = ` agree ' , to 4 = ` strongly agree ' ) .
we expected that drug usage is associated with high n , and low a and c. the darker dimension of personality can be described in terms of high n and low a and c , whereas much of the anti - social behaviour in non - clinical persons appears underpinned by high n and low c @xcite .
the so - called ` _ _ negative urgency _ _ ' is the tendency to act rashly when distressed , and characterized by high n , low c , and low a @xcite .
the negative urgency is partially proved below for users of most of the illegal drugs . in addition
, our findings suggest that o is higher for drug users .
the second measure used was the barratt impulsiveness scale ( bis-11 ) @xcite .
the bis-11 is a 30-item self - report questionnaire , which measures the behavioural construct of impulsiveness , and comprises three subscales : motor impulsiveness , attentional impulsiveness , and non - planning .
the ` motor ' aspect reflects acting without thinking , the ` attentional ' component poor concentration and thought intrusions , and the ` non - planning ' a lack of consideration for consequences @xcite .
the scale s items are scored on a four - point likert scale .
this study modified the response range to make it compatible with previous related studies @xcite .
a score of five usually connotes the most impulsive response although some items are reverse - scored to prevent response bias .
items are aggregated , and the higher bis-11 scores , the higher the impulsivity level @xcite . the bis-11 is regarded a reliable psychometric instrument with good test - retest reliability ( spearman s rho is equal to 0.83 ) and internal consistency ( cronbach s alpha is equal to 0.83 ; @xcite .
the third measurement tool employed was the impulsiveness sensation - seeking ( impss ) .
although the impss combines the traits of impulsivity and sensation - seeking , it is regarded as a measure of a general sensation - seeking trait @xcite .
the scale consists of 19 statements in true - false format , comprising eight items measuring impulsivity ( * imp * ) , and 11 items gauging sensation - seeking ( * ss * ) . the impss is considered a valid and reliable measure of high risk behavioural correlates such as , substance misuse @xcite .
participants were questioned concerning their use of 18 legal and illegal drugs ( alcohol , amphetamines , amyl nitrite , benzodiazepines , cannabis , chocolate , cocaine , caffeine , crack , ecstasy , heroin , ketamine , legal highs , lsd , methadone , magic mushrooms , nicotine and volatile substance abuse ( vsa ) ) and one fictitious drug ( semeron ) which was introduced to identify over - claimers .
it was recognised at the outset of this study that drug use research regularly ( and spuriously ) dichotomises individuals as users or non - users , without due regard to their frequency or duration / desistance of drug use @xcite . in this study ,
finer distinctions concerning the measurement of drug use have been deployed , due to the potential for the existence of qualitative differences amongst individuals with varying usage levels . in relation to each drug , respondents were asked to indicate if they never used this drug , used it over a decade ago , or in the last decade , year , month , week , or day .
this format captured the breadth of a drug - using career , and the specific recency of use .
different categories of drug users are depicted in fig [ categoriesfig:1 ] .
analysis of the classes of drug users shows that part of the classes are nested : participants which belong to the category ` used in last day ' also belong to the categories ` used in last week ' , ` used in last month ' , ` used in last year ' and ` used in last decade ' .
there are two special categories : ` never used ' and ` used over a decade ago ' ( see fig [ categoriesfig:1 ] ) .
data does not contain a definition of users and non - users groups .
formally only a participant in the class ` never used ' can be called a non - user , but it is not a seminal definition because a participant who used a drug more than decade ago can not be considered a drug user for most applications .
there are several possible way to discriminate participants into groups of users and non - users for binary classification : 1 . two isolated categories ( ` never used ' and ` used over a decade ago ' ) are placed into the class of non - users with a green background in fig [ categoriesfig:1 ] and all other categories into the class ` users ' as the simplest version of binary classification .
this classification problem is called ` _ _ decade - based _ _ ' user / non - user separation .
the categories ` used in last decade ' , ` used over a decade ago ' and ` never used ' are merged to form a group of non - users and all other categories are placed into the group of users .
this classification problem is called ` _ _ year - based _ _ ' .
3 . the categories ` used in last year ' , ` used in last decade ' , ` used over a decade ago ' and ` never used ' are combined to form a group of non - users and all three other categories are placed into the group of users .
this classification problem is called ` _ _ month - based _ _ ' .
4 . the categories ` used in last week ' and ` used in last month ' are merged to form a group of users and all other categories are placed into the group of non - users .
this classification problem is called ` _ _ week - based _ _ ' .
we begin our analysis from the decade - based user / non - user separation because it is a relatively well balanced classification problem , that is , there are sufficiently many users in the united group ` used in last decade - year - month - week ' for all drugs in the database . if the problem is not directly specified then it is the decade - based classification problem .
we also perform analysis for the year- , month- , and week - based user / non - user separation .
it is useful to group drugs with highly correlated usage for this purpose ( see section ` pleiades of drugs ' ) .
the proportion of drug users among all participants is different for different drugs and for different classification problems .
the data set comprises 1885 individuals without any missing data .
table [ tab:1a ] shows the percentage of drug users for each drug and for each problem in the database .
it is necessary to mention that the sample is biased to a higher proportion of drug users .
this means that for the general population the fraction of an illegal drug users is expected to be significantly lower @xcite .
-2.25in0 in .the number and fraction of drug users [ cols="<,^,^,^,^ " , ] -2.25in0 in + heroinpl , ecstasypl , benzopl & @xmath5 & & @xmath5 & @xmath6 & @xmath6 + + heroinpl , ecstasypl , benzopl & @xmath5 & & @xmath5 & @xmath6 & @xmath6 + + heroinpl , ecstasypl & @xmath5 & & @xmath5 & @xmath6 & @xmath6 + benzopl & @xmath5 & @xmath6 & @xmath5 & @xmath6 & @xmath6 + + heroinpl , benzopl & @xmath5 & @xmath6 & @xmath5 & @xmath6 & @xmath6 + ecstasypl & @xmath5 & & @xmath5 & @xmath6 & @xmath6 + the introduction of moderate subcategories of t-@xmath7 for pleiades of drugs enables to separates the pleiades of drugs into two groups for the decade- , month- , and week - based user / non - user separation . while for year - based user / non - user separation it is only one group with the profile @xmath8 includes the users of heroinpl , ecstasypl and benzopl . for decade - based classification problem ,
the group with the profile @xmath8 includes the users of heroinpl and benzopl .
the group with the profile @xmath9 includes the users of ecstasypl .
for month- and week - based classification problem , the group with the profile @xmath10 includes the users of heroinpl and benzopl .
the group with the profile @xmath9 includes the users of ecstasypl .
the personality profile for pleiades of drugs also are strongly associated with belonging to groups of the users and non - users of each plead for decade- , year- , month- , and week - based classification problem ( see fig .
[ heroinpldecade ] , fig .
[ ecstasyplyear ] , fig .
[ benzoplmonth ] and [ heroinplweek ] ) . with respect to the sample means , scaledwidth=100.0% ] with respect to the sample means , scaledwidth=100.0% ] with respect to the sample means , scaledwidth=100.0% ] with respect to the sample means , scaledwidth=100.0% ] we applied the eight methods described in section ` risk evaluation methods ' and selected the best one for each pleiad for decade- , year- , month- , and week - based classification problem .
the results of the classifier selection are depicted in table [ tab:12a ] and the quality of classification is high .
the classification results are very satisfactory for each pleiad for decade- , year- , month , and week based problem .
we can compare the classifiers for one pleiad and for different problems ( see table [ tab:12a ] ) .
for example , * the best classifier for ecstasypl for the year - based user / non user separation is dt with seven attributes and has sensitivity 80.65% and specificity 80.72% . * the best classifier for heroinpl for the month - based user / non user separation is dt with five attributes and has sensitivity 74.18% and specificity 74.11% . *
the best classifier for benzopl for the week - based user / non user separation is dt with five attributes and has sensitivity 75.10% and specificity 75.76% .
-2.25in0 in the comparison of tables [ tab:12 ] and [ tab:12a ] shows that the best classifiers for the ecstasy and benzodiazepines pleiades are more accurate than the best classifiers for consumption of the ` central ' drugs of the pleiades , ecstasy and benzodiazepines even for the decade - based user definition .
classifiers for heroinpl may have slightly worse accuracy but these classifiers are more robust because they solve classification problems which have more balanced classes .
all other classifiers for pleiades of drugs are more robust too by the same reasons for all pleiades and definitions of users
. tables [ tab:12 ] and [ tab:12a ] for the decade - based user definition show that most of the classifiers for pleiades use more input features than the classifiers for individual drugs .
we can see from these tables that the accuracies of the classifiers for pleiades and for individual drugs do not differ drastically , but the use of a greater number of input features indicates more robust classifiers .
it is important to stress that usually pleiades are assumed to be disjoint .
we consider pleiades which are named by the central drug and have an intersection in the peripheral drugs .
for example , heroin and ecstasy pleiades have cocaine as an intersection .
this approach corresponds to the concept of ` soft clustering ' .
the successful construction of a classifier provides an instrument for the evaluation of the risk of drug consumption for each individual , along with the creation of a map of risk @xcite .
the risk map of ecstasy consumption on the basis of three input features is depicted in fig [ riskmapsfig:8 ] . from the pdfe - based risk maps ( fig [ riskmapsfig:8]a and [ riskmapsfig:8]b )
it can be observed that there is a considerable area of high risk ( indicated in blue ) for men aged between 25 - 34 years , but significantly less for females .
however , young males with the highest ss scores have significantly less risk than females with the same profiles .
dt - based risk maps ( fig [ riskmapsfig:8]c and [ riskmapsfig:8]d ) illustrate qualitatively the same shapes .
the risk maps provide a tool for the generation of hypotheses for further study .
we can create risk maps for pleiades of drugs as well .
our study demonstrates strong correlations between personality profiles and the risk of drug use .
this result supports observations from some previous works @xcite .
individuals involved in drug use are more likely to have higher scores for n , and low scores for a and c .
we analysed in detail the average differences in the groups of drug users and non - users for 18 drugs ( tables [ tab:5 ] , [ tab:5a ] , [ tab:5b ] , and [ tab:5c ] ) .
in addition to this analysis , we achieved much more detailed knowledge about the relationship between the personality traits , biographic data and the use of individual drugs or drug clusters by an individual patient . the analysed database contained 1885 participants and 12 features ( input attributes ) .
these features included five personality traits ( neo - ffi - r ) ; impulsivity ( bis-11 ) , sensation seeking ( impss ) , level of education , age , gender , country of residence , and ethnicity .
the data set included information on the consumption of 18 central nervous system psychoactive drugs : alcohol , amphetamines , amyl nitrite , benzodiazepines , cannabis , chocolate , cocaine , caffeine , crack , ecstasy , heroin , ketamine , legal highs , lsd , methadone , mushrooms , nicotine , and vsa ( output attributes ) .
there were limitations of this study since the collected sample was biased with respect to the general population , but it remained useful for risk evaluation .
we used three different techniques of feature ranking .
after input feature ranking we excluded ethnicity and country of residence .
it was impossible to completely exclude the possibility that ethnicity and country of residence may be important risk factors , but the dataset has not enough data for most of ethnicities and countries to prove the value of this information . as a result , 10 input features remained : age , edu . ,
n , e , o , a , c , imp . , ss , and gender .
our aim was to predict the risk of drug consumption for an individual .
all input features are ordinal or nominal . to apply data mining methods which were developed for continuous input features we apply catpca technique to quantify data .
we used four different definitions of drug users which differ with regard to the recency of the last drug consumption : the decade - based , year - based , month - based and week - based user / non user separation ( fig . [ categoriesfig:1 ] ) .
the day - based classification problem is also possible but there is not enough data on drug use during the last day for most drugs our findings allowed us to draw a number of important conclusions about the associations between personality traits and drug use .
all five personality factors are relevant traits to be taken into account when assessing the risk of drug consumption .
the mean scores for the groups of users of all 18 drugs are moderately high @xmath4 or neutral @xmath3 for n and o , and moderately low @xmath2 for a and c , except for crack usage for the week based classification problem which has a moderately low @xmath2 o score ( see table [ tab:5c ] and fig [ averagecrack ] ) .
users of legal drugs ( alcohol , chocolate , caffeine , and nicotine ) have neutral a and c scores @xmath3 , except nicotine users whose c score is moderately low @xmath2 . for lsd users in the year
based classification problem and for lsd and magic mushrooms users in the week based classification problem the a score is neutral @xmath3 . the impact of the e score is drug specific .
for example , for the decade - based user / non user definition the e score is negatively correlated with consumption of crack , heroin , vsa , and methadone ( e score is @xmath2 for their users ) .
it is has no predictive value for other drugs for the decade - based classification ( e score for users is @xmath3 ) , whereas in the year- , month- , and week - based classification problems all three possible values of e score are observed ( see tables [ tab:5 ] , [ tab:5a ] , [ tab:5b ] and [ tab:5c ] ) .
we confirm the previous researcher finding that the higher scores for n and o the lower scores for c and a lead to increased risk of drug use @xcite .
o score is marked by curiosity and open - mindedness ( and correlated with intelligence ) , and it is therefore understandable why higher o may be sometimes associated with drug use @xcite .
flory et al @xcite found that marijuana use to be associated with lower a and c , and higher o. these findings have been confirmed by our study .
our results improve the knowledge concerning the pathways leading to drug consumption .
it is known that significant predictors of alcohol , tobacco and marijuana use may vary according to the drug in question @xcite .
our study demonstrated that different attributes were important for different drugs .
we tested eight types of classifiers for each drug for the decade - based user definition .
loocv was used to evaluate sensitivity and specificity . in this study
we choose the method which provide maximal value of minimum among sensitivity and specificity as the best one .
the method with maximal sum of the sensitivity and specificity is selected as the best one for two methods with the same minimum among sensitivity and specificity .
there were classifiers with sensitivity and specificity greater than 70% for the decade - based user / non user separation for all drugs except magic mushrooms , alcohol , and cocaine ( table [ tab:12 ] ) .
this accuracy was unexpectedly high for this type of problem .
the poorest result was obtained for the prediction of alcohol consumption .
the best set of input features was defined for each drug ( table [ tab:12 ] ) .
an exhaustive search was performed to select the most effective subset of input features , and the best data mining methods to classify users and non - users for each drug .
there were 10 input features .
each of them is an important factor for risk evaluation for the use of some drugs .
however , there was no single most effective classifier using all input features .
the maximal number of attributes used in the best classifiers is six ( out of 10 ) and the minimal number is two . table [ tab:12 ] shows the best sets of attributes for user / nonuser classification for different drugs and for the decade - based classification problem .
this table together with its analogues for pleiads of drugs and all decade - year - month - week classification problems ( table [ tab:12a ] ) are important result of the analysis .
the dt for crack consumption used two features e and c only , and provided sensitivity of 80.63% and specificity of 78.57% .
the dt for vsa consumption used age , edu . ,
e , a , c , and ss , and provided sensitivity 83.48% and specificity 77.64% ( table [ tab:12 ] ) .
age was a widely used feature which was employed in the best classifiers for 14 drugs for the decade - based classification problem .
gender was used in the best methods for 10 drugs .
we found some unexpected outcomes .
for example , fraction of females which are alcohol users is greater than the fraction of males but the greater part of males consume caffeine ( coffee ) .
most of the features which are not used in the best classifiers are redundant but are not uninformative .
for example , the best classifier for ecstasy consumption used age , ss , and gender and had sensitivity 76.17% and specificity 77.16% .
there exist a dt for prediction of usage of the same drug , which utilizes age , edu .
, o , c , and ss with sensitivity 77.23% and specificity 75.22% , a dt with inputs age , edu . ,
e , o , and a with sensitivity 73.24% and specificity 78.22% , and an advanced @xmath0nn classifier with inputs age , edu .
, n , e , o , c , imp . , ss , and gender with sensitivity 75.63% and specificity 75.75% .
this means that for evaluating the risk of ecstasy usage all input attributes are informative but the required information can be extracted from a subset of attributes .
we demonstrated that there are three groups of drugs with strongly correlated consumption .
that is , the drug usage has a ` modular structure ' .
the idea to merge correlated attributes into ` modules ' is popular in biology .
the modules are called the ` correlation pleiades ' @xcite ( see section ` pleiades of drugs ' ) .
the modular structure contains three modules : the heroin pleiad , ecstasy pleiad , and benzodiazepines pleiad : * the _ heroin pleiad ( heroinpl ) _ includes crack , cocaine , methadone , and heroin . *
the _ ecstasy pleiad ( ecstasypl ) _ includes amphetamines , cannabis , cocaine , ketamine , lsd , magic mushrooms , legal highs , and ecstasy . * the _ benzodiazepines pleiad ( benzopl ) _ includes contains methadone , amphetamines , and cocaine .
the modular structure has a clear reflection in the correlation graph , fig [ strongdrugusfig:5 ] .
we define groups of users and non - users for each pleiad . in
most of the databases the classes of users and non - users for most of the individual drugs are imbalanced ( see table [ tab:1 ] ) but the merging the users of all drugs in one class ` drug users ' does not seem to be the best solution because of physiological , psychological and cultural differences between usage of different drugs .
we propose instead to use correlation pleiades for the analysis of drug usage as a solution to the class imbalance problem because for all three pleiades the classes of users and non - users are better balanced ( table [ tab:13a ] ) and the consumption of different drugs from the same pleiad is correlated .
we applied the eight methods described in the ` risk evaluation methods ' section and selected the best one for each problem for all pleiades .
the results of the classifier selection are presented in table [ tab:12a ] and the quality of the classification is high .
the majority of the best classifiers for pleiades of drugs has a better accuracy than the classifiers for individual drug usage .
( see table [ tab:12 ] and [ tab:12a ] ) .
the best classifiers for pleiades of drugs use more input features than the best classifiers for the corresponding individual drugs .
the classification problems for pleiades of drugs are more balanced .
therefore , we can expect that the classifiers for pleiades are more robust than the classifiers for individual drugs .
the user / non - user classifiers can be also used for forming of risk maps .
risk maps are useful tools for data visualisation and hypotheses generation .
these results are important as they examine the question of the relationship between drug use and personality comprehensively and engage the challenge of untangling correlated personality traits ( the ffm , impulsivity , and sensation seeking @xcite ) , and clusters of substance misuse ( the correlation pleiades ) .
the work acknowledged the breadth of a common behaviour which may be transient and leave no impact , or may significantly harm an individual .
we examined drug use behaviour comprehensively in terms of the many kinds of substances that may be used ( from the legal and anodyne , to the deeply harmful ) , as well as the possibility of behavioural over - claiming .
we built into our study the wide temporality of the behaviour indicative of the chronicity of behaviour and trends and fashions ( e.g. the greater use of lsd in the 1960s and 1970s , the rise of ecstasy in the 1980s , some persons being one - off experimenters with recreational drugs , and others using recreational substances on a daily basis ) .
we defined substance use in terms of behaviour rather than legality , as legislation in the field is variable .
our data were gathered before ` legal highs ' emerged as a health concern @xcite so we did not differentiate , for example , synthetic cannabinoids and cathinone - based stimulants ; these substances have been since widely made illegal .
we were nevertheless able to accurately classify users of these substances ( reciprocally , our data were gathered before cannabis decriminalisation in parts of north america , but again , we were able to accurately classify cannabis users ) .
we included control participants who had never used these substances , those who had used them in the distant past , up to and including persons who had used the drug in the past day , avoiding the procrustean data - gathering and classifying methods which may occlude an accurate picture of drug use behaviour and risk @xcite . such rich data and the complex methods used for analysis necessitated a large and substantial sample
. the study was atheoretical regarding the morality of the behaviour , and did not medicalise or pathologise participants , optimising engagement by persons with heterogeneous drug - use histories .
our study used a rigorous range of data - mining methods beyond those typically used in studies examining the association of drug use and personality in the psychological and psychiatric literature , revealing that decision tree methods were most commonly effective for classifying drug users .
we found that high n , low a , and low c are the most common personality correlates of drug use , these traits being sometimes seen in combination as an indication of higher - order stability and behavioural conformity , and , inverted , are associated with externalisation of distress @xcite .
low stability is also a marker of negative urgency @xcite whereby persons act rashly when distressed .
our research points to the importance of individuals acquiring emotional self - management skills anteceding distress as a means to reduce self - medicating drug - using behaviour , and the risk to health that injudicious or chronic drug use may cause .
[ [ s1_appendix ] ] s1 appendix .
+ + + + + + + + + + + + * mean for groups of users and non - users . * in this section we present mean t-@xmath7 for groups of users and non - users for decade , year , month , and week based user definitions respectively .
column @xmath11-value assesses the significance of differences of mean scores for groups of users and non - users : it is the probability of observing by chance the same or greater differences for mean scores if both groups have the same mean .
rows ` # ' contain number of users and non - users for the drugs .
+ & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score & 95% ci for mean & + + & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score&95% ci for mean & + + + # & & & + n & 50.13 & 49.67 , 50.59 & 48.19 & 45.77 , 50.61 & 0.116 + e & 50.06 & 49.60 , 50.52 & 50.04 & 47.61 , 52.42 & 0.988 + o & 50.04 & 49.58 , 50.51 & 48.81 & 46.45 , 51.17 & 0.318 + a & 49.93 & 49.47 , 50.39 & 52.51 & 50.26 , 54.77 & 0.036 + c & 49.94 & 49.48 , 50.40 & 53.31 & 51.05 , 55.56 & 0.006 + + # & & & + n & 51.71 & 50.95 , 52.46 & 49.14 & 48.58 , 49.69 & 0.001 + e & 49.71 & 48.89,50.53 & 50.26 & 49.72 , 50.80 & 0.251 + o & 53.05 & 52.34 , 53.77 & 48.28 & 47.72 , 48.84 & 0.001 + a & 48.39 & 47.60 , 49.18 & 50.94 & 50.39 , 51.48 & 0.001 + c & 47.04 & 46.29 , 47.80 & 51.76 & 51.22 , 52.30 & 0.001 +
n & 50.78 & 49.78 , 51.79 & 49.89 & 49.37 , 50.39 & 0.122 + e & 50.97 & 49.95 , 51.99 & 49.84 & 49.33 , 50.35 & 0.052 + o & 51.45 & 50.47 , 52.43 & 49.65 & 49.14 , 50.15 & 0.002 + a & 48.69 & 47.65 , 49.72 & 50.35 & 49.84 , 50.85 & 0.004
+ c & 48.08 & 46.10 , 49.07 & 50.54 & 50.04 , 51.05 & 0.001 + n & 52.83 & 52.12 , 53.54 & 48.15 & 47.59 , 48.72 & 0.001 + e & 49.07 & 48.31 , 49.83 & 50.74 & 50.19 , 51.30 & 0.001 + o & 52.66 & 51.97 , 53.34 & 48.17 & 47.59 , 48.75 & 0.001 + a & 48.28 & 47.53 , 49.02 & 51.22 & 50.67 , 51.78 & 0.001 + c & 47.70 & 46.98 , 48.41 & 51.69 & 51.12 , 52.25 & 0.001 + n & 51.08 & 50.52 , 51.65 & 47.98 & 47.25 , 48.71 & 0.001 + e & 49.75 & 49.17 , 50.33 & 50.70 & 49.98 , 51.41 & 0.053 + o & 52.48 & 51.96 , 52.99 & 44.95 & 44.20,45.70 & 0.001 + a & 48.84 & 48.28,49.40 & 52.42 & 51.69,53.15 & 0.001 + c & 48.15 & 47.60,48.70 & 53.92 & 53.27,54.65 & 0.001 +
n & 50.06 & 49.60 , 50.51 & 50.29 & 46.36 , 54.21 & 0.894 + e & 50.05 & 49.59 , 50.51 & 50.80 & 47.71 , 53.89 & 0.660 + o & 50.05 & 49.59 , 50.51 & 47.37 & 44.42 , 50.32 & 0.117 + a & 50.05 & 49.59 , 50.50 & 48.66 & 44.61 , 52.70 & 0.416
+ c & 50.03 & 49.58 , 50.49 & 51.57 & 47.87 , 55.27 & 0.366 +
n & 51.85 & 51.09 , 52.60 & 49.04 & 48.48 , 49.59 & 0.001 + e & 50.33 & 49.55 , 51.12 & 49.91 & 49.35 , 50.46 & 0.374 + o & 52.60 & 51.89 , 53.30 & 48.51 & 47.94 , 49.08 & 0.001 + a & 47.71 & 46.93 , 48.50 & 51.34 & 50.81,51.88 & 0.001 + c & 47.49 & 46.76 , 48.22 & 51.53 & 50.98 , 52.09 & 0.001 +
n & 50.06 & 49.61 , 50.52 & 50.08 & 46.59,53.57 & 0.991 + e & 50.13 & 49.67 , 50.59 & 46.76 & 43.82 , 49.69 & 0.043 + o & 50.11 & 49.65 , 50.57 & 44.59 & 41.67,47.52 & 0.001 + a & 49.99 & 49.54 , 50.45 & 51.11 & 47.71 , 54.51 & 0.504 + c & 49.99 & 49.53 , 50.44 & 53.59 & 50.79 , 56.40 & 0.029 +
n & 53.08 & 51.64 , 54.52 & 49.72 & 49.25 , 50.20 & 0.001 + e & 48.80 & 47.33 , 50.27 & 50.20 & 49.73 , 50.68 & 0.066 + o & 52.91 & 51.60 , 54.21 & 49.67 & 49.19 , 50.15 & 0.001 + a & 47.02 & 45.46 , 48.58 & 50.36 & 49.89 , 50.83 & 0.001 + c & 46.19 & 44.70 , 47.68 & 50.50 & 50.03 , 50.96 & 0.001 +
n & 51.30 & 50.58 , 52.02 & 49.24 & 48.67 , 49.82 & 0.001 + e & 50.56 & 49.80 , 51.31 & 49.73 & 49.17 , 50.30 & 0.081 + o & 53.62 & 52.96 , 54.28 & 47.60 & 47.03 , 48.17 & 0.001 + a & 48.49 & 47.75 , 49.23 & 51.03 & 50.47 , 51.60 & 0.001 + c & 47.30 & 46.59 , 48.00 & 51.89 & 51.33 , 52.45 & 0.001 +
n & 54.60 & 53.29 , 55.92 & 49.49 & 49.01 , 49.96 & 0.001 + e & 48.42 & 46.94 , 49.90 & 50.27 & 49.80 , 50.74 & 0.011 + o & 54.25 & 53.04 , 55.47 & 49.46 & 48.98 , 49.94 & 0.001 + a & 45.53 & 44.00 , 47.06 & 50.59 & 50.12 , 51.05 & 0.001 + c & 45.91 & 44.55 , 47.26 & 50.59 & 50.11 , 51.06 & 0.001 + n & 51.42 & 50.40 , 52.43 & 49.75 & 49.25 , 50.26 & 0.005 + e & 50.36 & 49.23 , 51.48 & 49.99 & 49.50 , 50.49 & 0.542 + o & 53.87 & 52.90 , 54.84 & 49.12 & 48.62 , 49.62 & 0.001 + a & 47.80 & 46.67 , 48.94 & 50.53 & 50.04 , 51.01 & 0.001 + c & 46.86 & 45.81 , 47.92 & 50.79 & 50.30 , 51.28 & 0.001 +
n & 51.49 & 50.77 , 52.22 & 49.09 & 48.52 , 49.66 & 0.001 + e & 49.71 & 48.94 , 50.47 & 50.30 & 49.75 , 50.86 & 0.206 + o & 54.30 & 53.67 , 54.92 & 47.08 & 46.51 , 47.65 & 0.001 + a & 48.61 & 47.86 , 49.36 & 50.98 & 50.42 , 51.53 & 0.001 + c & 46.98 & 46.27 , 47.72 & 52.14 & 51.60 , 52.68 & 0.001 +
n & 50.87 & 50.05 , 51.69 & 49.72 & 49.18 , 50.26 & 0.023 + e & 50.07 & 49.16 , 50.98 & 50.06 & 49.54 , 50.58 & 0.986 + o & 55.25 & 54.54 , 55.96 & 47.80 & 47.27 , 48.33 & 0.001 + a & 48.44 & 47.57 , 49.31 & 50.68 & 50.16 , 51.21 & 0.001
+ c & 47.54 & 46.71 , 48.36 & 51.12 & 50.59 , 51.65 & 0.001 +
n & 53.41 & 52.47 , 54.35 & 49.11 & 48.61 , 49.62 & 0.001 + e & 47.97 & 46.88 , 49.05 & 50.66 & 50.17 , 51.15 & 0.001 + o & 53.77 & 52.86 , 54.69 & 48.93 & 48.42 , 49.44 & 0.001 + a & 47.07 & 46.03 , 48.10 & 50.86 & 50.37 , 51.35 & 0.001 + c & 46.21 & 45.22 , 47.20 & 51.15 & 50.66 , 51.64 & 0.001 + n & 50.73 & 49.98,51.47 & 49.67 & 49.10 , 50.24 & 0.027 + e & 50.15 & 49.35,50.96 & 50.01 & 49.47 , 50.55 & 0.765 + o & 54.36 & 53.71,55.01 & 47.46 & 46.90 , 48.02 & 0.001 + a & 48.55 & 47.78,49.32 & 50.88 & 50.33 , 51.43 & 0.001
+ c & 47.54 & 46.81,48.27 & 51.53 & 50.97 , 52.08 & 0.001 +
n & 50.97 & 50.41 , 51.52 & 48.22 & 47.45 , 48.99 & 0.001 + e & 49.98 & 49.41 , 50.54 & 50.24 & 49.48 , 50.99 & 0.599 + o & 51.47 & 50.92 , 52.01 & 47.01 & 46.26 , 47.77 & 0.001 + a & 49.20 & 48.65 , 49.75 & 51.69 & 50.92 , 52.46 & 0.001 + c & 48.64 & 48.08 , 49.20 & 52.95 & 52.24 , 53.67 & 0.001 +
n & 52.88 & 51.57 , 54.20 & 49.67 & 49.19,50.15 & 0.001 + e & 48.96 & 47.45 , 50.47 & 50.22 & 49.74,50.69 & 0.075 + o & 54.20 & 53.00 , 55.41 & 49.42 & 48.93,49.90 & 0.001 + a & 47.30 & 45.92 , 48.68 & 50.40 & 49.92,50.87 & 0.001 + c & 45.22 & 43.88 , 46.56 & 50.73 & 50.26,51.20 & 0.001 + + & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score & 95% ci for mean & + + & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score&95% ci for mean & + + + # & & & + n & 49.96 & 49.49 , 50.43 & 50.55 & 48.87 , 52.24 & 0.5012 + e & 50.12 & 49.64 , 50.59 & 48.52 & 46.89 , 50.14 & 0.0642 + o & 50.07 & 49.60 , 50.54 & 49.11 & 47.46 , 50.76 & 0.2712 + a & 49.95 & 49.48 , 50.42 & 50.70 & 49.03 , 52.37 & 0.3921 + c & 49.94 & 49.47 , 50.41 & 50.73 & 49.03 , 52.44 & 0.3782 + + # & & & + n & 52.40 & 51.43 , 53.36 & 49.28 & 48.77 , 49.78 & 0.0001 + e & 49.48 & 48.45 , 50.50 & 50.16 & 49.66 , 50.66 & 0.2431 + o & 53.83 & 52.92 , 54.74 & 48.85 & 48.34 , 49.35 & 0.0001 + a & 47.47 & 46.45 , 48.48 & 50.76 & 50.27 , 51.26 & 0.0001
+ c & 45.91 & 44.94 , 46.88 & 51.23 & 50.74 , 51.72 & 0.0001 +
n & 51.57 & 49.85 , 53.29 & 49.88 & 49.41 , 50.35 & 0.0633 + e & 50.25 & 48.44 , 52.06 & 49.98 & 49.51 , 50.45 & 0.7777 + o & 51.97 & 50.36 , 53.59 & 49.85 & 49.38 , 50.32 & 0.0136 + a & 46.35 & 44.64 , 48.07 & 50.28 & 49.81 , 50.74 & 0.0001 + c & 46.92 & 45.25 , 48.60 & 50.23 & 49.77 , 50.70 & 0.0002 +
n & 53.69 & 52.84 , 54.54 & 48.54 & 48.03 , 49.05 & 0.0001 + e & 48.91 & 47.97 , 49.85 & 50.43 & 49.92 , 50.94 & 0.0053 + o & 53.00 & 52.18 , 53.83 & 48.81 & 48.28 , 49.34 & 0.0001 + a & 47.55 & 46.64 , 48.46 & 50.97 & 50.46 , 51.48 & 0.0001 + c & 47.06 & 46.19 , 47.94 & 51.16 & 50.65 , 51.68 & 0.0001 +
n & 51.10 & 50.46 , 51.74 & 48.76 & 48.14 , 49.38 & 0.0001 + e & 49.73 & 49.08 , 50.39 & 50.30 & 49.69 , 50.91 & 0.2177 + o & 53.68 & 53.13 , 54.23 & 45.85 & 45.22 , 46.48 & 0.0001 + a & 48.77 & 48.13 , 49.40 & 51.39 & 50.76 , 52.02 & 0.0001
+ c & 47.32 & 46.70 , 47.95 & 53.02 & 52.42 , 53.62 & 0.0001 +
n & 50.01 & 49.56 , 50.47 & 49.45 & 46.04 , 52.86 & 0.7445 + e & 49.99 & 49.53 , 50.45 & 50.49 & 47.75 , 53.24 & 0.7156 + o & 50.04 & 49.58 , 50.50 & 48.27 & 45.59 , 50.94 & 0.1945 + a & 50.02 & 49.56 , 50.48 & 49.17 & 45.72 , 52.62 &
0.6259 + c & 49.97 & 49.51 , 50.42 & 51.41 & 47.81 , 55.01 & 0.4254 +
n & 52.16 & 51.19 , 53.13 & 49.39 & 48.88 , 49.89 & 0.0001 + e & 50.93 & 49.88 , 51.98 & 49.74 & 49.24 , 50.23 & 0.0439 + o & 52.79 & 51.87 , 53.71 & 49.21 & 48.70 , 49.72 & 0.0001 + a & 46.84 & 45.81 , 47.86 & 50.90 & 50.41 , 51.39 & 0.0001 + c & 46.81 & 45.86 , 47.75 & 50.91 & 50.40 , 51.41 & 0.0001 +
n & 49.99 & 49.53 , 50.45 & 50.36 & 47.75 , 52.96 & 0.7827 + e & 50.10 & 49.64 , 50.56 & 46.92 & 44.34 , 49.50 & 0.0181 + o & 50.10 & 49.64 , 50.56 & 47.07 & 44.45 , 49.69 & 0.0265 + a & 49.96 & 49.50 , 50.42 & 51.15 & 48.61 , 53.69 & 0.3605 + c & 49.91 & 49.45 , 50.37 & 52.79 & 50.14 , 55.43 & 0.0362 +
n & 54.06 & 51.78 , 56.35 & 49.82 & 49.36 , 50.28 & 0.0005 + e & 49.24 & 47.20 , 51.28 & 50.03 & 49.57 , 50.50 & 0.4548 + o & 52.90 & 50.72 , 55.08 & 49.87 & 49.41 , 50.33 & 0.0084 + a & 46.65 & 44.17 , 49.12 & 50.15 & 49.69 , 50.61 & 0.0070 + c & 45.08 & 42.92 , 47.25 & 50.22 & 49.76 , 50.67 & 0.0001 +
n & 50.71 & 49.82 , 51.61 & 49.73 & 49.21 , 50.25 & 0.0624 + e & 51.51 & 50.60 , 52.42 & 49.43 & 48.91 , 49.95 & 0.0001 + o & 54.10 & 53.32 , 54.89 & 48.45 & 47.92 , 48.97 & 0.0001 + a & 48.55 & 47.64 , 49.45 & 50.55 & 50.03 , 51.07 & 0.0002 + c & 47.02 & 46.17 , 47.87 & 51.13 & 50.61 , 51.65 & 0.0001 +
n & 55.37 & 53.63 , 57.12 & 49.64 & 49.18 , 50.10 & 0.0001 + e & 47.68 & 45.68 , 49.69 & 50.15 & 49.69 , 50.62 & 0.0192 + o & 53.45 & 51.80 , 55.11 & 49.77 & 49.30 , 50.24 & 0.0001 + a & 44.37 & 42.25 , 46.50 & 50.38 & 49.92 , 50.83 & 0.0001 + c & 45.45 & 43.52 , 47.39 & 50.30 & 49.84 , 50.77 & 0.0001 +
n & 51.52 & 50.20 , 52.84 & 49.81 & 49.33 , 50.29 & 0.0175 + e & 51.02 & 49.48 , 52.57 & 49.87 & 49.40 , 50.34 & 0.1619 + o & 54.18 & 52.88 , 55.49 & 49.48 & 49.00 , 49.96 & 0.0001
+ a & 47.90 & 46.42 , 49.38 & 50.26 & 49.79 , 50.73 & 0.0030 + c & 46.34 & 44.91 , 47.76 & 50.45 & 49.98 , 50.93 & 0.0001 +
n & 51.27 & 50.41 , 52.12 & 49.46 & 48.93 , 49.99 & 0.0004 + e & 50.02 & 49.11 , 50.92 & 49.99 & 49.48 , 50.51 & 0.9659 + o & 54.49 & 53.76 , 55.22 & 48.08 & 47.55 , 48.62 & 0.0001 + a & 48.09 & 47.23 , 48.96 & 50.81 & 50.29 , 51.34 & 0.0001 + c & 46.56 & 45.71 , 47.41 & 51.47 & 50.95 , 51.98 & 0.0001 + n & 49.98 & 48.97 , 50.99 & 50.00 & 49.50 , 50.51 & 0.9691 + e & 50.72 & 49.63 , 51.81 & 49.82 & 49.32 , 50.31 & 0.1403 + o & 56.29 & 55.48 , 57.11 & 48.41 & 47.91 , 48.91 & 0.0001 + a & 49.05 & 47.98 , 50.12 & 50.24 & 49.74 , 50.74 & 0.0484 + c & 47.74 & 46.71 , 48.77 & 50.57 & 50.07 , 51.07 & 0.0001 +
n & 53.74 & 52.65 , 54.84 & 49.23 & 48.75 , 49.72 & 0.0001 + e & 47.75 & 46.46 , 49.03 & 50.46 & 49.99 , 50.94 & 0.0001 + o & 53.81 & 52.76 , 54.86 & 49.22 & 48.73 , 49.71 & 0.0001 + a & 46.53 & 45.33 , 47.73 & 50.71 & 50.23 , 51.19 & 0.0001 + c & 46.01 & 44.87 , 47.15 & 50.82 & 50.33 , 51.30 & 0.0001 +
n & 50.33 & 49.40 , 51.26 & 49.90 & 49.38 , 50.42 & 0.4311 + e & 50.71 & 49.66 , 51.76 & 49.79 & 49.29 , 50.28 & 0.1179 + o & 55.72 & 54.96 , 56.49 & 48.29 & 47.78 , 48.80 & 0.0001 + a & 48.51 & 47.48 , 49.53 & 50.45 & 49.95 , 50.95 & 0.0009 + c & 47.30 & 46.36 , 48.24 & 50.81 & 50.30 , 51.32 & 0.0001 +
n & 51.16 & 50.55 , 51.76 & 48.51 & 47.85 , 49.18 & 0.0001 + e & 49.80 & 49.17 , 50.42 & 50.26 & 49.61 , 50.91 & 0.3094 + o & 51.91 & 51.31 , 52.51 & 47.55 & 46.89 , 48.20 & 0.0001 + a & 49.04 & 48.42 , 49.65 & 51.24 & 50.58 , 51.90 & 0.0001 + c & 47.98 & 47.38 , 48.57 & 52.60 & 51.95 , 53.25 & 0.0001 + n & 53.81 & 51.67 , 55.95 & 49.80 & 49.34 , 50.26 & 0.0004 + e & 49.55 & 47.32 , 51.79 & 50.02 & 49.56 , 50.49 & 0.6832 + o & 53.59 & 51.58 , 55.60 & 49.81 & 49.35 , 50.27 & 0.0004 + a & 47.48 & 45.35 , 49.61 & 50.13 & 49.67 , 50.60 & 0.0174 + c & 45.31 & 43.18 , 47.44 & 50.25 & 49.79 , 50.71 &
0.0001 + + & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score & 95% ci for mean & + + & & & * @xmath11-value * + & mean t - score & 95% ci for mean & mean t - score&95% ci for mean & + + + # & & & + n & 49.85 & 49.35 , 50.35 & 50.69 & 49.61 , 51.78 & 0.1669 + e & 50.62 & 50.12 , 51.11 & 47.14 & 46.11 , 48.17 & 0.0001 + o & 50.22 & 49.73 , 50.72 & 48.96 & 47.87 , 50.06 & 0.0401 + a & 50.10 & 49.60 , 50.60 & 49.55 & 48.50 , 50.60 & 0.3575 + c & 50.15 & 49.66 , 50.65 & 49.28 & 48.22 , 50.35 & 0.1460 +
n & 52.78 & 51.49 , 54.07 & 49.60 & 49.12 , 50.08 & 0.0001 + e & 49.07 & 47.63 , 50.51 & 50.13 & 49.66 , 50.61 & 0.1665 + o & 52.94 & 51.69 , 54.20 & 49.57 & 49.09 , 50.06 & 0.0001 + a & 46.57 & 45.21 , 47.92 & 50.50 & 50.02 , 50.97 & 0.0001 + c & 45.06 & 43.75 , 46.37 & 50.71 & 50.24 , 51.19 & 0.0001 + n & 49.36 & 46.12 , 52.61 & 50.01 & 49.56 , 50.47 & 0.6911 + e & 49.83 & 46.38 , 53.28 & 50.00 & 49.55 , 50.46 & 0.9218 + o & 50.40 & 47.06 , 53.74 & 49.99 & 49.53 , 50.45 & 0.8083 + a & 45.43 & 41.87 , 49.00 & 50.10 & 49.65 , 50.56 & 0.0122 + c & 47.31 & 44.60 , 50.01 & 50.06 & 49.60 , 50.52 & 0.0490 +
n & 55.26 & 54.11 , 56.41 & 49.01 & 48.53 , 49.48 & 0.0001 + e & 48.06 & 46.79 , 49.33 & 50.37 & 49.89 , 50.84 & 0.0010 + o & 52.68 & 51.59 , 53.77 & 49.49 & 49.00 , 49.99 & 0.0001 + a & 46.72 & 45.47 , 47.97 & 50.62 & 50.14 , 51.10 & 0.0001 + c & 46.54 & 45.40 , 47.69 & 50.65 & 50.17 , 51.14 & 0.0001 +
n & 50.72 & 49.99 , 51.45 & 49.48 & 48.91 , 50.06 & 0.7142 + e & 50.06 & 49.31 , 50.80 & 49.96 & 49.40 , 50.52 & 0.5738 + o & 54.34 & 53.74 , 54.95 & 46.88 & 46.30 , 47.46 & 0.3233 + a & 48.66 & 47.93 , 49.38 & 50.97 & 50.39 , 51.54 & 0.5477 + c & 47.26 & 46.55 , 47.97 & 51.97 & 51.41 , 52.53 & 0.3800 +
n & 49.98 & 49.52 , 50.44 & 50.40 & 48.19 , 52.60 & 0.0002 + e & 50.03 & 49.57 , 50.49 & 49.43 & 47.38 , 51.49 & 0.1295 + o & 50.05 & 49.58 , 50.52 & 49.13 & 47.36 , 50.90 & 0.0008 + a & 50.04 & 49.57 , 50.50 & 49.36 & 47.19 , 51.53 & 0.0001 + c & 49.95 & 49.49 , 50.41 & 50.90 & 48.82 , 52.97 & 0.0002 + + # & & & + n & 52.87 & 51.29 , 54.46 & 49.74 & 49.27 , 50.21 & 0.3787 + e & 51.34 & 49.50 , 53.18 & 49.88 & 49.41 , 50.34 & 0.0078 + o & 52.53 & 51.00 , 54.06 & 49.77 & 49.29 , 50.24 & 0.0560 + a & 45.75 & 44.05 , 47.46 & 50.39 & 49.93 , 50.86 & 0.7537 + c & 47.23 & 45.71 , 48.75 & 50.25 & 49.78 , 50.73 & 0.4728 +
n & 50.05 & 49.59 , 50.52 & 49.21 & 47.37 , 51.05 & 0.0092 + e & 50.16 & 49.69 , 50.63 & 47.68 & 45.93 , 49.43 & 0.8390 + o & 50.12 & 49.65 , 50.58 & 48.29 & 46.47 , 50.11 & 0.0001 + a & 49.98 & 49.51 , 50.45 & 50.27 & 48.50 , 52.04 & 0.0001 + c & 49.95 & 49.49 , 50.42 & 50.67 & 48.77 , 52.56 & 0.0001 + n & 57.86 & 52.26 , 63.46 & 49.92 & 49.46 , 50.37 & 0.1931 + e & 45.97 & 41.30 , 50.64 & 50.04 & 49.59 , 50.50 & 0.0852 + o & 50.89 & 47.21 , 54.57 & 49.99 & 49.54 , 50.45 & 0.6165 + a & 42.98 & 36.34 , 49.62 & 50.08 & 49.62 , 50.53 & 0.0380 + c & 45.14 & 40.37 , 49.91 & 50.05 & 49.60 , 50.51 & 0.0448 +
n & 49.53 & 48.18 , 50.89 & 50.07 & 49.59 , 50.55 & 0.4649 + e & 52.24 & 50.83 , 53.65 & 49.67 & 49.20 , 50.15 & 0.0008 + o & 54.41 & 53.23 , 55.59 & 49.36 & 48.88 , 49.84 & 0.0001 + a & 48.10 & 46.75 , 49.45 & 50.28 & 49.80 , 50.76 & 0.0029 + c & 47.27 & 45.96 , 48.59 & 50.40 & 49.92 , 50.88 & 0.0001 + n & 56.70 & 54.05 , 59.34 & 49.81 & 49.35 , 50.26 & 0.0001 + e & 45.58 & 42.06 , 49.10 & 50.13 & 49.67 , 50.58 & 0.0130 + o & 52.48 & 49.63 , 55.34 & 49.93 & 49.47 , 50.39 & 0.0823 + a & 42.18 & 39.00 , 45.35 & 50.23 & 49.77 , 50.68 & 0.0001 + c & 43.36 & 40.35 , 46.37 & 50.19 & 49.74 , 50.65 & 0.0001 + + # & & & + n & 51.29 & 49.14 , 53.45 & 49.94 & 49.48 , 50.41 & 0.2270 + e & 49.62 & 46.74 , 52.49 & 50.02 & 49.56 , 50.47 & 0.7851 + o & 54.79 & 52.59 , 56.98 & 49.79 & 49.33 , 50.25 & 0.0001 + a & 46.90 & 44.15 , 49.66 & 50.14 & 49.68 , 50.59 & 0.0237 + c & 45.03 & 42.50 , 47.56 & 50.22 & 49.76 , 50.67 & 0.0001 +
n & 52.02 & 50.68 , 53.36 & 49.70 & 49.23 , 50.18 & 0.0015 + e & 49.10 & 47.59 , 50.61 & 50.13 & 49.66 , 50.60 & 0.2002 + o & 54.37 & 53.22 , 55.53 & 49.36 & 48.88 , 49.84 &
0.0001 + a & 46.83 & 45.50 , 48.16 & 50.46 & 49.99 , 50.94 & 0.0001 + c & 45.30 & 44.01 , 46.60 & 50.69 & 50.21 , 51.16 & 0.0001 + n & 50.55 & 48.97 , 52.12 & 49.95 & 49.48 , 50.42 & 0.4717 + e & 51.28 & 49.53 , 53.04 & 49.88 & 49.41 , 50.34 & 0.1279 + o & 57.28 & 56.16 , 58.41 & 49.30 & 48.83 , 49.77 & 0.0001 + a & 48.92 & 47.35 , 50.48 & 50.10 & 49.63 , 50.58 & 0.1533 + c & 47.10 & 45.59 , 48.60 & 50.28 & 49.81 , 50.75 & 0.0001 + n & 54.53 & 53.00 , 56.06 & 49.55 & 49.08 , 50.02 & 0.0001 + e & 46.86 & 45.07 , 48.65 & 50.31 & 49.85 , 50.78 & 0.0003 + o & 52.89 & 51.37 , 54.40 & 49.71 & 49.24 , 50.18 & 0.0001 + a & 46.18 & 44.50 , 47.87 & 50.38 & 49.92 , 50.85 & 0.0001 + c & 45.44 & 43.83 , 47.05 & 50.45 & 49.99 , 50.92 & 0.0001 + + # & & & + n & 49.91 & 48.41 , 51.42 & 50.01 & 49.53 , 50.48 & 0.9064 + e & 50.31 & 48.50 , 52.12 & 49.97 & 49.51 , 50.44 & 0.7205 + o & 56.92 & 55.77 , 58.07 & 49.36 & 48.89 , 49.83 & 0.0001 + a & 48.57 & 46.94 , 50.19 & 50.13 & 49.66 , 50.60 & 0.0699 + c & 46.85 & 45.25 , 48.45 & 50.29 & 49.82 , 50.76 & 0.0001
+ + # & & & + n & 51.11 & 50.43 , 51.79 & 49.04 & 48.44 , 49.64 & 0.0001 + e & 49.98 & 49.29 , 50.66 & 50.02 & 49.42 , 50.62 & 0.9237 + o & 51.86 & 51.19 , 52.52 & 48.39 & 47.79 , 48.99 & 0.0001 + a & 49.02 & 48.34 , 49.71 & 50.85 & 50.25 , 51.44 & 0.0001 + c & 47.69 & 47.03 , 48.34 & 52.01 & 51.41 , 52.60 & 0.0001 + + # & & & + n & 51.34 & 47.48 , 55.20 & 49.98 & 49.52 , 50.43 & 0.4793 + e & 51.80 & 48.00 , 55.60 & 49.97 & 49.51 , 50.42 & 0.3378 + o & 54.65 & 51.65 , 57.66 & 49.91 & 49.46 , 50.37 & 0.0032 + a & 45.91 & 42.12 , 49.71 & 50.08 & 49.62 , 50.53 & 0.0336 + c & 47.22 & 43.63 , 50.81 & 50.05 & 49.60 , 50.51 & 0.1209 +
n & 49.82 & 49.28 , 50.36 & 50.37 & 49.56 , 51.19 & 0.2673 + e & 50.90 & 50.35 , 51.44 & 48.17 & 47.38 , 48.97 & 0.0001 + o & 50.08 & 49.54 , 50.63 & 49.83 & 49.03 , 50.62 & 0.6023 + a & 50.05 & 49.50 , 50.60 & 49.89 & 49.09 , 50.69 & 0.7454 + c & 50.19 & 49.64 , 50.74 & 49.61 & 48.81 , 50.42 & 0.2441 +
n & 52.86 & 51.27 , 54.45 & 49.73 & 49.26 , 50.20 & 0.0003 + e & 48.50 & 46.78 , 50.21 & 50.14 & 49.68 , 50.61 & 0.0695 + o & 52.87 & 51.27 , 54.47 & 49.73 & 49.26 , 50.20 & 0.0003 + a & 46.77 & 45.09 , 48.44 & 50.31 & 49.84 , 50.77 & 0.0001 + c & 45.29 & 43.72 , 46.85 & 50.45 & 49.98 , 50.91 & 0.0001 + + # & & & + n & 49.64 & 43.98 , 55.29 & 50.00 & 49.55 , 50.46 & 0.8922
+ e & 45.85 & 39.05 , 52.64 & 50.04 & 49.59 , 50.49 & 0.2106 + o & 49.02 & 43.02 , 55.02 & 50.01 & 49.56 , 50.46 & 0.7320 + a & 44.36 & 38.95 , 49.77 & 50.05 & 49.60 , 50.50 & 0.0409 + c & 44.81 & 39.74 , 49.89 & 50.05 & 49.59 , 50.50 & 0.0447 + n & 56.56 & 55.08 , 58.05 & 49.31 & 48.85 , 49.77 & 0.0001 + e & 46.15 & 44.61 , 47.70 & 50.40 & 49.93 , 50.87 & 0.0001 + o & 52.18 & 50.72 , 53.64 & 49.77 & 49.30 , 50.25 & 0.0022 + a & 46.57 & 44.94 , 48.21 & 50.36 & 49.89 , 50.83 & 0.0001 + c & 46.20 & 44.75 , 47.65 & 50.40 & 49.93 , 50.87 & 0.0001 +
n & 50.62 & 49.82 , 51.42 & 49.68 & 49.13 , 50.22 & 0.0555 + e & 50.17 & 49.36 , 50.97 & 49.91 & 49.37 , 50.46 & 0.6056 + o & 54.70 & 54.05 , 55.35 & 47.54 & 46.99 , 48.09 & 0.0001 + a & 48.78 & 48.00 , 49.56 & 50.64 & 50.09 , 51.19 & 0.0001 + c & 47.45 & 46.67 , 48.23 & 51.34 & 50.80 , 51.88 & 0.0001 +
n & 49.95 & 49.44 , 50.45 & 50.20 & 49.18 , 51.23 & 0.6601 + e & 50.22 & 49.72 , 50.73 & 49.16 & 48.14 , 50.17 & 0.0648 + o & 49.89 & 49.37 , 50.40 & 50.42 & 49.48 , 51.36 & 0.3270 + a & 50.20 & 49.70 , 50.70 & 49.23 & 48.19 , 50.27 & 0.0988 + c & 50.13 & 49.62 , 50.63 & 49.51 & 48.50 , 50.53 & 0.2864 +
n & 53.24 & 50.50 , 55.99 & 49.89 & 49.44 , 50.35 & 0.0191 + e & 52.03 & 49.22 , 54.84 & 49.93 & 49.48 , 50.39 & 0.1464 + o & 51.40 & 48.83 , 53.97 & 49.95 & 49.49 , 50.41 & 0.2727 + a & 43.73 & 40.79 , 46.68 & 50.21 & 49.75 , 50.66 & 0.0001 + c & 46.72 & 44.32 , 49.11 & 50.11 & 49.65 , 50.57 & 0.0072 +
n & 50.07 & 49.59 , 50.55 & 49.46 & 48.10 , 50.81 & 0.3998 + e & 50.20 & 49.72 , 50.68 & 48.54 & 47.27 , 49.81 & 0.0166 + o & 50.04 & 49.56 , 50.52 & 49.71 & 48.40 , 51.03 & 0.6450 + a & 49.93 & 49.45 , 50.41 & 50.48 & 49.15 , 51.81 & 0.4462 + c & 49.95 & 49.47 , 50.43 & 50.37 & 49.01 , 51.72 & 0.5690 + + # & & & + n & 55.26 & 45.53 , 64.99 & 49.97 & 49.52 , 50.42 & 0.2540 + e & 46.46 & 38.43 , 54.50 & 50.02 & 49.57 , 50.47 & 0.3480 + o & 48.01 & 43.74 , 52.28 & 50.01 & 49.56 , 50.47 & 0.3242 + a & 39.61 & 30.12 , 49.11 & 50.06 & 49.61 , 50.51 & 0.0343 + c & 44.28 & 36.73 , 51.83 & 50.03 & 49.58 , 50.49 & 0.1211 + n & 50.28 & 47.99 , 52.58 & 49.99 & 49.53 , 50.45 & 0.8031 + e & 53.37 & 50.71 , 56.03 & 49.84 & 49.39 , 50.30 & 0.0110 + o & 56.15 & 54.15 , 58.15 & 49.71 & 49.25 , 50.17 & 0.0001 + a & 48.56 & 46.38 , 50.75 & 50.07 & 49.61 , 50.53 & 0.1837 + c & 46.98 & 44.78 , 49.18 & 50.14 & 49.68 , 50.60 & 0.0065 +
n & 58.66 & 55.84 , 61.47 & 49.86 & 49.41 , 50.32 & 0.0001 + e & 44.77 & 39.80 , 49.73 & 50.08 & 49.63 , 50.53 & 0.0376 + o & 52.41 & 48.00 , 56.81 & 49.96 & 49.51 , 50.42 & 0.2681 + a & 41.48 & 38.25 , 44.70 & 50.13 & 49.68 , 50.59 & 0.0001 + c & 43.04 & 38.95 , 47.12 & 50.11 & 49.66 , 50.56 & 0.0015 + + # & & & + n & 50.47 & 46.78 , 54.16 & 49.99 & 49.54 , 50.45 & 0.7952 + e & 47.23 & 42.01 , 52.46 & 50.06 & 49.61 , 50.51 & 0.2822 + o & 54.39 & 50.87 , 57.90 & 49.91 & 49.46 , 50.37 & 0.0148 + a & 44.50 & 40.15 , 48.85 & 50.11 & 49.66 , 50.56 & 0.0133 + c & 44.99 & 41.13 , 48.85 & 50.10 & 49.65 , 50.55 & 0.0113 + + # & & & + n & 53.13 & 51.32 , 54.94 & 49.77 & 49.30 , 50.23 & 0.0005 + e & 47.12 & 45.06 , 49.18 & 50.22 & 49.76 , 50.67 & 0.0044 + o & 53.16 & 51.53 , 54.79 & 49.76 & 49.30 , 50.23 & 0.0001 + a & 46.25 & 44.62 , 47.87 & 50.28 & 49.81 , 50.75 & 0.0001 + c & 44.50 & 42.88 , 46.11 & 50.41 & 49.95 , 50.88 & 0.0001 + n & 50.28 & 47.82 , 52.73 & 49.99 & 49.53 , 50.45 & 0.8196 + e & 52.70 & 49.90 , 55.51 & 49.90 & 49.44 , 50.35 & 0.0531 + o & 57.57 & 55.78 , 59.35 & 49.71 & 49.25 , 50.17 & 0.0001 + a & 50.03 & 47.88 , 52.17 & 50.00 & 49.54 , 50.46 & 0.9793 + c & 46.98 & 44.90 , 49.06 & 50.11 & 49.65 , 50.58 & 0.0045 +
n & 54.99 & 53.14 , 56.84 & 49.66 & 49.20 , 50.12 & 0.0001 + e & 45.27 & 43.10 , 47.43 & 50.32 & 49.87 , 50.78 & 0.0001 + o & 51.87 & 50.02 , 53.72 & 49.87 & 49.41 , 50.34 & 0.0406 + a & 46.00 & 43.84 , 48.15 & 50.27 & 49.82 , 50.73 & 0.0002 + c & 45.74 & 43.89 , 47.59 & 50.29 & 49.83 , 50.76 & 0.0001 + + # & & & + n & 49.79 & 46.82 , 52.75 & 50.01 & 49.55 , 50.46 & 0.8844 + e & 53.71 & 50.31 , 57.12 & 49.91 & 49.46 , 50.37 & 0.0309 + o & 57.89 & 55.73 , 60.05 & 49.81 & 49.35 , 50.27 & 0.0001 + a & 50.14 & 46.71 , 53.57 & 50.00 & 49.54 , 50.45 & 0.9352 + c & 48.03 & 45.33 , 50.74 & 50.05 & 49.59 , 50.51 & 0.1459 +
n & 51.32 & 50.59 , 52.04 & 49.10 & 48.52 , 49.67 & 0.0001 + e & 49.91 & 49.17 , 50.65 & 50.06 & 49.49 , 50.63 & 0.7485 + o & 51.57 & 50.85 , 52.28 & 48.92 & 48.35 , 49.50 & 0.0001 + a & 49.04 & 48.31 , 49.78 & 50.66 & 50.09 , 51.23 & 0.0007 + c & 47.69 & 46.99 , 48.38 & 51.59 & 51.01 , 52.16 & 0.0001 +
n & 50.92 & 46.02 , 55.82 & 49.99 & 49.54 , 50.44 & 0.6974 + e & 52.60 & 47.66 , 57.53 & 49.97 & 49.52 , 50.42 & 0.2828 + o & 56.30 & 53.63 , 58.97 & 49.93 & 49.47 , 50.38 & 0.0001 + a & 46.36 & 41.38 , 51.34 & 50.04 & 49.59 , 50.49 & 0.1407 + c & 49.99 & 45.49 , 54.49 & 50.00 & 49.55 , 50.45 & 0.9955 + who .
prevention of mental disorders : effective interventions and policy options : summary report .
geneva : world health organization ; 2004 .
available : http://www.who.int/mental_health/evidence/en/ prevention_of_mental_disorders_sr.pdf[http://www.who.int / mental_health / evidence / en/ prevention_of_mental_disorders_sr.pdf ] .
beaglehole r , bonita r , horton r , adams c , alleyne g , asaria p , baugh v , bekedam h , billo n , casswell s , et al .
priority actions for the non - communicable disease crisis . the lancet . 2011 ; 377(9775):14381447 .
bickel wk , johnson mw , koffarnus mn , mackillop j , murphy g james .
the behavioral economics of substance use disorders : reinforcement pathologies and their repair
. annual review of clinical psychology . 2014 ; 10:641677 .
roncero c , daigre c , barral c , ros - cucurull e , grau - lpez l , rodrguez - cintas l , tarifa n , casas m , valero s. neuroticism associated with cocaine - induced psychosis in cocaine - dependent patients : a cross - sectional observational study .
plos one . 2014 ; 9(9):e106,111 .
flory k , lynam d , milich r , leukefeld c , clayton r. the relations among personality , symptoms of alcohol and marijuana abuse , and symptoms of comorbid psychopathology : results from a community sample . experimental and clinical psychopharmacology .
2002 ; 10(4):425434 .
andreassen cs , griffiths md , gjertsen sr , krossbakken e , kvam s , pallesen s. the relationships between behavioral addictions and the five - factor model of personality .
journal of behavioral addictions . 2013 ; 2(2 ) : 9099 .
turiano na , whiteman sd , hampson se , roberts bw , mroczek dk . personality and substance use in midlife : conscientiousness as a moderator and the effects of trait change .
journal of research in personality . 2012
; 46(3):295305 .
haider ah , edwin dh , mackenzie ej , bosse mj , castillo rc , travison tg , group ls , et al .
the use of the neo - five factor inventory to assess personality in trauma patients : a two - year prospective study .
journal of orthopaedic trauma . 2002
; 16(9):660667 .
kopstein an , crum rm , celentano dd , martin ss .
sensation seeking needs among 8th and 11th graders : characteristics associated with cigarette and marijuana use . drug and alcohol dependence .
2001 ; 62(3):195203 .
yasnitskiy l , gratsilev v , kulyashova j , cherepanov f. possibilities of artificial intellect in detection of predisposition to drug addiction .
perm university herald series `` philosophy psychology sociology '' . 2015 ; 1(21):6173 .
valero s , daigre c , rodrguez - cintas l , barral c , gom - i - freixanet m , ferrer m , casas m , roncero c. neuroticism and impulsivity : their hierarchical organization in the personality characterization of drug - dependent patients from a decision tree learning perspective . comprehensive psychiatry . 2014 ; 55(5):12271233 .
bulut f , bucak i .
an urgent precaution system to detect students at risk of substance abuse through classification algorithms .
turkish journal of electrical engineering & computer sciences . 2014
; 22(3):690707 .
rioux c , castellanos - ryan n , parent s , sgu jr , the interaction between temperament and the family environment in adolescent substance use and externalizing behaviors : support for diathesis stress or differential susceptibility ? developmental review .
2016 ; 40 : 117150 .
weissman dg , schriber ra , fassbender c , atherton o , krafft c , robins rw , hastings pd , guyerb ae , earlier adolescent substance use onset predicts stronger connectivity between reward and cognitive control brain networks .
developmental cognitive neuroscience .
2015 ; 16:121129 .
fehrman e , egan v. drug consumption , collected online march 2011 to march 2012 , english - speaking countries .
icpsr36536-v1 . ann arbor , mi : inter - university consortium for political and social research [ distributor ] , 2016 - 09 - 09 .
deposited by mirkes e. http://doi.org/10.3886/icpsr36536.v1 stanford ms , mathias cw , dougherty dm , lake sl , anderson ne , patton jh .
fifty years of the barratt impulsiveness scale : an update and review . personality and individual differences . 2009 ; 47(5):385395 .
egan v , deary i , austin e. the neo - ffi : emerging british norms and an item - level analysis suggest n , a and c are more reliable than o and e. personality and individual differences . 2000 ; 29(5):907920 .
armbruster ws , di stilio vs , tuxill jd , flores tc , runk jl .
covariance and decoupling of floral and vegetative traits in nine neotropical plants : a re - evaluation of berg s correlation - pleiades concept .
american journal of botany .
1999 ; 86(1):3955 .
omote h , sugiyama k. method for drawing intersecting clustered graphs and its application to web ontology language . in proceedings of the 2006 asia - pacific symposium on information visualisation - volume 60 2006 jan 1 ( pp .
89 - 92 ) .
australian computer society , inc .
available : http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.294.2209&rep=rep1&type=pdf hoare j , moon d. ( eds . ) drug misuse declared : findings from the 2009/10 .
british crime survey home office statistical bulletin 13/10 ; 2010 .
available : https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/116321/hosb1310.pdf settles re , fischer s , cyders ma , combs jl , gunn rl , smith gt . negative urgency : a personality predictor of externalizing behavior characterized by neuroticism , low conscientiousness , and disagreeableness .
journal of abnormal psychology . 2012
; 121(1):160172 .
garca - montes jm , zaldvar - basurto f , lpez - ros f , molina - moreno a. the role of personality variables in drug abuse in a spanish university population .
international journal of mental health and addiction . 2009 ; 7(3):475487 .
fossati p , ergis am , allilaire jf .
problem - solving abilities in unipolar depressed patients : comparison of performance on the modified version of the wisconsin and the california sorting tests . psychiatry research . 2001 ; 104(2):145156 . mcdaniel sr , mahan je .
an examination of the impss scale as a valid and reliable alternative to the sss - v in optimum stimulation level research . personality and individual differences . 2008 ; 44(7):15281538 .
lee sy , poon wy , bentler pm . a two - stage estimation of structural equation models with continuous and polytomous variables .
british journal of mathematical and statistical psychology .
1995 ; 48(2):339358 .
martinson eo , hamdan ma .
maximum likelihood and some other asymptotically efficient estimators of correlation in two way contingency tables .
journal of statistical computation and simulation . 1971 ; 1(1):4554 .
gorban an , zinovyev ay . principal graphs and manifolds . in : olivas es , guerrero jdm , sober mm , benedito jrm , lpez ajs , editors .
handbook of research on machine learning applications and trends : algorithms , methods , and techniques .
hershey new york .
igi global ; 2009 .
2859 .
naikal n , yang ay , sastry ss .
informative feature selection for object recognition via sparse pca . in : computer vision
( iccv ) , 2011 ieee international conference on , ieee ; 2011 .
doi : 10.1109/iccv.2011.6126321 clarkson kl .
nearest - neighbor searching and metric space dimensions . in : shakhnarovich g , darrell t , indyk p , editors .
nearest - neighbor methods for learning and vision : theory and practice .
mit press ; 2006 .
1559 .
sofeikov ki , tyukin iy , gorban an , mirkes em , prokhorov dv , romanenko iv .
learning optimization for decision tree classification of non - categorical data with information gain impurity criterion . in : neural networks ( ijcnn )
joint confe . on , ieee ; 2014 .
35483555 .
dietterich t , kearns m , mansour y. applying the weak learning framework to understand and improve c4.5 . in : icml , proc .
of the 13th int .
conf . on machine learning .
san francisco : morgan kaufmann ; 1996 .
. 96104 .
mirkes em , alexandrakis i , slater k , tuli r , gorban an .
computational diagnosis of canine lymphoma .
journal of physics : conference series . 2014
; 490(1):012135 .
available : http://stacks.iop.org/1742-6596/490/i=1/a=012135 . | the problem of evaluating an individual s risk of drug consumption and misuse is highly important .
an online survey methodology was employed to collect data including big five personality traits ( neo - ffi - r ) , impulsivity ( bis-11 ) , sensation seeking ( impss ) , and demographic information .
the data set contained information on the consumption of 18 central nervous system psychoactive drugs .
correlation analysis demonstrated the existence of groups of drugs with strongly correlated consumption patterns .
three correlation pleiades were identified , named by the central drug in the pleiade : ecstasy , heroin , and benzodiazepines pleiades .
an exhaustive search was performed to select the most effective subset of input features and data mining methods to classify users and non - users for each drug and pleiad .
a number of classification methods were employed ( decision tree , random forest , @xmath0-nearest neighbors , linear discriminant analysis , gaussian mixture , probability density function estimation , logistic regression and nave bayes ) and the most effective classifier was selected for each drug .
the quality of classification was surprisingly high with sensitivity and specificity ( evaluated by leave - one - out cross - validation ) being greater than 70% for almost all classification tasks .
the best results with sensitivity and specificity being greater than 75% were achieved for cannabis , crack , ecstasy , legal highs , lsd , and volatile substance abuse ( vsa ) . |
to monitor high rate beam fluency gas ionization chambers based devices @xcite can be used . in order to obtain a 2d profile of radiation dose or fluency
it is necessary to have a highly segmented anode @xcite @xcite .
this represents a major problem when the size of the chamber is large compared with the anode segmentation pitch . in the present work we describe a new and simple solution developed to give real two dimensional readout on a gas ionization chamber .
the method consists on scanning electronically the detector active area in the perpendicular direction to the readout strips orientation ( strips instrumented with the measuring electro - meters ) .
gated charge transfer is a well known method which has been used in different radiation detectors like , for example , gated gas - filled time projection chambers @xcite in high energy physics .
@xmath5 v / m .
dimensions are given in @xmath0m.,width=377 ] the transfer of electrons drifting from the detector volume to the collector electrode strips can be blocked by means of a local electric field that dominates the drift field in the proximity of the gate electrode strips ( see simulation in figure [ puerta ] ) . in this way we provide a single stage control of the charge transfer from the gas drift region to the collector electrodes . in this work
we profit from the great development of micro pattern gas detectors ( mpgd ) @xcite in the last years , specially the advances on kapton etching techniques . in our case
we integrated a gating grid using a two metal layer circuit based in a kapton foil glued on a fr4 epoxy substrate to provide mechanical rigidity .
the use of a 25@xmath0 m thick kapton foil to produce a two level readout circuit allows to use a small gating voltage .
to design and optimize the geometry of the gate and collector electrodes we simulated the electric field map of a transfer gap cell through maxwell3d and matlab programs . for the final design it was chosen a 1:2:1 ratio corresponding to the transfer gap height : gate spacing : gate width . with this geometry we expected to have a low electron transfer transparency
( well bellow 1% ) using a relatively low positive ( + 5v ) _ `` closed gate '' _ voltage and moderate drift fields ( @xmath6 @xmath5 v / m ) . a wider gate strip spacing would increment the electron transparency for a given negative _ `` open gate '' _ voltage , but also higher positive voltages would be needed to block efficiently the electron drift . on the other hand the local electric field
present around the gate strip either in open or closed mode should not affect significantly the electric field in the drift region .
this is another important reason to use a small transfer gap between the gate and collector strips .
the readout circuit layout of this prototype is described in figure [ esquema ] .
we used a 25@xmath0 m thick kapton foil metal coated on both sides with 5@xmath0 m copper .
after etching of the copper , we obtained 35@xmath0 m wide and 1 mm wide strips on the gate and collector plane respectively .
the foil was glued on a 1 mm fr4epoxy substrate and subsequently kapton in the gap between gate strips was removed obtaining the figure [ esquema ] micro - pattern . in order to provide a simple connection procedure
a 64 pin 1/10 inch pitch standard connector was included in the board edge design for both gate and collector strips .
this choice implies a 1.27 mm pitch in the detector active area .
the collector strips were thus made 1 mm wide with 1.27 pitch while 17 gate strips were joined in a group in the detector edge to achieve the previous effective pitch .
the drift gas gap was built using a g10 6 mm height frame glued on the readout board , and this volume was closed with a drift electrode made of 200@xmath0 m thick g10 copper clad on one side .
v / m.,width=377 ]
the experimental transparency for electrons obtained as a function of the gate voltage is shown in figure [ conmutacion ] .
this measurements were done using pure ar at atmospheric pressure , and corresponds to the ratio of the ionization current under x ray irradiation measured at the collector and at the drift electrode .
the plot shows that very small transparency values ( @xmath7 ) can be obtained for low positive voltages at _ `` closed mode''_. the transparency increases when applying negative voltages at the gating electrodes , showing that there is a transition region centered on -15v , and can reach values very close to 1 ( _ `` open mode '' _ ) when the gate voltage is below -20v . if the drift field is increased , the gate voltage has also to be increased proportionally to maintain the same transparency value .
the fraction of current produced by electrons arriving to the gate electrode is the complementary value of the transparency calculated for the collector .
we have chosen -17v as _ open gate _
voltage as this value can be commutated using standard cmos analogue switches .
if we consider the width of the gate strips @xmath8 ( 35 @xmath0 m in figure [ esquema ] ) and the corresponding space between gate strips @xmath9 ( 40 @xmath0 m in figure [ esquema ] ) , the expected electron transparency @xmath10 is equal to the optical limit defined by @xmath11 being @xmath12 and @xmath13 the gate and drift voltages respectively , @xmath14 the height of the gate strips over the collector electrodes and @xmath15 the gas gap below the drift electrode .
the measured maximum transparency significantly deviates from this expectation due to the trapezoidal etching of the kapton layer that partially covers the collector electrodes in the space between gate strips .
to demonstrate the working principle of our position sensitive charge transfer scan readout method , a small area of the prototype was instrumented with charge integration electronics . to drive the detector ,
two separate electronic boards were used , one with the commutation circuit to sequentially commutate from closed to open gate voltage on the gate strips and another board with the necessary electronic readout channels to integrate the charge collected by each individual collector strip ( integrators board ) . to integrate the collector current we used a precision switched ivc102 @xcite integrating amplifier ( with an internal feedback capacitor set to 10 pf ) . to sequentially change the voltage value applied to each gate strip
a shift register was used , built by means of daisy chained d type flip flops ( hef40174b ) whose outputs were connected to double analogue switches ( ad7592di ) .
these analogue switches commute from closed gate voltage value ( that keeps the electron transparency at lowest values ) , to a negative open gate voltage value giving a high electron transparency for the selected gate strip ( typically @xmath1670% in our setup ) .
a scheme of the readout circuit is shown on figure [ esquema2 ] .
for the synchronous control of the two electronic boards we used a personal computer pci embedded card .
the first prototype was instrumented with a total number of 96 effective pixels ( on an active area of 15.2 @xmath1 10.2 mm@xmath4 ) , corresponding to 12 gate channels and 8 collector channels .
the total capacitance at the detector between gate and collector plane is 7nf , meaning a 2pf capacity per pixel .
this gate
collector capacity gives a 2@xmath0s transient time during commutation , considering the 3k@xmath17 impedance of the ivc102 ( with switches s1+s2 closed ) .
this dead time is small compared with the typical integration times used ( of the order of milliseconds ) .
the total time required to obtain an image is equal to the integration time needed to integrate the charge transfered by each gate strip times the number of gate strips .
figure [ ranura ] and [ anillo ] show two x ray images obtained with this prototype using a chromium x ray tube : a 1.5 mm slit between two 5 mm thick aluminum plates and a 5 mm screw nut were illuminated .
the @xmath18 and @xmath19 gate voltage values used were + 5v and -17v respectively , and the drift field applied in the image was 1.7@xmath110@xmath20 v / m in pure ar . to obtain
this images a scan readout cycle time of less than 10 seconds was required , with a collector current value of 7 pa in the pixels with maximum signal .
the charge transparency value when the gate electrode is set to the _ closed _ state should be ideally zero .
but the real value differs from zero by a small amount ( @xmath21 0.1% ) .
this causes a leakage current , of a value proportional to the detector area , that can seriously distort the image . nevertheless , if the leakage current per effective pixel is small we can correct this effect by what we have called _ differential readout_. the correction term is measured by integrating the detector current during the same time interval used for the standard readout with all the gate electrodes
closed_. irradiated with soft x
rays.,width=566 ] this value will then be subtracted as a pedestal from the values obtained with the gate electrodes on _ opened _ state . in this way
the image will not be dramatically affected by the non zero transparency value at the _ closed _ gate state .
we have proved the working principle of a simple and reliable readout method for 2d position sensitive gas ionization detectors .
this two dimensional charge transfer readout solution allows to cover large detector areas giving a high number of effective pixels and minimizing the number of readout electronic channels . by using a kapton
insulated two layer readout circuit , charge transfer does not affect the drift field and the control of the charge transfer process can be done with very low voltage values .
the detector can be used not only with gases but with other photo - conductive media like non polar liquids ( @xmath21 isooctane or tetrametilsilane ) . for low beam intensity or fast readout cycle applications , an intermediate gas avalanche device , like a gas electron multiplier @xcite , can be added to improve its sensitivity . considering moderate gas gains around 100
, an image could be readout with a cycle period 100 times faster than an equivalent ionization chamber , leading to a fast beam imaging device .
we are grateful to manuel sanchez from cern est / dem group for his permanent technical support and collaboration . | we have constructed and tested a 2d position sensitive parallel - plate gas ionization chamber with scanned charge transfer readout . the scan readout method described here
is based on the development of a new position dependent charge transfer technique .
it has been implemented by using gate strips perpendicularly oriented to the collector strips .
this solution reduces considerably the number of electronic readout channels needed to cover large detector areas . the use of a 25@xmath0 m thick
kapton etched circuit allows high charge transfer efficiency with a low gating voltage , consequently needing a very simple commutating circuit .
the present prototype covers 8@xmath18 @xmath2 with a pixel size of @xmath3 mm@xmath4 . depending on the intended use and beam characteristics
a smaller effective pixel is feasible and larger active areas are possible .
this detector can be used for x ray or other continuous beam intensity profile monitoring . , , , , , , , , gas ionization chamber , position sensitive detector , scan readout 29.40.cs , 29.40.gx , 07.85.fv |
it has been recently shown @xcite that sodium vanadium oxide is the second example of the inorganic spin - peierls compound following @xmath6 @xcite . similarly to @xmath6 the strong decrease of the susceptibility @xmath7 in @xmath8
is found below @xmath9 as a consequence of the opening of a spin - gap .
the spin - peierls transition is one of the most interesting phenomena observed in low - dimensional quantum spin systems .
it occurs in crystals containing linear chains of half - integer spin coupled by antiferromagnetic exchange interaction .
the spin - peierls transition is analogue of the peierls transition in metal and results in dimerisation of the lattice induced by magnetic fluctuations .
the lattice dimerisation and the existence of the spin - gap are also confirmed by x - ray , inelastic neutron scattering @xcite , nmr @xcite and esr @xcite measurements . using these techniques @xcite ,
the spin gap was estimated to be in the range from @xmath10 to @xmath11 .
the previously published infrared @xcite and raman @xcite spectra show the appearance of new modes below the transition temperature , but nothing was said about their origin .
based on an early measurement of the @xmath0 crystal structure @xcite , it has been shown that this oxide , grown in single crystalline form under ambient conditions , has an orthorhombic unit cell with parameters @xmath12 , @xmath13 , @xmath14 , @xmath15 and the space group @xmath16 ( @xmath17 ) .
such a crystalline structure assumes two kinds of vanadium chains along the b - axis .
one is magnetic @xmath18 ( @xmath19 ) and the other one is a nonmagnetic @xmath20 ( @xmath21 chains ) .
these chains are aligned parallel to each other by sharing an edge of adjacent @xmath22 pyramids , thus forming layers in @xmath23 plane .
@xmath24 atoms are situated between layers as intercalates .
this magnetic structure is expected to be truly one - dimensional since the chains caring the spin are separated from each other by nonmagnetic chains along @xmath25-axis . the problem with this structure is missing physical argument for such a strong charge modulation .
the most recent structural analysis @xcite shows that all vanadium atoms are equivalent and in mixed valent state .
such a crystal structure is also predicted from polarized raman scattering measurements @xcite . even without charge modulation , due to strong electron correlations , it is possible to have formation of antiferromagnetic heisenberg chains @xcite .
it is also possible that an induced zigzag type of charge ordering along the ladders of v - ions is formed @xcite .
however , in this paper we consider only 1-d antiferromagnetic heisenberg model as an explanation for the magnetic behaviour of this compound .
this is based on the magnetic susceptibility measurements of @xmath0 powder @xcite which shows a broad maximum at @xmath26 , followed by rapid decrease below @xmath27 .
such a susceptibility behaviour can be well described as linear heisenberg antiferromagnet in the high temperature phase and by bulaevskii theory @xcite in dimerized phase , indicating that there is no frustration in this material . in this work we present low temperature raman spectra of @xmath0 single crystal .
the assignation of the observed modes is done on the basis of monte carlo calculation of excited states of heisenberg dimerized hamiltonian , without frustration .
we found a good agreement between measured and calculated magnetic excitations for @xmath1 and @xmath28 .
the singlet - triplet and singlet - singlet excitations as well as two - magnon mode are observed to be at @xmath3 , @xmath4 and @xmath5 .
the single crystals of @xmath8 with a size of approximately @xmath29 are obtained as described in @xcite .
the spectral lines of an argon - ion laser are used as an excitation source .
the light beam was focused on the sample surface using cylindrical lens with an average power of @xmath30 .
the scattered light was collected with an objective , aperture 1:1.4 .
the monohromator used was a jobin ivon model u1000 , equipped with a pelletier - effect - cooled rca 31034 photomultiplier with a conventional photon counting system .
the temperature was controlled with closed - cycle helium cryostat ( leybold ) .
numerical calculation were carried out on a 4-processor origin 2000 computer , courtesy of institute of physics computer facilities ( ipcf ) .
the polarized raman spectra measured with @xmath31 laser line at temperature of @xmath32 are presented in .
the new modes found at temperatures below transition temperature @xmath33 are denoted in the graph .
the lowest excitation is observed at @xmath34 , with the largest intensity in @xmath23 polarized spectra .
this mode is observed in @xmath35 polarization as well .
the continuum of magnetic excitations starts at @xmath36 which is exactly two times @xmath34 energy and is observed only in @xmath23 configuration .
the @xmath37 mode is found to be largest in @xmath35 spectrum .
the other modes in the spectra are high temperature phonons assigned on the basis of valence shell model calculation @xcite .
the broad structure present in @xmath38 spectra is assigned as electronic continuum which appears due to @xmath39 electronic transitions of @xmath18 ions @xcite . to further analyse the origin of new modes we measured the raman spectra at different temperatures from @xmath40 to @xmath41 , .
we found that @xmath34 mode exhibits the shift towards lower energy of @xmath42 by increasing temperature . besides the @xmath36 mode that shows similar shift , we observed additional structure at @xmath43 in the @xmath44 spectrum , denoted with an arrow . to resolve these two structures we performed higher resolution measurements ( resolution@xmath45 ) , see inset . by increasing the temperature
this part of the spectrum rapidly looses intensity and the difference between these two modes is no longer observable .
moreover , the existence of the continuum is confirmed by observing the fano - type asymmetry of the @xmath46 phonon with energy overlapping with that of the continuum .
the asymmetry of this mode is becoming less pronounced by increasing temperature and finally becoming complete lorenzian in the @xmath47 spectrum .
other modes in the spectra do not show frequency shift and we assigne them as phonons becoming active due to doubling of the unit cell .
here we present the results of the numerical calculation based on green s function monte carlo technique ( gfmc ) @xcite .
we assumed that @xmath0 can be modelled as 1-d heisenberg antiferomagnet with modulated nn coupling and no frustration .
the relevance of this model comes from susceptibility measurements @xcite were it was shown that low temperature behaviour can be well described with the hamiltonian : @xmath48 where @xmath49 are spin-1/2 operators .
further , we assumed the periodic boundary conditions so that only rings with even numbers of spins were studied .
we have calculated energies of the ground , the first excited triplet ( @xmath50 ) and quintuplet ( @xmath51 ) states for rings with @xmath52 spins .
all these energies are calculated exactly with no approximation .
the resulting triplet and quintuplet gaps , @xmath53 and @xmath54 , are shown in for @xmath55 . the extrapolation to @xmath56 is done using a @xmath57 function .
similar simpler functions were used in @xcite to extrapolate gaps obtained by lanczos diagonalization for systems up to @xmath58 . here , we have chosen more complicated ansatz to fit points for smaller @xmath59 also .
the estimated error for the extrapolated values of triplet and quintuplet gaps are less then 1% .
the size of symbols in figures do not represent this error .
further details of extrapolation procedure can be found in @xcite .
the extrapolated results for all @xmath60 considered are given in .
it is found that the singlet - triplet gap can be excellently fitted by the simple formula @xmath61 , @xmath62 .
this was also found in the recent study @xcite . in the limit
@xmath63 this formula coincides with leading three orders of the perturbation theory , @xcite .
the @xmath64 region was of no concern to us , and we do not know were the crossover to the known @xcite critical behaviour @xmath65 occurs .
the singlet - quintuplet gap can not be fitted in such a simple fashion .
we have found that it lies above the double singlet - triplet gap value , at the energy @xmath66 .
this is consistent with the rpa study @xcite were this mode was identified as an anti - bound two - magnon state .
the anti - boundedness @xmath67 decreases towards zero by increasing @xmath60 to 1 .
according to the estimated error of our calculation the existence of a quintuplet state at energy differenet then 2@xmath53 is established for @xmath68 , see inset .
to begin discussion , first we compare our gfmc calculation with other numerical results .
the ground state energy is in excellent agreement with bethe ansatz solution @xcite in the limit @xmath69 and with the results for isolated dimmers when @xmath63 @xcite .
moreover , our values for the triplet gap coincide precisely with the exact diagonalization results @xcite .
the most important point , which is one of our main numerical results , is exact calculation for the lowest lying quintuplet state , @xmath51 .
as allready mentioned above , this mode at energy @xmath66 was identified as an anti - bound two - magnon state @xcite . according to that study one
should also expect singlet , two - magnon bound state , at the energy @xmath70 .
such an excited singlet state can not be calculated exactly using gfmc but it can be estimated using the above formula . the low temperature susceptibility measurements @xcite can be well described with bulaevskii theory , indicating the spin - peierls transition at 34 k. the fit gave value of the exchange integral @xmath71 and dimerization @xmath60 to be @xmath72 and @xmath1 . using slightly different parameters @xmath73 and @xmath1 we calculated singlet - triplet and singlet - quintuplet transition energies to be @xmath74 and @xmath75 .
these values are obtained by a linear interpolation between calculated values for @xmath76 and @xmath55 .
the value for singlet - singlet transition @xmath77 is obtained using above estimate .
this value is in good agreement with exact diagonalization study which gives singlet to triplet gap ratio of @xmath78 @xcite . in the raman
spectra the @xmath34 feature coincides with our numerical result for the triplet gap and it is possible that this mode comes from one - magnon scattering process .
this value is in agreement with other measurements of the spin - gap .
moreover , this mode shifts towards lower energies by increasing temperature and by decreasing sodium concentration @xcite .
it was proved that sodium deficiency strongly suppresses the sp transition @xcite , which is an additional proof for the magnetic origin of this mode .
still , behaviour of this mode in magnetic field remains to be studied .
some preliminary results show no change of this mode in the magnetic fields up to @xmath79 @xcite . the two - magnon mode is found to be at @xmath36 which is exactly twice the gap energy .
it appears as a well - defined peak due to the one - dimensionality of the magnetic interaction .
then the structure at @xmath43 can be described as singlet - singlet excitation with boundness @xmath80 , which is in agreement with our calculation .
it is interesting to compare our results with that of @xmath6 which is the first inorganic material exhibiting the spin - peierls transition .
the magnetic mode at @xmath81 observed in @xmath6 @xcite is described as excitation from the ground state to singlet state with singlet to triplet gap ratio of @xmath82 .
such a ratio is a consequence of frustration , which is quite large in this material . from our raman spectra we found that in @xmath0 this ratio is around @xmath83 , which is consistent with assumption that frustration is negligible in this material .
moreover , the broad magnon band , centred at about twice the maximum magnon energy is not observed in our raman spectra which again proves existence of no frustration in @xmath0 @xcite .
the appearance of magnetic scattering in @xmath23 polarized configuration can not be explained using exchange scattering mechanism of fleury and loudon @xcite , since this theory predicts the magnetic scattering in configuration parallel to the chains , ie . in @xmath35 direction .
our previous study of raman spectra @xcite , using different laser line energies shows existence of resonance effects which may influence selection rules for magnetic scattering .
it is also shown that @xmath34 feature , in the spectra measured with @xmath84 laser line consists of two modes separated by @xmath85 .
still , the mechanism for the apearance of the second mode is not yet established .
it was proposed that this mode , as well as , the @xmath37 comes from crystal field excitations of @xmath18 ions .
finally , we conclude that magnetic excitations in the raman spectra of @xmath0 can be well described using gfmc calculation of excited states of dimerized 1-d heisenberg antiferomagnet with @xmath73 , @xmath1 and no frustration .
this work is supported by the serbian ministry of science and technology under projects no .
01e10m1 , 01e15 and 01e09 .
99 isobe m and ueda y 1996 _ j. phys .
soc . japan _ * 65 * 1178 . | we present low temperature raman spectra of @xmath0 single crystal .
the assignation of the observed modes is done on the basis of monte carlo calculation of excited states of heisenberg dimerized hamiltonian , without frustration .
we found a good agreement between measured and calculated magnetic excitations for @xmath1 and @xmath2 .
the singlet - triplet and singlet - singlet excitations as well as two - magnon mode are observed to be at @xmath3 , @xmath4 and @xmath5 . |
40004000 = 1000 # 1
40004000 = 1000 the distortion of the density field in redshift - space by peculiar motion is an old subject ( see e.g. ) .
it is commonly held that on small scales , virialized motion causes the stretching of structures along the line of sight , creating what is often called the fingers - of - god , while on large scales , coherent infall into high density regions causes compression , thereby making structures appear more wall - like in the tangential directions than they are in real - space .
the reader is referred to hamilton @xcite for a comprehensive review .
it has also been recognized for some time , however , that the classic linear prediction by kaiser @xcite overpredicts the compression , even at relatively large scales where the corresponding density field in real - space is quite linear ( @xmath0 ) .
conversely , the dilution effect embodied by the fingers - of - god appears to extend to rather large scales .
( see e.g. . )
we will refer to this phenomenon as translinear redshift distortions .
this effect has often been discussed in the context of attempts to measure @xmath2 ( @xmath3 , where @xmath4 is the bias ) using the ratio of the real - space to redshift - space two - point function or the quadrupole - to - monopole ratio of the redshift - space power spectrum ( see e.g. ) .
it is found , both from observations , as well as from n - body simulations , that the linear theory predictions for these ratios fail even at very large scales , where perturbation theory is known to work well in real - space . in general
, one finds that the amount of line - of - sight squashing of the redshift - space two - point function is overpredicted by the linear perturbative formulae .
the large uncertainties in published values of @xmath2 are at least in part due to our poor understanding of this effect ( see table 1 of )
. a related phenomenon has also been observed in studies of one - point statistics of the density field in redshift - space , albeit in a slightly more subtle form .
hivon et al .
@xcite computed the leading perturbative contribution to @xmath1 and @xmath5 in redshift - space .
they observed that the agreement on large scales between the predicted values and the measured values from n - body simulations is not as good as it is in the case of real - space . only at very small @xmath6 ( the linear fluctuation amplitude in real - space ; in other words , very large smoothing scales ) did they obtain rough agreements .
in fact , above some moderate @xmath6 ( @xmath7 ) , the value of @xmath8 in redshift - space generally lies below that in real - space , suggesting an effect that resembles the fingers - of - god .
recently , taylor & hamilton @xcite and fisher & nusser @xcite ( see also hatton & cole for detailed comparisons with simulations ) revisited this issue by computing the quadrupole - to - monopole ratio of the redshift - space power spectrum using the zeldovich approximation (; henceforth as za ) .
both pairs of authors were able to obtain accurate predictions for the shape of the quadrupole - to - monopole ratio on translinear scales .
in particular , they demonstrated analytically that the departure from the linear prediction could take place at relatively large scales , and they interpreted the success of the za in accounting for this effect as an indication that the failure of the linear prediction on translinear scales have more to do with coherent infall rather than virialized motion .
it is our aim here to continue this line of investigation of nonlinear effects in redshift distortions , but we will focus our attention on one - point statistics . following earlier work by mcgill @xcite , we explore the role of caustics and multiple - streaming in redshift - space .
it was pointed out by mcgill @xcite , who used one - dimensional zeldovich dynamics , that redshift - space caustics could form even when the real - space density field is only mildly nonlinear .
we will quantify this by computing the average number of streams in both real and redshift - space using the za .
as will be shown , at @xmath9 , the degree of multiple - streaming could be significant in redshift - space but negligible in real - space .
this is covered in [ nostreams ] .
the above finding indicates that the real to redshift - space mapping enhances the level of nonlinearity .
an interesting feature of multiple - streaming in redshift - space is that , unlike their real - space counterparts , redshift - space multiple streams to a first approximation can flow freely past each other ( at least before real - space caustics actually form ) , because they arise from physically distinct regions in real - space .
this might provide an interesting explanation for why the famous kaiser formulae overpredicts the amount of redshift - space compression along the line of sight , even at moderate @xmath6 s .
redshift - space multiple - streams could then provide a mechanism for counteracting the linear compression .
the free - crossing of redshift - streams in the intermediate regime is not unlike the familiar phenomenon of the thickening of the za pancake . in this sense , the za , which can be viewed as a local approximation where interaction is ignored ( see for a review ;
see also ) , is well - suited for studying this phenomenon .
the formalism for computing the average number of streams was in fact originally developed to calculate the full one - point probability distribution function ( pdf hereafter ) of density .
our work here is a natural extension of earlier work by kofman et al .
@xcite who computed the pdf in real - space using the za
. we will also use the same methodology to compute the pdf as well as the skewness @xmath1 in both real and redshift - space in the small @xmath6 limit .
the formalism is laid out in [ formalism ] , the computational method is explained in
[ method ] , and the results of the pdf and @xmath1 calculation are discussed in [ zpdf ] and [ s3 ] . we particularly focus our attention on the high density regime of the pdf .
catastrophe theory tells us that caustics generally induce a @xmath10 tail , which , however , is not observed in n - body simulations with cdm - type ( cold - dark - matter ) power spectra , either in real or redshift - space .
in fact , an intriguing property of the pdf in cdm models is that it is well - fit by a single parameter lognormal distribution ( in both real and redshift - space ) , where the caustic - tail is absent ( for real space , see e.g. ) .
we explore the reason for it in [ smoothing ] , and conclude that smoothing in redshift - space ( or in real - space ) , due to both finite resolution or discreteness and small - scale velocity dispersions , plays an important role in erasing the caustic - feature in the pdf . to understand properly the behavior of the pdf then requires a non - local calculation , which is beyond the scope of this paper .
we briefly discuss possible lines of attack in [ conclude ] . for the purpose of this article ,
we denote by @xmath11 and @xmath12 the densities in real ( or eulerian ) and redshift space respectively , normalized by the mean density ( i.e. @xmath13 ) .
the evolution of @xmath11 is determined by the mapping from lagrangian to eulerian space , according to mass conservation : similarly , the evolution of the redshift - space density is determined by the mapping from eulerian to redshift - space : @xmath19 where @xmath20 is the redshift - space coordinate , @xmath21 is the mean redshift of interest , @xmath22 is the hubble parameter at redshift @xmath21 , @xmath23 is the peculiar velocity and @xmath24 is the unit vector along the line of sight .
the redshift - space density @xmath12 is then given by @xmath25\right| = \rho \ , \left| { \rm det}^{-1 } \left [ { \partial s^i \over \partial x^j } \right ] \right| = \rho \left| 1 + { { 1+z}\over h } { \partial { \bf v}\over \partial x^i}\cdot{\bf e } e^i \right|^{-1 } , \label{rhoz}\ ] ] where eq .
( [ rho ] ) has been used for the second equality .
we have used the remote - observer approximation for the third equality , in other words , setting the term @xmath26 .
this is equivalent to assuming the directional unit vector does not change appreciably for a small displacement in the transverse direction .
in addition , an interesting matrix identity has also been used : @xmath27 = 1 + { \bf a } \cdot { \bf b}$ ] , where @xmath28 and @xmath29 are arbitrary vectors .
the za consists of assuming @xmath30 where @xmath31 is the linear time - dependent growth factor
. growing mode ( in other words , irrotational ) initial condition then implies @xmath32 where @xmath33 is the displacement potential and @xmath34 in component form is @xmath35 .
putting the above into eq .
( [ rhoz ] ) , we arrive at the following expression for @xmath12 @xmath38 where @xmath39 and @xmath40 @xcite , can be found in and , but we will adopt @xmath41 , which is a very good approximation at @xmath42 . ] with @xmath43 being the scale factor and @xmath44 being the fraction of critical density in matter .
therefore , for a given @xmath44 , @xmath12 is uniquely determined by the line - of - sight unit vector @xmath45 and the displacement matrix @xmath46 .
the probability distribution of @xmath12 is then dictated by the probability distributions of the later two quantities . for gaussian random initial conditions ,
the probability distribution , in lagrangian space , of the displacement matrix @xmath46 can be computed exactly .
one can always go to a frame in which @xmath46 is diagonal , keeping in mind that the line - of - sight is randomly oriented with respect to the eigenvectors of this matrix .
all we need then is the probability distribution of its eigenvalues .
we will denote this probability distribution in lagrangian space by @xmath49 , where @xmath50 s ( @xmath51 ) are the eigenvalues of @xmath52 , @xmath53 .
thus , the density in redshift - space can ultimately be expressed in terms of the eigenvalues and orientation of the local lagrangian deformation tensor . the statistical properties of @xmath12
can then be inferred from the well - known statistics of the deformation tensor .
( [ rhoz3 ] ) , ( [ rhoz4 ] ) represents a redshift - space generalization of the familiar zeldovich formulae for the density in real - space . just like the zeldovich result in real space , it sheds light on the main features of the local structures in redshift space
the first three factors in eq .
( [ rhoz3 ] ) obviously arise from the mapping from the lagrangian to eulerian ( or real ) space and the last factor comes from that from real to redshift - space .
the latter mapping strongly depends on the orientation of the deformation tensor with respect to the line of sight @xmath24 . in fig .
[ sergei ] , we show an example of the density evolution of of a fluid element , with the eigenvalues today ( @xmath56 ) of @xmath57 and @xmath58 .
two cases are considered : when the line of sight is aligned with the axis of fastest collapse ( @xmath59 ) ( short dashed line ) , and when the line of sight is aligned with the eigenvector associated with @xmath60 ( long dashed line ) .
the evolution of the real - space density is shown as a solid line .
the real - space density formally diverges at @xmath56 because of caustic - formation .
the redshift - space density tends to become nonlinear even earlier . in particular , this redshift - space amplification works most efficiently if the axis of the fastest collapse is aligned with the line of sight .
what we have here can be interpreted as a nonlinear manifestation of the linear effect noticed by kaiser ( 1987 ) .
note how the redshift - space density decreases after ( redshift- ) caustic formation ( at @xmath61 , for the short dashed line ) .
this corresponds to the free - crossing of redshift - space multiple - streams , which counteracts the kaiser compression of structures along the line of sight .
it is also important to emphasize that eq .
( [ rhoz3 ] ) only gives us a local picture of the density distribution .
as in the familiar case of the za in real - space , our local lagrangian treatment suggests an anisotropic structure , except that the structure is preferentially transverse to the line of sight .
however , it should be cautioned that the above equations tell us nothing about the global distribution of these local structures , whether they are aligned in a filament or a sheet is actually determined by the correlation function of shear ( see ) .
moreover , small - scale velocity dispersions are likely to at least smear out the caustics , which we will come back to later .
the only situation in which we can say something about the global distribution of structures using our local lagrangian description is for a truncated initial power spectrum , in other words with a displacement field that is smoothed on large scales in lagrangian space , similar to what is done in the truncated zeldovich approximation ( ) .
the lagrangian pdf of ( redshift - space ) density is : @xmath62 where @xmath63 is the dirac delta function , @xmath64 in its argument is a function of @xmath65 and @xmath45 as expressed in eq .
( [ rhoz3 ] ) , @xmath66 is given by eq .
( [ pei ] ) , and @xmath67 is the probability that a given lagrangian element has the set of eigenvalues @xmath50 s falling in the prescribed ranges .
in other words , @xmath68 tells us the fraction of the lagrangian volume that has the redshift - space density @xmath69 . to obtain the redshift - space pdf of the ( redshift - space ) density , two effects need to be taken into account .
the redshift - space pdf is supposed to tell us the fraction of the redshift space volume that has the redshift - space density within a particular range , say @xmath69 .
the first effect arises from the fact that the lagrangian and redshift - space pdfs differ by a multiplicative volume factor : a given lagrangian volume corresponds to a different redshift - space volume .
this factor is none other than @xmath12 itself . in other words ,
the redshift - space pdf should be equal to the lagrangian pdf divided by @xmath12 .
if there were no multiple - streaming , this would be the whole story .
the second effect that has to be taken into account is illustrated schematically in fig .
[ qsmap ] .
multiple - streaming is said to occur in places where the mapping from @xmath37 to @xmath20 is no longer one - to - one .
the expression for @xmath12 in eq .
( [ rhoz ] ) is defined for every lagrangian coordinate @xmath37 . in other words , it is a single - stream redshift - space density . in regions where multiple - streaming occurs ,
the actual redshift - space density , at a particular @xmath20 say , should be a sum of @xmath12 s over all @xmath37 s that map onto the same @xmath20 .
such a non - local calculation is beyond the scope of this paper . instead , following kofman et al .
@xcite , we assume the fraction of the total redshift - space volume that is occupied by these multiple - stream - regions is not large ( i.e. mildly nonlinear regime ) . however , there are inevitably some such regions .
a renormalizing factor @xmath70 is then needed when computing the redshift - space pdf : @xmath71 where @xmath72 is given in eq .
( [ pl ] ) and @xmath73 the factor of @xmath74 takes care of the first effect mentioned above : it is the ratio of redshift - space volume to lagrangian - space volume . however
, since @xmath75 is the single - stream redshift - space density , an overcounting of the total redshift - space volume over the total lagrangian - space volume occurs .
the true ratio of the two should be unity .
@xmath70 gives the ratio of the total _ single - stream _ redshift - space volume to the total lagrangian - space volume ( the latter is also equal to the true total redshift - space volume with no overcounting ) , which is in general larger than @xmath76 because of multiple - streaming ( or , in other words , overcounting ) .
this quantity provides the correct renormalizing factor for @xmath77 .
an additional bonus of the above formalism is that we have a quantitative measure of the degree of multiple - streaming in @xmath70 .
it tells us the average number of streams at an arbitrary point in redshift - space .
[ qsmap ] provides a nice illustration of how @xmath70 should be interpreted .
suppose @xmath78 is the total true ( no overcounting ) volume in redshift - space ( i.e. the `` volume '' shown on the y - axis ) .
suppose further that @xmath79 is the part of @xmath78 that is in single - stream regions , and @xmath80 is the part that is in triple - stream regions i.e. @xmath81 .
recall that @xmath70 is the ratio of the total _ single - stream _ redshift - space volume to the total true redshift - space volume , which means @xmath82 , where @xmath83 and @xmath84 are the probabilities that a given region is in single - stream and triple - stream regimes respectively .
hence , @xmath70 is exactly what one would refer to as the average number of streams at a point in redshift - space .
obviously , the argument extends to higher ( but always odd ) number of streams .
note furthermore that while eq .
( [ pz ] ) gives only an approximate redshift - space pdf in the limit of little multiple - streaming ( @xmath70 close to 1 ) , eq .
( [ ns ] ) is an exact expression for the average number of streams , at least within the framework of the za .
this should be kept in mind in the discussion of the next section .
lastly , note that the above expressions ( eq . [ [ rhoz3 ] ] , [ [ pl ] ] , [ [ pz ] ] , [ [ ns ] ] ) reduce to their real - space counterparts in the limit @xmath85 , in other words , when the real to redshift - space mapping is trivial .
the quantities we are interested in are @xmath86 and @xmath70 in eq .
( [ pz ] ) and ( [ ns ] ) .
both of them require performing the integral in eq .
( [ pl ] ) .
kofman et al .
@xcite succeeded in reducing the analog of this integral in real - space ( i.e. setting @xmath85 in eq .
[ [ rhoz3 ] ] ) to a one - dimensional integral which has to be done numerically .
redshift - space distortions introduce extra complications .
we will instead perform the integral in eq .
( [ pl ] ) numerically right from the beginning .
the delta function restriction @xmath87 $ ] is most easily dealt with using monte carlo techniques ( see for a similar calculation in the context of the pdf of the optical depth rather than the density in redshift - space ) . in other words , we generate realizations of @xmath50 s and @xmath45 s , and identify combinations of them which satisfy the delta function restriction .
the directional unit vector @xmath45 is easily handled because it is randomly distributed over the 2-sphere , with the probability distribution given by eq .
( [ pei ] ) .
on the other hand , the probability distribution for the @xmath50 s is somewhat more complicated .
the probability distribution of the eigenvalues of the displacement matrix @xmath46 , for a gaussian random field , was first calculated by doroshkevich @xcite .
we adopt a form for this distribution which is better suited for the monte carlo computation outlined above ( following ; see also ) .
let @xmath88 s be the eigenvalues of the displacement matrix @xmath46 , and let @xmath89 , @xmath90 and @xmath91 be variables defined through @xmath50 s as follows @xmath92 = { \delta\over 2 } { \rm cos \gamma } [ 1,1,1 ] + { \delta\over \sqrt{5}}{\rm sin}\gamma [ { \rm cos}({{\alpha+2\pi}\over 3 } ) , { \rm cos}({{\alpha-2\pi}\over 3 } ) , { \rm cos}({\alpha \over 3 } ) ] \ , , \label{angles}\ ] ] where the @xmath89 ranges from @xmath93 to infinity , and both @xmath90 and @xmath91 are between @xmath93 and @xmath94 .
the above ranges enforce the ordering @xmath95 .
it can be shown that for any given combination of @xmath96 s in this order , there corresponds a unique set of @xmath89 , @xmath90 and @xmath91 .
the probability distribution of @xmath50 s is given by ( ):
@xmath97 ^ 2\right ) d \delta^2 \right ] \left[{8\over 3\pi } { \rm sin}^4\gamma d\gamma\right ] \left[{1\over 2 } { \rm sin}\alpha d\alpha\right ] \ , .
\label{plambda}\ ] ] where @xmath6 is the root - mean - squared ( rms ) amplitude of the linear real - space density fluctuation . a nice feature of the set of variables @xmath89 , @xmath90 and @xmath91 is that their joint probability distribution factorizes .
@xmath98 is distributed like a gamma - deviate .
@xmath99 can be integrated exactly to give @xmath100 where @xmath101 , so that @xmath102 is distributed uniformly between @xmath103 and @xmath76 .
lastly , @xmath104 is also uniformly distributed between @xmath103 and @xmath76 .
standard numerical methods exist for generating variables distributed in the above manner ( see ) .
note that the only free parameters in the above calculation are @xmath6 and @xmath44 , or more precisely , @xmath105 .
the former controls the distribution of @xmath50 s through eq .
( [ plambda ] ) , while the latter determines the degree of redshift - space distortion ( eq .
[ [ rhoz2 ] ] or [ [ rhoz3 ] ] ) .
[ avestream ] shows the result of the calculation outlined above for @xmath70 , the average number of lagrangian streams , defined in eq .
( [ ns ] ) .
this is shown as a function of @xmath6 ( eq . [ plambda ] ) , the rms amplitude of the linear real - space density fluctuation , for three different values of @xmath44 .
the @xmath106 case corresponds to the average number of streams in real - space .
several points should be noted .
first , redshift - space distortion , or peculiar motion , has the effect of increasing the amount of multiple - streaming over that in real - space .
second , a larger @xmath44 means more redshift - space distortion , hence a higher average number of streams . at @xmath107 ,
the average number of streams in real - space is very close to 1.0 , while the average number of streams in redshift - space is 1.1 and 1.2 for @xmath108 and @xmath109 respectively .
this means multiple - streaming can be insignificant in real - space , but going on in earnest in redshift - space
. might not seem too different from @xmath110 .
however , it should be kept in mind this means , at any given point in redshift - space , one finds on the average @xmath111 streams . in other words ,
the total single - stream redshift - volume is @xmath112 higher than the total true redshift - volume . ] in other words , even at relatively small @xmath6 , when the real - space density field is quite linear , redshift - space distortion introduces inherently non - perturbative effects through multiple - streaming .
the above result perhaps explains the success of perturbation theory , which is by nature a single - stream theory , in estimating various statistics of the density field in real - space , at @xmath0 ( many references should be cited here , see and for short reviews .
see also and for loop corrections which can be important at large enough @xmath6 . ) .
the same does not appear to be true for statistics of the redshift - space density field , however .
for instance , it is well - known that the linear - theory redshift distortion factor deduced by kaiser @xcite fails to describe the quadrupole - to - monopole ratio of the redshift - space power spectrum even at very large scales ( see hatton & cole and references therein for a recent discussion ) .
this failure is an indication of the importance of non - perturbative effects such as those due to multiple - streaming in redshift - space .
the fact that multiple - streaming in real - space is less severe than multiple - streaming in redshift - space has another interesting implication .
recall the lagrangian to real - space to redshift - space mapping : @xmath113 .
the multiple-(lagrangian)-streaming in redshift - space arises not so much from the @xmath114 mapping ( otherwise , the average number of lagrangian streams in real - space would not be so close to @xmath76 for , say , @xmath107 ) , but rather from the @xmath115 mapping .
this is interesting because the multiple - streams at a given @xmath20 then mostly come from distinct places in real space i.e. they do not interact strongly , unlike the case of multiple-(lagrangian)-streams at the same @xmath116 .
one well - known failure of the zeldovich approximation is that it does not account for the gravitational interaction between different lagrangian streams around a real - space caustic i.e. the different streams ( unphysically ) freely go past each other at a zeldovich pancake . around a redshift - space caustic ,
however , the multiple streams can ( approximately ) free - stream past each other because they are not located at the same real - space location .
this might provide a mechanism for partially canceling the line - of - sight compression of structures pointed out by kaiser ( 1987 ) .
we compute the pdf of redshift - space density according to eq .
( [ pz ] ) for @xmath117 and @xmath118 , and for 2 different @xmath6 s : the result is shown in fig .
[ zapdf ] .
the @xmath118 curves correspond also to the pdfs of the real - space density .
we have chosen @xmath6 s small enough so that the average number of streams is close to @xmath76 ( see fig . [ avestream ] ) .
this is to ensure the validity of the approximation involved in eq .
( [ pz ] ) .
it can be seen that peculiar motion has the effect of making the pdf look more nonlinear in redshift - space than it is in real - space .
this should be expected based on the kaiser @xcite effect alone : large scale coherent infall into initial density enhancements ( in real - space ) shifts such regions further out into the high - density tail of the pdf , while at the same time creating lower - density voids thereby making the pdf peak at lower densities . at higher @xmath6 s , translinear distortions , or redshift - space multiple - streaming ,
would start to cancel the kaiser compression , and begin to cause dilution around the initial density enhancements . at
still higher @xmath6 s , a qualitatively different dilution effect , the finger - of - god , kicks in due to the formation of virialized clusters .
this happens after multiple - streaming becomes important in real - space .
unfortunately , the pdf as expressed in ( [ rhoz3 ] ) is incapable of taking multiple - streaming ( in either redshift or real - space ) properly into account , as we have explained in [ formalism ] .
nonetheless , we will attempt to probe into the regime of multiple - streaming by considering the effect of caustics on the high @xmath12 tail of the pdf in [ smoothing ] .
how good is the za - predicted pdf as a description of the true pdf ?
we show in fig .
[ simpdfv ] a comparison of the za pdf versus the pdf measured from an @xmath117 cdm simulation , for two different output times , smoothed at the same comoving scale .
the linear cdm power spectrum has @xmath119 , and the box size is @xmath120 , simulated on a @xmath121 grid .
the reader is referred to the hydra consortium ( http://coho.astro.uwo.ca/pub/data.html ; see also ) for further details of the simulation .
the one free parameter we have to fix for the za prediction is the linear fluctuation amplitude @xmath6 .
it can be seen that , as expected , @xmath6 is always smaller than the fluctuation amplitude of the evolved - and - smoothed density field , which we call @xmath122 .
we also show lognormal distributions which reproduce the evolved fluctuation amplitude of the simulation outputs .
in other words , @xmath123 where @xmath124 is chosen to equal @xmath125 .
it is impressive how well it fits the n - body data out to large @xmath12 s .
the same phenomenon has been observed in real - space for a similar cosmological model ( ) . the error - bars are estimated by dividing the box into 4 different subsamples , and computing the dispersion .
it should be emphasized that we have purposefully chosen outputs , with suitable smoothing , that are only mildly nonlinear , so that the average number of lagrangian streams is very close to @xmath76 , making eq .
( [ pz ] ) a good approximation .
the overall agreement between the za pdf and the n - body data is decent at these low values of @xmath126 .
however , the za pdf tends to overpredict the probability of high @xmath12 .
it is in fact well - known that the za pdf has a long tail of @xmath127 , due to the presence of caustics .
this is true in both real and redshift - space .
it implies that , unfortunately , @xmath126 has to be left as a free parameter in the above comparison because the za pdf does not yield a well - defined @xmath122 .
we will revisit this issue in [ smoothing ] .
the methodology outlined in [ method ] can be easily adopted to compute other one - point statistics , such as @xmath1 : @xmath128 as in the case of @xmath122 , because of the caustic - induced tail of the za pdf , @xmath1 is , strictly speaking , undefined except in the limit of vanishing @xmath6 .
the finite number of realizations used in our monte carlo method provides a natural regularization by truncating at some high @xmath12 s , thereby giving a finite @xmath1 .
as one decreases the input parameter @xmath6 in the monte carlo integration , the output @xmath1 converges eventually to the correct za predicted value for infinitesimal @xmath6 .
our results are summarized in table [ s3tab ] ( see also e.g. for the real - space za prediction ) .
we find that the rate of convergence for @xmath5 , @xmath129 , etc becomes progressively slower , and so we only give the @xmath1 values here .
it can be seen that peculiar motion tends to slightly increase @xmath1 over its real - space value , as the reader might have guessed from the pdfs shown in fig .
[ zapdf ] .
this behavior is consistent with what is found by hivon et el .
@xcite in the case of the exact dynamics , in the small @xmath6 limit .
however , as is well - known in the case of real - space , in the perturbative limit , the za systematically gives lower @xmath1 than the exact dynamics . as emphasized by hivon et al .
@xcite , the perturbative prediction breaks down even at relatively small @xmath6 s ( say @xmath130 ) , or large smoothing scales , and the actual n - body @xmath1 is generally _ smaller _ in redshift - space than in real - space .
they attributed it to some kind of finger - of - god effect on large scales , which we view as in part due to multiple - streaming in redshift - space ( that is distinct from the familiar small - scale finger - of - god ) .
unfortunately , the za yields a meaningful skewness only in the vanishing @xmath6 limit , making it difficult to investigate the impact of these effects on @xmath1 . why does the za pdf predict a long high-@xmath12 tail , and why is it not observed in n - body simulations , at least for cdm - type models ?
we believe the answer to the latter at least in part has to do with smoothing due to finite resolution or discreteness as well as small - scale velocity dispersions . but let us first briefly review where the high @xmath12 tail comes from . the reader is referred to shandarin & zeldovich @xcite for a review .
most of our arguments below apply equally well to real - space caustics .
a caustic is located at a point where @xmath12 or @xmath131 diverges . as zeldovich @xcite originally argued , this is generically due to the vanishing of only one of the eigenvalues of the matrix @xmath132 .
hence , we can focus our attention on the direction that aligns with the eigenvector associated with the vanishing eigenvalue .
we will denote the relevant one - dimensional coordinates simply by @xmath133 and @xmath134 .
the situation is depicted in fig .
[ qsmap ] .
taylor expanding @xmath133 as a function of @xmath134 around @xmath135 ( corresponding to @xmath136 where the caustic is ) , one can see that to the lowest order , @xmath133 is a quadratic function of @xmath134 i.e. @xmath137 in other words , @xmath12 should fall off as @xmath138 or @xmath139 , around the caustic .
note that , strictly speaking , @xmath12 should be given by the sum over the contribution from each multiple - stream around the caustic ( as in the case depicted in fig .
[ qsmap ] ) , but the singularity due to @xmath138 or @xmath139 dominates .
note also that this singular behavior of @xmath12 occurs only on one side of the caustic in @xmath133-space . to understand the caustic s influence on the pdf ,
the crucial point to remember is that one has equal probability of locating at any position in @xmath133-space with respect to the location of the caustic .
therefore , @xmath140 , with @xmath141 independent of @xmath133 , from which the @xmath127 behavior follows .
the premise is that asymptotically high @xmath12 s occur only around caustics .
we believe the answer lies in final ( as opposed to initial or lagrangian ) smoothing .
there are two sources of final smoothing : one due to finite resolution or the discrete nature of the simulations ( or observations ) , and the other due to smearing by small - scale velocity dispersions .
two conditions have to be met for the caustic - induced high @xmath12 tail to survive these kinds of inevitable smoothing in @xmath133-space .
first , it is important that the scale of the caustic in @xmath133-space , @xmath142 , is larger than the smoothing scale , @xmath143 , where the smoothing scale can be taken to be whichever is dominant of the two kinds of smoothing mentioned above @xmath144 .
the smoothing scale @xmath145 is associated with the finite final resolution , and @xmath146 is the velocity dispersion along the line of sight .
the scale of the caustic in @xmath133-space is related to its counterpart in @xmath134-space by eq .
( [ sqc ] ) : @xmath147 where @xmath148 is the scale of the caustic in @xmath134-space . a good definition for @xmath148 is given by the magnitude of @xmath149 such that the next higher order term ignored in eq .
( [ sqc ] ) becomes significant .
in other words , @xmath150^{-1 } [ \partial^2 s / \partial q^2]$ ] .
one can use even higher order terms for its definition , but it would not affect our arguments below . the second condition has to do with the density profile of a smoothed caustic .
assume that the first condition above is met , the smoothed profile of the caustic would be given by @xmath152 where @xmath153 is some dimensionless profile which , at its peak , is equal to @xmath154 .
this is a very generic prediction quite independent of the form or origin of the smoothing .
this means the smoothed caustic peaks at @xmath155 .
now , if one would like such a smoothed caustic to contribute to the high @xmath12 tail , it is important that @xmath156 .
hence , the second condition for the survival of the caustic - induced feature in the pdf is : @xmath157 therefore , the only remaining question is what the typical @xmath159 should be .
this can be estimated using the definition suggested before : that @xmath148 is given by @xmath160^{-1}$ ] @xmath161 $ ] evaluated at the caustic .
one can calculate exactly what the average value of this ratio should be , for gaussian initial conditions , imposing the constraint that the first derivative vanishes .
all one needs is the joint probability distribution of all the derivatives up to the third one _ at a single point_. however , there is no need to do any calculation .
the only relevant scale in this problem is the _ initial _ smoothing scale of the density field ( or the displacement field ) in @xmath134-space , let us call it @xmath162
. therefore @xmath163 . for cdm - type power spectra ,
the initial smoothing scale @xmath164 is of the order of a few grid - spacings ( unless the initial power spectrum is deliberately smoothed or truncated ) , while the final smoothing scale @xmath165 is at least also a few grid - spacings ( and probably even larger due to small - scale velocity dispersions ) .
this implies the condition in eq .
( [ condition3 ] ) is never met , hence explaining why the theoretical caustic - induced tail of the pdf is not observed , at least in cdm simulations . in other words , to understand the high-@xmath12 behavior of the pdf properly , both multiple - streaming and smoothing have to be properly taken into account .
this also points to a weakness of the calculation presented in [ formalism ] .
because final smoothing in @xmath133-space is not taken into account in the formalism , effects like what is discussed above are missed .
we hope to pursue improvement of the calculation in this direction in the future . using the zeldovich approximation
, we find that peculiar motion significantly raises the degree of multiple - streaming .
we quantify this by computing the average number of lagrangian - streams in both real and redshift - space . as is shown in fig .
[ avestream ] , multiple - streaming can be insignificant in real - space , but non - negligible in redshift - space , even at moderate values of @xmath6 .
this implies that most of the multiple - streaming in redshift - space arises from the @xmath166 mapping ( real or eulerian to redshift - space ) , rather than from the @xmath167 mapping ( lagrangian to eulerian space ) . that the level of nonlinearity is amplified by the real to redshift - space mapping implies non - perturbative effects have to be taken into account in the calculation of redshift distortions , even on scales where the real - space density field is relatively linear .
an interesting feature of redshift - space multiple - streaming is that redshift - space multiple - streams can freely flow past each other , unlike their real - space counterparts .
this counteracts the familiar large scale compression of structures along the redshift direction first pointed out by kaiser ( 1987 ) , but it is physically distinct from the stretching of structures due to virialized motions on small scales .
this might offer an explanation of why , even on relatively large scales , linear theory overpredicts the amount of reshift - space squashing of the two - point function along the line of sight . to make this quantitative , however
, requires a detailed calculation , which we will discuss briefly below .
we have computed also using the za the probability distribution function ( pdf ) of density , as well as @xmath1 , in both real and redshift - space . at the largest scales , or small @xmath6
limit , @xmath1 does not appear to change significantly from real to redshift - space , consistent with the finding of hivon et al .
we have compared the za pdf with that measured from a cdm n - body simulation ( fig [ simpdfv ] ) .
we find that the pdf in redshift - space is well - fit by a lognormal distribution , similar to its counter - part in real - space ( ) . this suggests the pdf in both real and redshift - space can be described by a single - parameter family of functions ( @xmath124 in eq .
[ [ plognorm ] ] ) .
this is certainly worth exploring with more n - body simulations and different cosmological models ( see ) .
the simulation pdf reveals no caustic - induced high @xmath12 tail , contrary to the za prediction .
the same has been observed before in real - space as well ( ) .
we argue in [ smoothing ] that this is at least in part due to the fact that final smoothing in redshift - space ( or real - space ) is not taken into account in our formalism .
smoothing naturally arises because of finite resolution or discreteness , as well as due to small - scale velocity dispersions .
we give the criterion for the survival of the caustic feature through smoothing in eq .
( [ condition3 ] ) : the final - smoothing scale has to be smaller than the lagrangian - smoothing scale . in practice , we expect caustics to be efficiently erased by final - smoothing , unless the initial power spectrum is sufficiently smoothed or truncated .
this is reminiscent of a related situation in gravitational lensing : that caustic features can only be observed if the source - size is sufficiently small .
as mentioned before , a proper treatment of the statistics of the density field in redshift - space requires tackling the issues of multiple - streaming and final - smoothing .
while such a calculation is beyond the scope of the present paper , taylor & hamilton @xcite and fisher & nusser @xcite have already made significant progress in using the za to compute the quadrupole - to - monopole ratio of the redshift - space power spectrum .
they use the following expression for the fourier transform of the redshift - space density : @xmath168 where @xmath169 as a function of @xmath37 is the displacement from the lagrangian position @xmath37 to the redsfhit - space position @xmath170 .
it comes from the following basic equation for the redshift - space density : @xmath171 , which implicitly allows for multiple - streaming by automatically summing over density contributions from all possible streams .
the key here is that the mapping from real to redshift - space is treated exactly , instead of perturbatively , hence the non - perturbative nature of the redshift - space density field is taken into accout even if the real - space density field is still quite linear .
these authors find that the za predicts the right - shape for the quadrupole to monopole ratio but underestimates the zero - crossing scale ( see also hatton & cole ) .
one can take their treatment one step further by using approximations that are closer to the exact dynamics ( see e.g. scoccimarro ) .
whatever the approach , it appears likely that , because of the prominence of multiple - streaming in redshift - space , a pure perturbative calculation would be inadequate .
non - perturbative effects such as those due to multiple - streaming ( or even pre - multiple - streaming ) have to be taken into account explicitly .
lastly , our lagrangian analysis here still leaves open the question of how the local pancakes or caustics are spatially distributed globally , whether they preferentially lie along filaments or sheets .
calculations have shown that filamentary structures in real - space are preferred for gaussian initial conditions ( e.g. , ) .
it remains to be seen how the redshift - space mapping might alter this picture .
our local calculation , however , does provide a valid description of the global distribution of structures if the initial power spectrum is sufficiently smoothed , as is done in the truncated zeldovich approximation .
lh is grateful to roman scoccimarro for many useful discussions and help with the simulations .
the cdm simulations analyzed in this work were obtained from the data bank of n - body simulations provided by the hydra consortium ( http://coho.astro.uwo.ca/ pub / data.html ) and produced using the hydra n - body code ( ) .
lh thanks dick bond and the canadian institute for theoretical astrophysics for hospitality and an excellent working environment .
this work was in part supported by the doe and the nasa grant nag 5 - 7092 at fermilab .
ss acknowledges the support of epscor 1998 grant , grf grant at uk , and thanks the canadian institute for theoretical astrophysics for hospitality . | we examine several aspects of redshift distortions by expressing the redshift - space density in terms of the eigenvalues and orientation of the local lagrangian deformation tensor .
we explore the importance of multiple - streaming using the zeldovich approximation ( za ) , and compute the average number of streams in both real and redshift - space .
it is found that multiple - streaming can be significant in redshift - space but negligible in real - space , even at moderate values of the linear fluctuation amplitude ( @xmath0 ) . moreover , unlike their real - space counter - parts , redshift - space multiple - streams can flow past each other with minimal interactions .
such nonlinear redshift - space effects , which are physically distinct from the fingers - of - god due to small - scale virialized motions , might in part explain the well - known departure of redshift distortions from the classic linear prediction by kaiser ( 1987 ) , even at relatively large scales where the corresponding density field in real - space is well described by linear perturbation theory .
we also compute using the za the probability distribution function ( pdf ) of density , as well as @xmath1 , in real and redshift - space , and compare it with the pdf measured from n - body simulations .
the role of caustics in defining the character of the high density tail is examined .
it is found that ( non - lagrangian ) smoothing , due to both finite resolution or discreteness and small - scale velocity dispersions , is very effective in erasing caustic structures , unless the initial power spectrum is sufficiently truncated .
ifundefinedchapter # 1 |
clusters of galaxies can be detected in the x - ray band due to the emission of the intracluster ( ic ) gas .
this gas is hot enough to change the brightness of the cosmic microwave background ( cmb ) photons through inverse compton scattering ( zeldovich & sunyaev 1969 ) .
the sunyaev - zeldovich ( sz ) effect caused by individual clusters has been measured for tens of clusters ( see birkinshaw 1999 for a review ) .
eventually , the planck satellite will produce an all - sky catalogue likely to contain thousands of sz sources ( da silva et al .
1999 ) . together with the effect of single sources
, the overall cluster population induces distortions and temperature anisotropies on the cmb .
the sunyaev - zeldovich effect has a well known frequency dependence that helps to distinguish it from other foregrounds and methods have already been proposed to separate and measure this contribution ( hobson et al .
1998 ) .
extensive theoretical work has been devoted to analyze the effect of clusters on the cmb radiation ( cole & kaiser 1988 , bartlett & silk 1994 , colafrancesco et al .
1994 , atrio - barandela & mcket 1999 -hereafter paper i- , komatsu & kitayama , 1999 ) .
barbosa et al .
( 1996 ) noticed that the value of the mean comptonization parameter , that measures the amplitude of the blackbody distortion , depends on which are the less massive clusters that produce a significant effect . on the other hand
, cole & kaiser ( 1988 ) remarked that temperature anisotropies were dominated by massive clusters at moderate redshifts .
since clusters of a given mass contribute differently to temperature anisotropies and distortions , in this letter we show how the number density of clusters with enough hot gas to produce a detectable sz effect can be estimated .
this number is marginally dependent on the cluster model and it varies by at most a factor of four in different cosmologies .
following paper i , we assume that clusters are spherical in shape , with typical size the virial radius @xmath3 , virial mass @xmath4 and have virialized at redshift @xmath5 .
we also assume that the ic gas is isothermal , distributed smoothly and with a radial distribution well fitted by the spherical isothermal @xmath6 model @xmath7^{-3\beta/2}$ ] ( jones & forman 1984 ) , where @xmath8 are the central electron density and the core radius of the cluster , respectively . for convenience
, we shall assume that @xmath9 .
similar results to those presented here are obtained for any value of @xmath6 within the observed range ( @xmath10 , see markevitch et al .
further , we assume that the dynamical evolution of the ic gas is well described by the entropy - driven model by bower ( 1997 ) since this model seems to provide an adequate description of clusters with redshifts @xmath11 ( mushotzky & scharf 1997 ) . to compute the spectral distortion and temperature anisotropies induced by hot ic gas we need to translate the properties of a sample of clusters at low redshifts into their equivalents at earlier epochs .
for the spherical collapse model , the virial radius scales with mass and redshift as @xmath12 . if the electron temperature is proportional to the velocity dispersion of the dark matter then @xmath13 .
in the entropy - driven model , the electron central density scales as @xmath14 . finally , the core radius scales as @xmath15 . in this scaling relations , @xmath16 are the current average core and virial radius , the central gas temperature and electron density of a @xmath17m@xmath1 cluster , respectively , while @xmath18 parametrizes the rate of core entropy evolution .
mushotzky and scharf ( 1997 ) found @xmath19 .
the effect of the ic gas is both to distort the blackbody spectrum and to induce temperature anisotropies on the cmb .
a measure of such distortion is the mean comptonization parameter defined as @xmath20 in this expression @xmath21 gives the dependence of the sz effect with frequency @xmath22 ; @xmath23 is the cmb mean temperature , @xmath24 is the cluster number density per unit of mass , @xmath25r_c t_g n_c$ ] with physical constants having their usual meaning and @xmath26 gives the probability that a particular line of sight crosses a cluster , with @xmath27 the angular distance to the cluster . if we introduce @xmath28 then @xmath29 is the averaged line of sight through a cluster .
the effect of all clusters on an angular scale @xmath30 can be obtained by adding in quadrature the contribution of each single cluster .
if we assume that clusters are poisson distributed on the sky the radiation power spectrum is ( cole & kaiser 1988 ) @xmath31 where @xmath32 is the fourier transform of the angular profile of the ic gas .
this expression represents the contribution of the foreground cluster population to the power spectrum of cmb temperature anisotropies .
the variance of the temperature field is given by @xmath33 in this equation , @xmath34 represents the window function of the experiment .
this temperature anisotropy , together with other secondary contributions , adds in quadrature with the `` intrinsic '' ( cosmological ) signal to give the pattern of temperature anisotropies on the sky ( see paper i for details ) . to compute the number density of clusters as a function of mass and redshift we use the press - schechter approximation , known to describe accurately the abundance of clusters in numerical simulations at different redshifts @xmath35 in this expression
, @xmath36 is the peak - height threshold , @xmath37 is the rms mass fluctuation within a top - hat filter measured in units of @xmath38 , the rms mass fluctuation on a sphere of @xmath39mpc today , @xmath40 is the threshold overdensity of the spherical collapse model and @xmath41 is the bias factor .
we integrate numerically eqs . ( [ ycmean ] ) and ( [ dtt ] ) for three cosmological models with the same baryon fraction @xmath42 : two flat models , ( 1 ) standard cdm , with @xmath43 , ( 2 ) @xmath44cdm with @xmath45 and @xmath46 as indicated by the recent measurements of high redshift supernovae ( perlmutter et al .
1999 ) and ( 3 ) an open model , ocdm , with @xmath45 and no cosmological constant , consistent with cluster dynamics and evolution ( bahcall 1999 ) . as average parameters , for a @xmath17m@xmath1 cluster we took : @xmath47mpc , @xmath48k , @xmath49mpc and @xmath50@xmath51 .
we considered different gas evolutionary histories , by taking @xmath52 .
the matter power spectrum was normalized to @xmath53mpc@xmath54 ( einasto et al .
1999b ) and we considered the following bias factors : @xmath55 .
the first corresponds to the bias of qdot galaxies and the second is an average value obtained from several catalogues ( einasto et al 1999a ) .
also , we considered @xmath56 ( tozzi & governato , 1998 , governato et al .
the simplicity of the press - schechter approach lies on the fact that the number density depends only on the ratio @xmath57 . with the values quoted above
, @xmath58 varies in the range 2 to 3 and we shall quote our results for both limits . to compute the effect of the cluster population on the cmb radiation it is necessary to determine first if groups of galaxies of about @xmath59m@xmath1 have enough hot gas to produce a significant contribution .
the scaling relations of sec . [ 2 ] have been found to be accurate for clusters above @xmath0m@xmath1 .
however , they could very well be extended below this mass scale . in this respect , it is necessary to specify the integration range of eqs .
( [ ycmean ] ) and ( [ pq ] ) .
we tried three different lower mass limits : @xmath60 and @xmath61 ( in units of @xmath0m@xmath1 ) .
the main conclusion was that when changing the mass limit one order of magnitude @xmath62 could vary a factor 3 to 10 , depending on the model , but the temperature anisotropy @xmath63 varied by less than 30% . in this calculations and in the rest of the paper , temperature anisotropies
were computed using the window function of the suzie experiment ( church et al .
for the models considered above , the anisotropies in the rayleigh - jeans regime ranged from @xmath64k for scdm with @xmath65 to @xmath66k for @xmath44cdm with @xmath67 .
integration of eq ( [ pq ] ) shows that @xmath68 reaches a maximum around @xmath69 . at those scales
komatsu & kitayama ( 1999 ) have concluded that the increase in temperature anisotropy induced by cluster correlations is negligible . to illustrate the different behavior of temperature anisotropies and distortions with the lower mass integration limit we shall consider the behavior of the radiation power spectrum at @xmath70 .
this scale is close to the maximum of the sz effect and is a good compromise between the smallest angular scale that will be resolved by planck and the maximum of the suzie window function . in fig .
1 we show @xmath71 for different cosmological models . notice that the integrand of @xmath62 increases at the low mass end , but that of the radiation power spectrum reaches a maximum around @xmath0m@xmath1 .
its exact location is weakly dependent on model parameters . a qualitative explanation of fig .
1 can be given : since @xmath72 on a mass scale of @xmath73m@xmath1 , in the upper mass limit @xmath74 and @xmath75 are damped by the exponential factor in the press - schechter formula ; in the low mass limit , the exponential factor is close to unity and @xmath76 but @xmath77 where @xmath78 is the slope of the matter power spectrum at @xmath59m@xmath1 , roughly @xmath79 ( einasto et al .
1999a ) . to conclude , the gas in groups contributes significantly to comptonization but little to temperature anisotropies .
this fact can be used to estimate the number density of clusters that produce a significant sz effect . for this purpose , in paper i we introduced the parameter @xmath80 .
it represents a relation similar to the approximate solution of the full kinetic equation for the change of photon distribution due to inverse compton scattering for a single cluster : @xmath81 . as an example , in fig .
2a we plot @xmath82 with respect to lower mass integration limit for three cosmological models with @xmath83 : solid line ( scdm ) , dot - dashed line ( ocdm ) and dashed line ( @xmath44cdm ) .
thick lines correspond to @xmath84 and thin lines @xmath85 .
similar behavior can be observed for other gas evolution histories and model parameters . by inspection of eqs .
( [ ycmean ] ) and ( [ pq ] ) , the comptonization parameter and the power spectrum scale linearly with the number density of clusters . therefore
, one expects that @xmath82 will scale as @xmath86 . in fig .
2b we plot @xmath87 for the same models as in fig .
notice that the value of @xmath87 is almost constant and nearly independent on the threshold @xmath58 .
the actual value varies by a factor of two when the total matter content changes from @xmath88 to 0.3 but does not depend on the geometry of the cosmological model , i.e. on the value of @xmath44 . in fig .
3 we plot @xmath82 as a function of gas evolution history , assuming that the number density of all the objects contributing to the sz effect is @xmath89mpc@xmath90 . in each panel ,
thick lines correspond to a threshold @xmath84 and thin lines to @xmath85 .
solid lines correspond to a lower mass integration limit of @xmath0m@xmath1 and dot - dashed lines to @xmath59m@xmath1 .
notice that @xmath82 varies by less than @xmath91 for different gas evolutionary histories .
let as remark that , if @xmath92 is known , @xmath82 permits to obtain @xmath93 . vice versa , if @xmath93 could be determined by other means , @xmath2 could be determined .
if no assumption is made about the cosmology , @xmath93 can be determined within a factor of four .
the results of previous section indicates that @xmath82 is mostly sensitive to the number density of clusters , and for a given cosmological model is weakly dependent on the exact modeling of the cluster population .
therefore , a measurement of both the mean distortion of the cmb radiation and the temperature anisotropies induced by clusters of galaxies will permit to estimate the number density of objects that have enough hot gas to produce a detectable effect on the cmb . this conclusion is not limited by the validity of the scaling relations of sec . [ 2 ] .
one must expect a similar behavior for a different cluster model , even though the level of temperature anisotropies and distortions will be different .
if the hot ic gas was uniformly distributed in the universe , it would still produce blackbody distortions but will not give rise to temperature anisotropies , and @xmath82 would be zero .
only when the gas is clumped in clusters , two different lines of sight , seeing different column densities , will have different temperatures .
therefore , @xmath62 scales with the number of clusters along the line of sight while for poisson distributed clusters @xmath94 scales as the square root and @xmath95 , as shown .
therefore , independently of the exact model used to describe the average cluster properties , @xmath82 measures how the hot gas is distributed in the universe . finally , if the cluster number density is known , @xmath82 could be used as an estimator of the total matter density @xmath2 . to summarize ,
clusters of @xmath0m@xmath1 mass dominate the contribution to temperature anisotropies .
therefore , even if the adequacy of the scaling relations of sec .
[ 2 ] below that mass scale has not been established , temperature anisotropies can be reliably calculated using eqs .
( [ pq ] ) and ( [ dtt ] ) .
if the entropy - driven model proves not to be adequate to describe groups of galaxies , one would have to use hydrodynamical simulations such as those of da silva et al .
( 1999 ) to compute @xmath62 but @xmath82 would still provide an estimate of the number density of sz clusters .
comparison of this density with that of x - ray clusters will undoubtably help our understanding of cluster formation and evolution .
* acknowledgments . *
we thank the referee , j. einasto , for very constructive comments and suggestions .
this research was supported by spanish german integrated actions ha 97/39 .
fab would like to acknowledge the support of the junta de castilla y len , grant sa40/97 .
atrio barandela , f. & mcket , j. 1999 , , 515 , 465 ( paper i ) bahcall , n. a. 1999 in `` proceedings of the american astronomical society meeting '' , 194 , 3002 barbosa , d. , bartlett , j. g. , blanchard , a. , oukbir , j. 1996 , a & a , 314 , 13 bartlett , j.g . ,
silk , j. 1994 , , 423 , 12 birkinshaw , m. 1999 , phys . rep .
, 310 , 97 bower , r.g .
1997 , mnras , 288 , 355 colafrancesco , s. , mazzotta , p. , rephaeli , y. & vittorio , n. 1994 , , 433 , 454 cole , s. & kaiser , n. 1989 , mnras , 233 , 637 church , s. e. et al .
1997 , 484 , 523 da silva , a. c. , barbosa , d. , liddle , a. r. & thomas , p. a. 1999 , preprint astro - ph/9907224 einasto , j. et al .
1999a , , 519 , 441 einasto , j. et al . 1999b , , 519 , 456 governato et al .
1999 , mnras , submitted .
preprint astro - ph/9810189 hobson , m. p. , jones , a. w. , lasenby , a. n. , bouchet , f. r. 1998 , , 300 , 1 jones , c. & forman , w. 1984 , , 276 , 38 komatsu , e. & kitayama , t. 1999 , preprint astro - ph/9908087 markevitch , m. , forman , w.r . , sarazin , c.l .
& vikhlinin , a. 1997 , , 503 , 77 mushotzky , r.f . & scharf , c.a .
1997 , , 482 , l13 perlmutter , s. et al .
1999 , , 517 , 565 press , w.h . , & schechter , p. 1974 , , 187 , 425 tozzi , p. & governato , f. 1998 , in `` the young universe '' s. dodorico , a. fontana , and e. giallongo , eds .
asp conference series , 146 , 461 zeldovich , ya .
b. , sunyaev , r. 1969 , , 4 , 301 | if the mean properties of clusters of galaxies are well described by the entropy - driven model , the distortion induced by the cluster population on the blackbody spectrum of the cosmic microwave background radiation is proportional to the total amount of intracluster gas while temperature anisotropies are dominated by the contribution of @xmath0m@xmath1 clusters .
this result depends marginally on cluster parameters and it can be used to estimate the number density of clusters with enough hot gas to produce a detectable sunyaev - zeldovich effect . comparing different cosmological models , the relation depends mainly on the density parameter @xmath2 .
if the number density of clusters could be estimated by a different method , then this dependence could be used to constrain @xmath2 . |
a significant property of clusters of a normal phase embedded in a superconducting medium consists in their capability to trap a magnetic flux and hold the vortices from moving under the action of the lorentz force . as a result
, these clusters can act as effective pinning centers @xcite- @xcite .
this property is widely used in creating composite superconducting materials with high critical current values @xcite , @xcite .
the morphologic characteristics of the normal phase clusters essentially affects the dynamics of trapped magnetic flux , especially if the clusters have fractal boundaries @xcite-@xcite . in the present work we consider in detail the geometric probability properties of these fractal clusters and their influence on the critical current and dynamics of trapped magnetic flux near the transition of the superconductor into a resistive state
let us consider a superconductor containing fragments of a normal phase .
suppose that dimension of these fragments along one direction significantly exceed the other sizes .
similar columnar defects are of great interest in creating artificial pinning centers @xcite , @xcite-@xcite . if such a superconducting structure is cooled in a magnetic field directed along the axis of alignment of these defects below the critical temperature , then the distribution of magnetic flux trapped in the clusters of a normal phase will be two - dimensional . this can be done especially easily with a superconducting film where such clusters are formed near defects at the boundary with the substrate during the growth process and are directed transversely to the film plane @xcite , @xcite .
let us suppose that the film surface fraction covered by the normal phase is below the percolation threshold for the transfer of magnetic flux ( i.e. , less than 50% for 2d - percolation @xcite ) . in this case
the fraction of the superconducting phase exceeds the percolation threshold , so there is a superconducting percolation cluster that can carry a transport current in the film plane .
this kind of structure provides for the effective pinning and thereby raises the critical current , for the magnetic flux is trapped in isolated clusters of the normal phase and the vortices can not leave them without crossing the surrounding superconducting space . as the current increases ,
the trapped flux will remain unchanged until the vortices begin to break away from those clusters for which the pinning force is less than the lorentz force caused by the transport current .
when the magnetic flux is breaking away from the pinning centers , each vortex must cross the infinite superconducting cluster . in this case
the vortices will move primarily along the weak links that connect the clusters of a normal phase to each other @xcite , @xcite-@xcite .
these weak links form readily in high - temperature superconductors ( hts ) characterized by a small coherence length @xcite , @xcite .
structural defects , which could act simply as scattering centers at a large coherence length , create weak links in hts .
there is a large variety of weak links in a wide range of spatial scales in hts @xcite , @xcite-@xcite . at an atomic level
weak links are created by structural point defects , primarily by oxygen vacancies @xcite , @xcite . on a mesoscopic scale weak links are efficiently formed on twin boundaries @xcite , @xcite , @xcite , @xcite .
twins can be located at a distance of several tens of nanometers from each other , therefore even a single crystal can have a fine substructure created by twins .
finally , on a macroscopic scale weak links are created by various structural defects , such as boundaries of grains and crystallites or barriers arising from the secondary degrading of a non - stoichiometric crystal into the domains with a high and low oxygen content @xcite-@xcite . moreover
, a magnetic field favors the coherence length to further decrease @xcite , so weak link formation is facilitated still more . in conventional low - temperature superconductors , which are of large coherence length
, weak links are formed due to the proximity effect at sites where the distance between neighboring clusters of a normal phase is minimum .
thus , regardless of their origin , weak links form channels for vortex transport where the pinning force is less than the lorentz force caused by the transport current . according to their configuration each cluster of the normal phase
has its own value of the depinning current that makes the contribution to the overall statistical distribution of the critical currents .
when a transport current is passed through the sample , vortices will be broken away first from clusters of a smaller pinning force and , accordingly , a smaller critical current of depinning .
so the change in the trapped magnetic flux will be proportional to the number of the normal phase clusters which have the critical current less than the preset value @xmath0 .
therefore , the relative decrease in the flux can be expressed through the function of cumulative probability @xmath1 for the distribution of critical currents of the clusters : @xmath2 where @xmath3 the right - hand side of eq .
( [ eq1 ] ) is the probability that any @xmath4-th cluster has a the critical current @xmath5 less than the given upper bound @xmath6 . on the other hand ,
the magnetic flux trapped in a single cluster is proportional to its area @xmath7 .
therefore , the decrease in the trapped flux can be expressed through the function of cumulative probability @xmath8 for the distribution of the areas of normal phase clusters , which is proportional to the number of clusters of area smaller than a given value of @xmath7@xmath9 where @xmath10 generally , the distribution of cluster areas can be described by the gamma - distribution , for which the function of cumulative probability has the form @xmath11 where @xmath12 is the incomplete gamma function , @xmath13 is euler gamma function , and @xmath14 and @xmath15 are the parameters of the gamma - distribution that determine the mean area of the normal phase cluster @xmath16 and its standard deviation @xmath17 .
the distribution function of cluster areas can be found from the geometric probability analysis of electron photomicrographs of superconducting films @xcite , @xcite , @xcite .
for example , in the practically important case of ybco films with columnar defects @xcite the exponential distribution is realized , which is a special case of the gamma - distribution of eq .
( [ eq3 ] ) with @xmath18 . to clarify how the transport current affects the trapped magnetic flux ,
it is necessary to find the relationship between the distribution of critical currents of clusters of eq .
( [ eq1 ] ) and the distribution of their areas of eq .
( [ eq2 ] ) .
the larger the normal phase cluster size , the more weak links rise from its perimeter bordering with the surrounding superconducting space , and therefore , the smaller is the critical current at which the magnetic flux breaks away from this cluster .
let us suppose that the weak link concentration per unit length of the perimeter is the same for all clusters and all the clusters of equal perimeters have the same pinning force .
then the critical current @xmath0 is inversely proportional to the perimeter @xmath19 of the normal phase cluster : @xmath20 , because the probability to find a weak link over the larger perimeter is higher .
it is suggested here that the vortex , driven by the lorentz force , passes through the weak link from one normal phase cluster to another with the probability of 100% . in this case magnetic flux
is transferred through the superconducting medium by josephson vortices . the probability to trap the vortex
when it moves through the weak link under the action of the lorentz force is negligible , inasmuch as the josephson penetration depth in the considered materials significantly exceeds the sizes of possible irregularities along the weak link . in accordance with these assumptions , in order to find the distribution function of eq .
( [ eq1 ] ) we have to know the relationship between the perimeter and the area of a cluster . as was first established in ref .
@xcite , fractal properties of the boundaries of normal phase clusters essentially affect the dynamics of magnetic flux in superconductors . for fractal clusters the perimeter - area relation has the form @xcite @xmath21 where @xmath22 is the fractal dimension of the cluster boundary .
relation of eq .
( [ eq4 ] ) agrees with the generalized euclid theorem @xcite , according to that the ratios of the corresponding measures are equal when they are reduced to the same dimension .
so @xmath23 , and this relation holds true both for euclidean clusters ( for which the hausdorff
bezikovich dimension of the perimeter is equal to the topological dimension of a line , @xmath24 ) , and for fractal clusters ( for which the hausdorff
bezikovich dimension of the boundary strictly exceeds the topological one , @xmath25 ) .
note that it is just the statistical distribution of cluster areas , rather than their perimeters , is fundamental for deriving the critical current distribution . inasmuch as the hausdorff
bezikovich dimension of a fractal line exceeds unity , the perimeter of a fractal cluster is not well defined : it diverges when the precision of its measurement increases indefinitely @xcite . at the same time
, the topological dimension of the cluster area coincides with the hausdorff
bezikovich dimension ( both are equal to 2 ) .
therefore , the area of a surface confined by the fractal curve is a finite , well - defined quantity . analyzing the geometric characteristics of normal phase clusters
, we are considering the cross - section of extended columnar defect by the plane the transport current flows through .
therefore , though a normal phase cluster is a self - affine fractal @xcite , we can consider its geometric probability properties in the planar section only , where the boundary of cluster is statistically self - similar .
now , using relation of eq .
( [ eq4 ] ) between the fractal perimeter and area , together with the assumption of inverse proportionality of the critical current of a cluster to its perimeter , we arrive at the following expression for critical current of the cluster : @xmath26 , where @xmath27 is a form factor . taking into account our initial relations of eqs .
( [ eq1 ] ) and ( [ eq2 ] ) , we can find the distribution of critical currents in the general case of the gamma - distributed cluster areas : @xmath28 where @xmath29 @xmath30 is the complementary incomplete gamma function , @xmath31 is the dimensionless electric current , and @xmath32 is the critical current of the transition into a resistive state . the found function of cumulative probability of eq .
( [ eq5 ] ) provides a complete description of the effect of the transport current on the trapped magnetic flux . using this function
, we can easily derive the probability density @xmath33 for the distribution of depinning currents , which has the form @xmath34 the probability density is normalized to unity over all possible positive values of the critical currents .
the relative change in the trapped flux @xmath35 , which can be directly found from equation ( [ eq5 ] ) , also determines the density of vortices @xmath36 broken away from pinning centers by the current of a given magnitude @xmath37@xmath38 where @xmath39 is the magnetic field , and @xmath40 is the magnetic flux quantum ( @xmath41 is planck constant , @xmath42 is the velocity of light , and @xmath43 is the electron charge ) .
figure [ figure1 ] shows how @xmath15-parameter of the gamma - distribution affects the distribution of the critical currents of the clusters . as an example , we have taken a value of the fractal dimension @xmath44 , which is near to the value @xmath45 found earlier @xcite for the normal phase clusters in ybco film structures . on the other hand ,
the value @xmath44 is intermediate between two limiting cases : @xmath24 for the euclidean clusters and @xmath46 for the clusters of the most fractality .
when the fractal dimension differs from unity so much , the fractal properties of the cluster structure of a superconductor are of great importance . in fig .
[ figure2 ] the cluster area distribution is presented .
the corresponding probability density has the following form @xmath47 where @xmath48 is the dimensionless cluster area , with @xmath49 and @xmath50 .
the function of cumulative probability for the dimensionless area is related with function of eq .
( [ eq8 ] ) by the formula @xmath51 and can be written as @xmath52 as is seen from fig.@xmath53[figure1 ] , when @xmath15-parameter decreases , the distribution @xmath54 spreads to the right , spanning higher and higher range of critical currents .
it might be expected that the highest current - carrying capability of the superconductor can be achieved in the limiting case of @xmath18 , when the gamma - distribution of eq .
( [ eq8 ] ) is reduced to the exponential one @xmath55 . in this case
the small clusters give the most contribution into the overall distribution ( curve 1 in fig .
[ figure2 ] ) . for @xmath18 expressions of eqs .
( [ eq5 ] ) and ( [ eq6 ] ) can be simplified @xmath56 @xmath57 where , as above , @xmath31 , and the critical current of the resistive transition can be calculated using a simpler formula : @xmath58 .
figure [ figure3 ] shows how the fractal dimension of the cluster boundaries affects the distribution of the critical currents of eq .
( [ eq10 ] ) . as is clearly seen from the figure , with increasing fractal dimension the critical current distribution spreads and shifts to the higher values of current .
this shift can be characterized by the dependence of the mean critical current on the fractal dimension , as is shown in fig .
[ figure4 ] . although the mode of the distribution _
mode_@xmath59 is linear with respect to the fractal dimension , the mathematical expectation obeys a much stronger superlinear law controlled by euler gamma function @xmath60 thus , the geometric probability properties of the normal phase clusters determine the statistical distribution of depinning currents . using this distribution along with equation ( [ eq1 ] ) , the relative change in the trapped flux under the action of the transport current can be found .
the effect of the transport current on the trapped magnetic flux is demonstrated in fig .
[ figure5 ] .
the change in the trapped flux was calculated using formula of eq .
( [ eq1 ] ) for the exponential - hyperbolic critical current distribution of eq .
( [ eq9 ] ) .
curves 1 and 5 in fig .
[ figure5 ] correspond to the limiting cases of the euclidean clusters ( @xmath24 ) and the clusters of maximum fractality ( @xmath46 ) , respectively ; they confine the region of changes in the trapped flux for all possible values of the fractal dimension .
the function @xmath61 reveals one important property ; namely , it is very flat near the coordinate origin .
it can readily be shown that all its derivatives become zero at the origin point : @xmath62 for any values of @xmath63 .
therefore even the taylor series expansion of the function in the vicinity of the origin converges to zero value but not to the quantity @xmath64 itself .
this mathematical feature has a clear physical sense : such a small transport current does not affect the trapped magnetic flux , because there are no pinning centers with such small critical currents in the total statistical distribution .
the decrease in the magnetic flux becomes noticeable only near the point of a resistive transition ( @xmath65 ) . a use of the exponential - hyperbolic distribution of critical currents of eq .
( 9 ) excludes the uncertainty caused by truncation of non - physical negative values of the depinning currents , which may happen , e.g. , in the case of the normal distribution @xcite , @xcite . as is seen from fig .
[ figure5 ] , the breaking of the vortices away is observed mainly at @xmath66 , when the sample undergoes a transition into a resistive state .
figure [ figure5 ] demonstrates the practically important property of a superconducting structure with fractal normal phase clusters : the fractality favors magnetic flux trapping , and thereby increases the critical current magnitude below which the sample remains in a superconducting state .
indeed , the transport current of the value @xmath67 causes the 43% of the total trapped magnetic flux to break away from the usual euclidean clusters ( @xmath24 , curve 1 ) , whereas this value equals only to 13.5% for the fractal normal phase clusters of the greatest possible fractal dimension ( @xmath46 , curve 5 ) .
this is equivalent to the pinning increase of 218% in the latter case .
the enhancement of pinning due to fractality can be characterized by the pinning gain factor @xmath68 which is equal to the relative decrease in the fraction of the magnetic flux broken away from the fractal clusters of dimension @xmath22 in comparison with the euclidean ones ( @xmath24 ) . in fig.@xmath53[figure6 ] the dependencies of the pinning gain factor on the transport current and on the fractal dimension are presented .
the maximum pinning gain is achieved when the boundaries of clusters have the greatest possible fractal dimension ( @xmath46 ) .
note that the pinning gain factor characterizes the properties of the superconductor in the range of transport currents corresponding to a resistive state ( @xmath66 ) . with smaller currents , the trapped magnetic flux virtually does not change , because there are no pinning centers with such small critical currents ( figs .
[ figure1 ] , [ figure3 ] ) and the breaking of the vortices away has not started yet . in the presence of a finite resistance any current flow
is accompanied by energy dissipation . as for any hard superconductor ( type - ii , with pinning centers ) the energy dissipation in a resistive state
does not mean the phase coherence destruction yet .
some dissipation accompanies any motion of the magnetic flux ; this effect can be observed in a hard superconductor even for small transport currents .
therefore , the critical current in such materials can not be determined as the greatest dissipationless current .
the superconducting state collapses only when the dissipation increases in an avalanche - like manner as a result of the thermo - magnetic instability .
the reason for the pinning gain caused by the fractality of the cluster boundaries lies in the fundamental properties of the statistical distribution of critical currents ( see fig . [ figure4 ] ) . in the case of euclidean clusters the mean value of the critical current calculated in accordance with eq .
( [ eq11 ] ) is equal to @xmath69 , whereas for the clusters of the most fractality ( @xmath46 ) it is infinitely high . as is seen from fig .
[ figure3 ] , when the fractal dimension increases , the contribution of the clusters with high depinning currents to the overall statistical distribution increases , too , which results in the enhancement of magnetic flux trapping . in a resistive state a hard superconductor can be adequately described by its voltage - current characteristics . using the fractal distribution of the critical currents of eq .
( [ eq6 ] ) , we can find the electric field caused by magnetic flux motion after the vortices have been broken away from pinning centers .
inasmuch as each cluster of the normal phase contributes to the overall distribution of the critical currents , the voltage across the superconductor @xmath70 is a response to the sum of effects made by contribution from each cluster .
this response can be expressed as a convolution integral @xmath71 where @xmath72 is the flux flow resistance .
this representation for the voltage across the sample is often used to consider the pinning of bunches of vortex lines in a superconductor @xcite , as well as to analyze the critical scaling of voltage - current characteristics @xcite , i.e. , in all the cases when the distribution of depinning critical currents occurs . in the following analysis
we will focus on the results of the properties of the exponential - hyperbolic distribution in eq .
( [ eq10 ] ) ; we will not consider questions associated with the possible dependence of the flux flow resistance @xmath72 on the transport current . in the simplest case ,
when all pinning centers have the same critical current @xmath73 , all vortices are released simultaneously at @xmath74 , and their density , in accordance with eq .
( [ eq7 ] ) , has the form @xmath75 where @xmath76 is the dirac delta function , and @xmath77 is the heaviside function .
the trapped flux would change at once by 100% in this case : @xmath78 .
thus , accordingly with equation ( [ eq12 ] ) , for the @xmath79-like distribution of the critical currents in the regime of viscous flow of the magnetic flux the voltage across the superconductor is controlled by a simple linear dependence @xmath80 . the corresponding voltage - current characteristic is shown in fig .
[ figure7 ] by the dashed line @xmath81 . in the same figure
the voltage - current characteristics of the superconductor in a still simpler approximation of the critical state model is represented , too ( the dashed curve @xmath82 ) . in this case
the response on any external action , which results in the appearance of an electric field in a hard superconductor , is the flow of the current that is equal to the critical one @xmath74 , regardless of the value of the voltage across the sample .
( the dimensionless critical current is equal to unity due to the normalization chosen above , @xmath31 ) for the fractal distribution of the critical currents the situation changes radically , because now the vortices are being broken away in a wide range of transport currents .
we shall consider the case of the exponential distribution of the cluster areas ( @xmath18 ) , where the current - carrying capability of the superconductor is maximum .
after substitution of the distribution of eq .
( [ eq10 ] ) into eq .
( [ eq12 ] ) followed by integration , the voltage across the sample can be expressed through the function of cumulative probability of eq .
( [ eq9 ] ) @xmath83 upon integration we get @xmath84 \label{eq13}\ ] ] in the limiting cases of @xmath24 and @xmath46 , equation ( [ eq13 ] ) can be simplified . for euclidean clusters ( @xmath24 ) the voltage across the superconductor has the form @xmath85
\label{eq14}\ ] ] where the complementary incomplete gamma function is expressed through the complementary error function @xmath86erfc@xmath87 . for clusters of maximum fractality ( @xmath46 ) ,
after substitution of the representation for the complementary incomplete gamma function @xmath88 , equation ( [ eq13 ] ) for the voltage takes its final form @xmath89 \label{eq15}\ ] ] where @xmath90 is the exponential integral function . in fig .
[ figure7 ] there are shown the voltage - current characteristics of a superconductor containing fractal clusters of a normal phase .
for all values of the fractal dimension there is a noticeable voltage drop starting from the transport current of @xmath65 that coincides with the value found earlier from the critical current distribution of eqs .
( [ eq5 ] ) and ( [ eq9 ] ) for the resistive transition current . equations ( [ eq14 ] ) and ( [ eq15 ] ) describe the dependencies of the voltage on the transport current for the limiting values of the fractal dimension . whatever the geometric probability properties of the normal phase clusters may be
, the voltage - current characteristics of a superconductor with these clusters will lie in the region confined by these dependencies ( curves 1 and 5 in fig .
[ figure7 ] ) .
as is seen from the figure , the fractality significantly reduces the electric field induced by magnetic flux motion . in fig .
[ figure8 ] the dependencies of the attenuation factor of dissipation @xmath91 on the transport current at various values of the fractal dimension are presented .
the decrease in the electric field with increasing fractal dimension is especially appreciable in the range of currents @xmath92 , where the pinning gain also has a maximum ( fig .
[ figure6 ] ) . both of these effects have the same nature , since their reason lies in the peculiarities of the fractal distribution of the depinning currents .
as is seen from figs .
[ figure3 ] and [ figure4 ] , an increase in the fractal dimension leads to considerable spreading of the tail of the distribution @xmath93 .
this means that more and more clusters of small sizes , which can best trap the magnetic flux , are being involved in the game . as a result ,
the density of vortices broken away from pinning centers by the lorentz force decreases , so the smaller part of magnetic flux can flow creating a lower electric field . in turn ,
the smaller the electric field , the smaller energy is dissipated when the transport current passes through the sample .
so the decrease in heat evolution , which could cause a transition into a normal state , leads to an increase in the current - carrying capability of the superconductor containing such fractal clusters .
thus , figs .
[ figure5]-[figure8 ] clearly demonstrate the most important result : the fractality of the normal phase clusters prevents destruction of the superconductivity by the transport current , and thereby enhances the current - carrying capability of the superconductor .
the fractal clusters essentially affect the dynamics of the magnetic flux trapped in the superconductor .
the crucial change of the depinning current distribution caused by increasing the fractal dimension of the clustes is the reason of this effect .
the fractality of the boundaries of the normal phase clusters intensifies pinning and slows down destruction of the superconductivity by the transport current .
this point provides principally new possibilities for increasing the critical currents of composite superconductors by optimizing their geometric morphological properties .
m. r. beasley , artificially - structured superconductors .
- in : _ percolation , localization and superconductivity _ , ed . by a. m.
goldman and s. a. wolf , nato asi series , ser.b , vol.109 , ( plenum press , new york , 1984 ) , pp.115 - 143 . | the effect of morphologic factors on magnetic flux dynamics and critical currents in percolative superconducting structures is considered .
the superconductor contains the fractal clusters of a normal phase , which act as pinning centers .
the properties of these clusters are analyzed in the general case of gamma - distribution of their areas .
the statistical characteristics of the normal phase clusters are studied , the critical current distribution is derived , and the dependencies of the main statistical parameters on the fractal dimension are found . the effect of fractal clusters of a normal phase on the electric field induced by the motion of the magnetic flux after the vortices have been broken away from pinning centers is considered .
the voltage - current characteristics of fractal superconducting structures in a resistive state for an arbitrary fractal dimension are obtained .
it is found that the fractality of the boundaries of normal phase clusters intensifies magnetic flux trapping and thereby increases the current - carrying capability of the superconductor . |
throughout this paper , all surfaces and 3-manifolds will be taken to be compact , orientable and piecewise - linear .
let @xmath0 be an orientable , irreducible @xmath1-manifold admitting a weakly reducible genus three heegaard splitting as a minimal genus heegaard splitting .
let us consider an element @xmath2 $ ] of the group of isotopy classes of orientation - preserving automorphisms of @xmath0 , say @xmath12 , and an automorphism @xmath13 in the isotopy class @xmath2 $ ] .
let @xmath14 $ ] be the isotopy class of a properly embedded ( possibly disconnected ) surface @xmath15 in @xmath0 .
since we can well - define the image @xmath2([f])$ ] as @xmath16 $ ] for an isotopy class @xmath14 $ ] and an element @xmath2\in mod(m)$ ] , if there is a correspondence @xmath2([f])=[f']$ ] between two isotopy classes @xmath14 $ ] and @xmath17 $ ] , then it would contain some information of @xmath2 $ ] even though it does not contain all information of @xmath2 $ ] .
but if @xmath15 does not divide @xmath0 into sufficiently small pieces , then one can expect that the correspondence contains not much information and if the genus of @xmath15 is large , then it would be hard to even just find a correspondence .
since @xmath0 admits a weakly reducible heegaard splitting of genus three , we can get the generalized heegaard splitting obtained by `` _
_ weak reduction _ _ '' , where it consists of two non - trivial heegaard splittings of genus two .
conversely , if there is a generalized heegaard splitting of @xmath0 consisting of two non - trivial heegaard splittings of genus two , then the `` _ _ amalgamation _ _ '' is a weakly reducible , genus three heegaard splitting of @xmath0 .
hence , we can make use of the correspondences between sets of surfaces in @xmath0 of genera at most two instead of surfaces of genus three or more .
since there have been many results about genus two heegaard splittings , this approach would make sense .
but the question is , how much information of elements of @xmath18 could be contained in a correspondence between two isotopy classes of generalized heegaard splittings consisting of two heegaard splittings of genus two ?
for @xmath2,[g]\in mod(m)$ ] and a generalized heegaard splitting @xmath19 consisting of two heegaard splittings of genus two , assume that @xmath2([\mathbf{h}])=[g]([\mathbf{h}])$ ] . even though the set of surfaces @xmath20 is isotopic to @xmath21
, we can not guarantee that @xmath2=[g]$ ] in @xmath12 , i.e. the difference @xmath6=[g]\cdot[f]^{-1}$ ] might not be the identity in @xmath12 . since two generalized heegaard splittings @xmath20 and @xmath21 are isotopic , we could expect that the amalgamations of them are also isotopic .
hence , there comes a natural expectation that there would be a representative @xmath5 of the difference @xmath6 $ ] such that @xmath5 preserves an embedding @xmath7 of the amalgamation obtained from @xmath20 .
hence , there would be the corresponding subset or subgroup of @xmath22 containing such representatives of @xmath6 $ ] ( @xmath22 is the group of isotopy classes of orientation - preserving automorphisms of @xmath0 preserving @xmath7 ) and this subset or subgroup would tell us how much information the correspondence loses for such elements of @xmath12 .
first , we will show that `` _ _ whether or not @xmath2 $ ] gives a correspondence between two weakly reducible , unstabilized heegaard surfaces of genus three _ _ '' can be interpreted as `` _ _ whether or not there exists a correspondence between two generalized heegaard splittings obtained by weak reductions from them by @xmath2 $ ] _ _ '' in theorem [ theorem - ghs - hs ] .
this gives an important motivation to understand @xmath2 $ ] as a correspondence between two generalized heegaard splittings instead of two heegaard splittings of genus three .
[ theorem - ghs - hs ] let @xmath23 and @xmath24 be weakly reducible , unstabilized , genus three heegaard splittings in an orientable , irreducible @xmath1-manifold @xmath0 and @xmath13 an orientation - preserving automorphism of @xmath0 .
then @xmath13 sends @xmath15 into @xmath7 up to isotopy if and only if @xmath13 sends a generalized heegaard splitting obtained by weak reduction from @xmath23 into a generalized heegaard splitting obtained by weak reduction from @xmath24 up to isotopy
. let @xmath25 be the set of isotopy classes of the generalized heegaard splittings consisting of two non - trivial heegaard splittings of genus two and @xmath26}$ ] the maximal subset of @xmath25 such that every element of @xmath26}$ ] gives the same isotopy class @xmath14 $ ] of amalgamation .
next , we will prove theorem [ theorem - main-2 ] which is the main theorem in this article .
[ theorem - main-2 ] let @xmath0 be an orientable , irreducible @xmath1-manifold having a weakly reducible , genus three heegaard splitting as a minimal genus heegaard splitting .
suppose that there is a correspondence between ( possibly duplicated ) two isotopy classes of @xmath25 by some elements of @xmath12 , say @xmath4\in \widetilde{\mathcal{ghs}}_{[f]}\rightarrow [ \mathbf{h}']\in \widetilde{\mathcal{ghs}}_{[f']}$ ] .
if @xmath2 $ ] , @xmath3\in mod(m)$ ] give the same correspondence , then there exists a representative @xmath5 of the difference @xmath6=[g]\cdot[f]^{-1}$ ] satisfying the follows . for a suitably chosen representative @xmath27 $ ] , 1 .
@xmath5 takes @xmath7 into itself and[thm12 - 1 ] 2 .
@xmath5 sends a uniquely determined weak reducing pair @xmath10 of @xmath7 into itself up to isotopy ( i.e. @xmath28 is isotopic to @xmath29 or @xmath30 in the relevant compression body and @xmath31 is isotopic to the other in the relevant compression body ) , where @xmath10 is determined naturally when we obtain @xmath7 by amalgamation from a representative @xmath8 of @xmath9$].[thm12 - 2 ] moreover , for any orientation - preserving automorphism @xmath11 of @xmath0 satisfying ( [ thm12 - 1 ] ) and ( [ thm12 - 2 ] ) , there exist two elements in @xmath12 giving the correspondence @xmath4\to[\mathbf{h}']$ ] such that @xmath11 belongs to the isotopy class corresponding to the difference between them . hence , the main theorem means that the difference between such two elements of @xmath12 comes from the subgroup of @xmath22 consisting of elements preserving the weak reducing pair @xmath10 , say @xmath32 .
this section introduces basic notations and summarizes the author s results in @xcite @xcite @xcite @xcite .
[ def - isotopy ] let @xmath0 be a manifold .
an _ ambient isotopy _ taking @xmath33 into @xmath34 is a family of maps @xmath35 , @xmath36 such that the associated map @xmath37 given by @xmath38 is continuous , @xmath39 is the identity , @xmath40 , and @xmath41 is a homeomorphism from @xmath0 to itself at any time @xmath42 . in this article
, we just say @xmath33 is _ isotopic _ to @xmath34 in @xmath0 by an isotopy @xmath41 if there is an ambient isotopy @xmath41 taking @xmath33 into @xmath34 .
an _ isotopy _ between two homeomorphisms @xmath43 for two manifolds @xmath44 and @xmath45 is a family of maps @xmath46 , @xmath36 such that the associated map @xmath47 given by @xmath48 is continuous , @xmath49 , @xmath50 , and @xmath51 is a homeomorphism at any time @xmath42 .
let @xmath52 be a homeomorphism such that @xmath53 for a submanifold @xmath54 .
if there is an isotopy @xmath51 such that @xmath49 and @xmath55 , then we say that `` _ _ we can isotope @xmath13 so that @xmath56 _ _ '' .
for example , if @xmath57 itself is isotopic to @xmath58 by an isotopy @xmath41 in @xmath45 , then we can isotope @xmath13 so that @xmath56 by taking the isotopy @xmath59
. if we can isotope @xmath13 so that @xmath60 , then we say that `` _ _ @xmath13 takes ( or sends ) @xmath33 into @xmath34 up to isotopy _ _ '' .
if a homeomorphism @xmath13 is isotopic to @xmath61 , then we say that @xmath13 and @xmath61 belong to the same _ isotopy class _ , where we will denote the isotopy class of a homeomorphism @xmath13 as @xmath2 $ ] .
if we assume @xmath62 , then there is the set of isotopy classes of orientation - preserving automorphisms of @xmath0 , say @xmath12 .
then we can well - define the operation @xmath2\cdot [ g]$ ] as @xmath63 $ ] and this gives a group structure on @xmath12 with the identity @xmath64 $ ] and the inverse @xmath2^{-1}=[f^{-1}]$ ] .
suppose that @xmath13 is an orientation - preserving automorphism of @xmath0 . if a submanifold @xmath65 is isotopic to @xmath66 in @xmath0 , i.e. @xmath67 and @xmath68 by an isotopy @xmath41 for @xmath42 , then the image @xmath69 is isotopic to @xmath70 by the isotopy @xmath71 for @xmath42 .
moreover , if @xmath13 is isotopic to @xmath72 by an isotopy @xmath51 for @xmath42 for two representatives @xmath13 and @xmath72 of @xmath2 $ ] , then the isotopy @xmath73 for @xmath42 sends @xmath74 into @xmath75 .
this means that we can well - define the image @xmath2([f])$ ] as @xmath16 $ ] for an isotopy class @xmath14 $ ] and an element @xmath2\in mod(m)$ ] .
[ def - compression ] a _ compression body _ is a @xmath1-manifold which can be obtained by starting with some closed , orientable , connected surface @xmath15 , forming the product @xmath76 , attaching some number of @xmath77-handles to @xmath78 and capping off all resulting @xmath77-sphere boundary components that are not contained in @xmath79 with @xmath1-balls . the boundary component @xmath79 is referred to as @xmath80 .
the rest of the boundary is referred to as @xmath81
. if a compression body @xmath82 is homeomorphic to @xmath83 , then we call it _
trivial _ and otherwise we call it _ nontrivial_. the cores of the @xmath77-handles defining a compression body @xmath82 , extended vertically down through @xmath83 , are called a _ defining set _ of @xmath77-disks for @xmath82 . a defining set for @xmath82
is _ minimal _ if it properly contains no other defining set .
note that we can define a compression body @xmath82 with non - empty minus boundary as a connected @xmath1-manifold obtained from @xmath76 for a ( possibly disconnected ) closed surface @xmath15 such that each component of @xmath15 is of genus at least one , followed by @xmath84-handles attached to @xmath85 , where @xmath79 becomes @xmath86 and the other boundary of @xmath82 becomes @xmath87 .
[ lemma - defining ] a genus @xmath88 compression body @xmath82 with minus boundary having a genus @xmath89 component has a unique minimal defining set up to isotopy and it consists of only one disk .
if @xmath86 is connected , i.e. @xmath86 consists of a genus @xmath89 surface , then there is a unique non - separating disk in @xmath82 up to isotopy . if @xmath86 is disconnected , i.e. @xmath86 consists of a genus @xmath89 surface and a torus
, then there is a unique compressing disk in @xmath82 up to isotopy , where it is separating in @xmath82 .
moreover , if we cut @xmath82 along the uniquely determined disk , then we get @xmath90 in any case .
therefore , we can obtain @xmath82 by attaching only one @xmath84-handle to @xmath90 corresponding to the disk .
this gives a way to determine @xmath82 by attaching only one @xmath77-handle to @xmath83 and therefore the relevant defining set is the singleton set consisting of the disk .
since this defining set consists of only one disk , it is a minimal defining set .
moreover , if there is a minimal defining set for @xmath82 , i.e. it consists of a disk , then the disk must be a compressing disk of @xmath82 otherwise the resulting compression body would be trivial .
hence , it must consist of a non - separating disk ( if @xmath86 is connected ) or a separating compressing disk ( if @xmath86 is disconnected ) by considering the shape of the resulting minus boundary .
hence , a minimal defining set for @xmath82 is uniquely determined up to isotopy by the argument in the start of the proof .
this completes the proof .
[ def - spine ] a _ spine _ of a compression body @xmath82 is a graph @xmath91 embedded in @xmath82 with some valence - one vertices possibly embedded in @xmath86 such that @xmath92 is homeomorphic to @xmath93 $ ] where @xmath94 is an open regular neiborhood of @xmath91 .
a spine @xmath91 of @xmath82 is _ minimal _ if it is a union of arcs , each of which has both ends on @xmath86 ( or at a single vertex if @xmath82 is a handlebody ) . a spine @xmath91 of a compression body @xmath82 is _ dual _ to a defining set @xmath95 for @xmath82
if each edge of @xmath91 intersects a single disk of @xmath95 exactly once , each disk of @xmath95 intersects exactly one edge of @xmath91 , and each ball of @xmath96 contains exactly one vertex of @xmath91 , and all vertices of @xmath91 in @xmath90 component of @xmath96 are contained in @xmath86 . a _
heegaard splitting _ of a @xmath1-manifold @xmath0 is an expression of @xmath0 as a union @xmath97 , denoted as @xmath23 ( or @xmath98 simply ) , where @xmath82 and @xmath99 are compression bodies that intersect in a transversally oriented surface @xmath100 .
we say @xmath15 is the _ heegaard surface _ of this splitting .
if @xmath82 or @xmath99 is homeomorphic to a product , then we say the splitting is _
trivial_. if there are compressing disks @xmath101 and @xmath102 such that @xmath103 , then we say the splitting is _ weakly reducible _ and call the pair @xmath104 a _ weak reducing pair_. if @xmath104 is a weak reducing pair and @xmath105 is isotopic to @xmath106 in @xmath15 , then we call @xmath104 a _ reducing pair_. if the splitting is not trivial and we can not take a weak reducing pair , then we call the splitting _
strongly irreducible_. if there is a pair of compressing disks @xmath107 such that @xmath108 intersects @xmath109 transversely in a point in @xmath15 , then we call this pair a _ canceling pair _ and say the splitting is _
stabilized_. otherwise , we say the splitting is _
unstabilized_. let @xmath15 be a surface of genus at least two in a compact , orientable @xmath1-manifold @xmath0 .
then the _ disk complex _
@xmath110 is defined as follows : a. vertices of @xmath110 are isotopy classes of compressing disks for @xmath15 .
b. a set of @xmath111 vertices forms an @xmath112-simplex if there are representatives for each that are pairwise disjoint .
hence , two compressing disks @xmath113 and @xmath114 of @xmath15 correspond to the same vertex in @xmath110 if and only if there exists an isotopy @xmath41 defined on @xmath0 such that ( i ) @xmath115 , ( ii ) @xmath116 , and ( iii ) @xmath117 for @xmath42 .
[ def - dvwf ] consider a heegaard splitting @xmath23 of an orientable , irreducible @xmath1-manifold @xmath0 .
let @xmath118 and @xmath119 be the subcomplexes of @xmath110 spanned by compressing disks in @xmath82 and @xmath99 respectively .
we call these subcomplexes _ the disk complexes of @xmath82 and @xmath99_. let @xmath120 be the subset of @xmath110 consisting of the simplices having at least one vertex from @xmath118 and at least one vertex from @xmath119
. we will denote the isotopy class @xmath121\in { \mathcal{d}_\mathcal{v}}(f)$ ] as @xmath101 or @xmath122 for the sake of convenience if there is no confusion .
@xmath118 and @xmath119 are contractible . from now on , we will consider only unstabilized heegaard splittings of an irreducible @xmath1-manifold . if a heegaard splitting of a compact @xmath1-manifold is reducible , then the manifold is reducible or the splitting is stabilized ( see @xcite ) .
hence , we can exclude the possibilities of reducing pairs among weak reducing pairs .
suppose @xmath123 is a compressing disk for @xmath124 .
then there is a subset of @xmath0 that can be identified with @xmath125 so that @xmath126 and @xmath127 .
we form the surface @xmath128 , obtained by _ compressing @xmath15 along @xmath123 _ , by removing @xmath129 from @xmath15 and replacing it with @xmath130 .
we say the two disks @xmath130 in @xmath128 are the @xmath131 of @xmath123 .
[ lemma-2 - 8 ] suppose that @xmath0 is an irreducible @xmath1-manifold and @xmath23 is an unstabilized heegaard splitting of @xmath0 .
if @xmath7 is obtained by compressing @xmath15 along a collection of pairwise disjoint disks , then no @xmath132 component of @xmath7 can have scars from disks in both @xmath82 and @xmath99 . if we add the assumption that the genus of the heegaard splitting is three , then we get the following important lemma .
[ lemma-2 - 9 ] suppose that @xmath0 is an irreducible @xmath1-manifold and @xmath23 is an unstabilized , genus three heegaard splitting of @xmath0 . if there exist three mutually disjoint compressing disks @xmath133 , @xmath134 and @xmath102 , then either @xmath133 is isotopic to @xmath29 , or one of @xmath105 and @xmath135 bounds a punctured torus @xmath136 in @xmath15 and the other is a non - separating loop in @xmath136 .
moreover , we can not choose three weak reducing pairs @xmath137 , @xmath138 , and @xmath139 such that @xmath140 and @xmath141 are mutually disjoint and non - isotopic in @xmath82 for @xmath142 .
note that `` _ _ one of @xmath105 and @xmath135 bounds a punctured torus @xmath136 in @xmath15 and the other is a non - separating loop in @xmath136 _ _ '' means that one of @xmath133 and @xmath29 , say @xmath29 , cuts off a solid torus from @xmath82 and @xmath133 is a meridian disk of the solid torus and therefore @xmath29 is a band sum of two parallel copies of @xmath133 in @xmath82 . in a weak reducing pair for a heegaard splitting @xmath23 ,
if a disk belongs to @xmath82 , then we call it a _
@xmath82-disk_. otherwise , we call it a _ @xmath99-disk_. we call a @xmath77-simplex in @xmath120 represented by two vertices in @xmath118 and one vertex in @xmath119 a _
@xmath82-face _ , and also define a _
@xmath99-face _ symmetrically .
let us consider a @xmath84-dimensional graph as follows . 1 .
we assign a vertex to each @xmath82-face in @xmath120 .
2 . if a @xmath82-face shares a weak reducing pair with another @xmath82-face , then we assign an edge between these two vertices in the graph .
we call this graph the _ graph of @xmath82-faces_. if there is a maximal subset @xmath143 of @xmath82-faces in @xmath120 representing a connected component of the graph of @xmath82-faces and the component is not an isolated vertex , then we call @xmath143 a _ @xmath82-facial cluster_. similarly , we define the _ graph of @xmath99-faces _ and a _ @xmath99-facial cluster_. in a @xmath82-facial cluster , every weak reducing pair gives the common @xmath99-disk , and vise versa .
if we consider an unstabilized , genus three heegaard splitting of an irreducible @xmath1-manifold , then we get the following lemmas .
[ lemma-2 - 14 ] suppose that @xmath0 is an irreducible @xmath1-manifold and @xmath23 is an unstabilized , genus three heegaard splitting of @xmath0 .
if there are two @xmath82-faces @xmath144 represented by @xmath145 and @xmath146 represented by @xmath147 sharing a weak reducing pair @xmath138 , then @xmath148 is non - separating , and @xmath149 , @xmath150 are separating in @xmath15 .
therefore , there is a unique weak reducing pair in a @xmath82-facial cluster which can belong to two or more faces in the @xmath82-facial cluster .
[ definition-2 - 15 ] by lemma [ lemma-2 - 14 ] , there is a unique weak reducing pair in a @xmath82-facial cluster belonging to two or more faces in the @xmath82-facial cluster .
we call it the _ center _ of a @xmath82-facial cluster .
we call the other weak reducing pairs _ hands _ of a @xmath82-facial cluster
. see figure [ figure3 ] . note that
if a @xmath82-face in a @xmath82-facial cluster is represented by two weak reducing pairs , then one is the center and the other is a hand .
lemma-2 - 14 ] means that the @xmath82-disk in the center of a @xmath82-facial cluster is non - separating , and those from hands are all separating .
moreover , lemma [ lemma-2 - 9 ] implies that ( i ) the @xmath82-disk in a hand of a @xmath82-facial cluster is a band sum of two parallel copies of that of the center of the @xmath82-facial cluster and ( ii ) the @xmath82-disk of a hand of a @xmath82-facial cluster determines that of the center of the @xmath82-facial cluster by the uniqueness of the meridian disk of the solid torus which the @xmath82-disk of the hand cuts off from @xmath82 .
-facial cluster in @xmath120 .
@xmath137 is the center and the other weak reducing pairs are hands .
[ figure3],width=170 ] [ lemma-2 - 16 ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] .
every @xmath82-face belongs to some @xmath82-facial cluster .
moreover , every @xmath82-facial cluster has infinitely many hands .
the next is the definition of `` _ _ generalized heegaard splitting _ _ '' originated from @xcite .
[ definition-2 - 9 ] a _ generalized heegaard splitting _ ( ghs )
@xmath19 of a @xmath1-manifold @xmath0 is a pair of sets of pairwise disjoint , transversally oriented , connected surfaces , @xmath151 and @xmath152 ( called the _ thick levels _ and _ thin levels _ , resp . ) , which satisfies the following conditions . 1 .
each component @xmath153 of @xmath154 meets a unique element @xmath155 of @xmath151 and @xmath155 is a heegaard surface in @xmath153 .
henceforth we will denote the closure of the component of @xmath154 that contains an element @xmath156 as @xmath157 .
2 . as each heegaard surface
@xmath158 is transversally oriented , we can consistently talk about the points of @xmath157 that are `` above '' @xmath155 or `` below '' @xmath155 .
suppose @xmath159 .
let @xmath157 and @xmath160 be the submanifolds on each side of @xmath161 .
then @xmath161 is below @xmath155 if and only if it is above @xmath162 .
3 . there is a partial ordering on the elements of @xmath152 which satisfies the following : suppose @xmath155 is an element of @xmath151 , @xmath161 is a component of @xmath163 above @xmath155 and @xmath164 is a component of @xmath163 below @xmath155
. then @xmath165 .
we denote the maximal subset of @xmath152 consisting of surfaces only in the interior of @xmath0 as @xmath166 and call it the _ inner thin levels_. if the corresponding heegaard splitting of @xmath157 is not trivial for every @xmath156 , then we call @xmath19 _ clean_. note that a ghs in this article is the same as a _ pseudo - ghs _ in @xcite since we allow a ghs to have product compression bodies and we do not encounter thin @xmath77-spheres .
the next is the definition of `` _ _ generalized heegaard splitting _ _ '' originated from @xcite .
[ definition - wr ] let @xmath0 be an orientable , irreducible 3-manifold .
let @xmath19 be an unstabilized heegaard splitting of @xmath0 , i.e. @xmath167 and @xmath152 consists of @xmath168 .
let @xmath133 and @xmath123 be disjoint compressing disks of @xmath15 from the opposite sides of @xmath15 such that @xmath169 has no @xmath77-sphere component .
( lemma [ lemma-2 - 8 ] guarantees that @xmath169 does not have a @xmath77-sphere component . )
define @xmath170 @xmath171 where we assume that each element of @xmath172 belongs to the interior of @xmath82 or @xmath99 by slightly pushing off @xmath173 or @xmath128 into the interior of @xmath82 or @xmath99 respectively and then also assume that they miss @xmath174 .
we say the ghs @xmath175 is obtained from @xmath19 by _ pre - weak reduction _ along @xmath104 .
the relative position of the elements of @xmath172 and @xmath176 follows the order described in figure [ figure2 ] . , width=377 ]
if there are elements @xmath177 and @xmath178 that cobound a product region @xmath179 of @xmath0 such that @xmath180 and @xmath181 then remove @xmath182 from @xmath172 and @xmath183 from @xmath176 .
this gives a clean ghs @xmath184 of @xmath0 from the ghs @xmath185 ( see lemma 5.4 of @xcite ) and we say @xmath184 is obtained from @xmath185 by _
cleaning_. we say the clean ghs @xmath184 of @xmath0 given by pre - weak reduction along @xmath104 , followed by cleaning , is obtained from @xmath19 by _ weak reduction _ along @xmath104 .
the next is the definition of `` _ _ amalgamation _ _ '' originated from @xcite . since the original definition identifies the product structures near the relevant thin level into the thin level itself , the union of submanifolds after amalgamation is not exactly the same as the union before amalgamation setwisely .
hence , we need to use another version of amalgamation .
[ the detailed version of `` _ _ partial amalgamation _ _ '' of section 3 of @xcite by using the terms in @xcite][def - amalgamation ] let @xmath33 and @xmath186 be submanifolds of @xmath0 such that @xmath187 is a ( possibly disconnected ) closed surface @xmath15 , where @xmath15 belongs to @xmath188 and @xmath189 .
suppose that @xmath33 and @xmath186 have non - trivial heegaard splittings @xmath190 and @xmath191 respectively , where @xmath192
. then we can represent @xmath193 as the union of @xmath194 and @xmath84-handles attached to @xmath195 and the symmetric argument also holds for @xmath196 .
especially , we can choose the product structures of the submanifolds @xmath197 and @xmath198 of @xmath199 and @xmath200 respectively ( hence @xmath201 and @xmath202 share @xmath15 as the common @xmath203-level ) such that the projections of attaching disks of the @xmath84-handles defining @xmath193 and @xmath196 in the @xmath84-levels of @xmath201 and @xmath202 into @xmath15 would be mutually disjoint .
let @xmath204 and @xmath205 ( @xmath206 or @xmath207 might be empty ) .
let @xmath208 and @xmath209 be the relevant projection functions defined in @xmath201 and @xmath202 respectively .
then we can extend the @xmath84-handles of @xmath193 until we meet @xmath15 by using @xmath210 through @xmath201 and also we can extend those of @xmath196 until we meet @xmath15 by using @xmath211 through @xmath202 . let @xmath212 ( @xmath213 resp . )
be the closure of the complement of the extended @xmath84-handles of @xmath193 in @xmath201 ( @xmath196 in @xmath202 resp . ) .
then we can see that @xmath214 is just expanded @xmath215 vertically down through @xmath212 and therefore it is a compression body and @xmath216 is also a compression body similarly .
if we define @xmath217\cup [ \text{the ( possibly extended)}~~1\text{-handles of } { \mathcal{w}}_1]\cup l_1 $ ] and @xmath218\cup[\text{the ( possibly extended)}~~1\text{-handles of } { \mathcal{v}}_2]\cup n_1 $ ] , then @xmath98 becomes a heegaard splitting of @xmath0 .
we call @xmath98 the _ amalgamation _ of @xmath219 and @xmath220 along @xmath15 with respect to the given @xmath84-handle structures of @xmath193 and @xmath196 and the pair @xmath221 ( see figure [ fig - amalgamation ] ) . and @xmath220 along @xmath15 [ fig - amalgamation],width=377 ] [ lemma - lackenby ] the amalgamation is well - defined up to ambient isotopy . despite of the existence of proposition [ lemma - lackenby ] , we need the precise definition as in definition [ def - amalgamation ] since we will analyze the exact differences between representatives of generalized heegaard splittings which induce the same amalgamation up to isotopy .
+ the following lemma means that the isotopy class of the generalized heegaard splitting obtained by weak reduction along a weak reducing pair does not depend on the choice of the weak reducing pair if the weak reducing pair varies in a fixed @xmath82- or @xmath99-facial cluster .
[ lemma-2 - 17 ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] .
every weak reducing pair in a @xmath82-face gives the same generalized heegaard splitting after weak reduction up to isotopy .
therefore , every weak reducing pair in a @xmath82-facial cluster gives the same generalized heegaard splitting after weak reduction up to isotopy .
moreover , the embedding of the thick level contained in @xmath82 or @xmath99 does not vary in the relevant compression body up to isotopy .
the next lemma gives an upper bound for the dimension of @xmath120 and restricts the shape of a @xmath1-simplex in @xmath120 .
[ lemma-2 - 18 ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] . then @xmath222 .
moreover , if @xmath223 , then every @xmath1-simplex in @xmath120 must have the form @xmath224 , where @xmath225 and @xmath226 . indeed , @xmath227 ( @xmath228 resp .
) is non - separating in @xmath82 ( in @xmath99 resp . ) and @xmath229 ( @xmath230 resp . ) is a band sum of two parallel copies of @xmath227 in @xmath82 ( @xmath228 in @xmath99 resp . ) .
the next lemma characterizes the possible generalized heegaard splittings obtained by weak reductions from @xmath23 into five types .
[ lemma - four - ghss ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] .
let @xmath231 be the generalized heegaard splitting obtained by weak reduction along a weak reducing pair @xmath104 from the heegaard splitting @xmath23 , where @xmath232 .
then this generalized heegaard splitting is one of the following five types ( see figure [ fig - heegaard - a ] ) .
a. each of @xmath233 and @xmath234 consists of a torus , where either[ghs - a ] a. @xmath133 and @xmath123 are non - separating in @xmath82 and @xmath99 respectively and @xmath235 is also non - separating in @xmath15 , b. @xmath133 cuts off a solid torus from @xmath82 and @xmath123 is non - separating in @xmath99 , c. @xmath123 cuts off a solid torus from @xmath99 and @xmath133 is non - separating in @xmath82 , or d. each of @xmath133 and @xmath123 cuts off a solid torus from @xmath82 or @xmath99 .
+ we call it a `` _ _ type ( a ) ghs _ _ '' .
b. one of @xmath233 and @xmath234 consists of a torus and the other consists of two tori , where either[lemma-3 - 1-b ] a. @xmath133 cuts off @xmath236 from @xmath82 and @xmath123 is non - separating in @xmath99,[lemma-3 - 1-b - i ] b. @xmath133 cuts off @xmath236 from @xmath82 and @xmath123 cuts off a solid torus from @xmath99,[lemma-3 - 1-b - ii ] c. @xmath123 cuts off @xmath236 from @xmath99 and @xmath133 is non - separating in @xmath82 , or[lemma-3 - 1-b - iii ] d. @xmath123 cuts off @xmath236 from @xmath99 and @xmath133 cuts off a solid torus from @xmath82.[lemma-3 - 1-b - iv ] + we call it a `` _ _ type ( b)-@xmath99 ghs _ _ '' for ( [ lemma-3 - 1-b - i ] ) and ( [ lemma-3 - 1-b - ii ] ) and `` _ _ type ( b)-@xmath82 ghs _ _ '' for ( [ lemma-3 - 1-b - iii ] ) and ( [ lemma-3 - 1-b - iv ] ) . c. each of @xmath233 and @xmath234 consists of two tori but @xmath237 is a torus , where each of @xmath133 and @xmath123 cuts off @xmath236 from @xmath82 or @xmath99.[lemma-3 - 1-c ] we call it a `` _ _ type ( c ) ghs _ _ '' .
d. each of @xmath233 and @xmath234 consists of two tori and @xmath237 also consists of two tori , where both @xmath133 and @xmath123 are non - separating in @xmath82 and @xmath99 respectively but @xmath235 is separating in @xmath15.[lemma-3 - 1-d ] we call it a `` _ _ type ( d ) ghs _ _ '' . as the summary of the previous observations ,
the generalized heegaard splitting @xmath231 is just a set of three surfaces @xmath238 obtained as the follows . 1 . the thick level @xmath239 ( @xmath240 resp .
) is obtained by pushing the genus two component of @xmath173 ( @xmath128 resp . ) off into the interior of @xmath82 ( of @xmath99 resp . ) and [ lemma-3 - 1 - 2nd-1 ] 2 .
the inner thin level @xmath241 is the union of components of @xmath174 having scars of both @xmath133 and @xmath123,[lemma-3 - 1 - 2nd-2 ] where we can see that if @xmath233 ( @xmath234 resp . )
has another component other than @xmath241 , then it belongs to @xmath242 ( @xmath86 resp . ) .
, width=453 ] from now on , we will use the notation @xmath238 as the generalized heegaard splitting obtained by weak reduction from a weakly reducible , unstabilized heegaard splitting @xmath23 of genus three along the weak reducing pair @xmath104 .
since every weak reducing pair in a @xmath82- or @xmath99-facial cluster @xmath243 gives a unique generalized heegaard splitting after weak reduction up to isotopy by lemma [ lemma-2 - 17 ] , we can say @xmath243 has a ghs of either type ( a ) , type ( b)-@xmath99 or type ( b)-@xmath82 by lemma [ lemma - four - ghss ] ( we exclude the possibility that @xmath243 has a ghs of type ( c ) or type ( d ) by lemma 3.7 of @xcite ) . in definition [ definition - bb - type - a ] , definition
[ definition - bb - type - b ] and definition [ definition - bb - type - cd ] , we will find a connected portion of @xmath120 , say a `` _ _ building block _ _ '' of @xmath120 , such that every weak reducing pair in a building block gives the same generalized heegaard splitting obtained by weak reduction up to isotopy .
[ definition - bb - type - a ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] .
let @xmath143 and @xmath244 be a @xmath82-facial cluster and a @xmath99-facial cluster such that they share the common center @xmath107 ( so @xmath108 and @xmath109 are non - separating in @xmath82 and @xmath99 respectively ) .
let @xmath245 be the union of all simplices of @xmath120 spanned by the vertices of @xmath246 .
let @xmath247 be a @xmath1-simplex of @xmath120 containing @xmath107 .
then @xmath248 for all possible @xmath29 and @xmath30 and therefore every weak reducing pair in @xmath245 gives the same generalized heegaard splitting up to isotopy of type ( a ) .
we call @xmath245 and @xmath107 a _ building block of @xmath120 having a type ( a ) ghs _ and the _ center _ of @xmath245 respectively .
[ definition - bb - type - b]@xmath249 1 .
_ a building block of @xmath120 having a type ( b)-@xmath99 ghs _ is a @xmath99-facial cluster having a type ( b)-@xmath99 ghs .
2 . _ a building block of @xmath120 having a type ( b)-@xmath82 ghs _ is a @xmath82-facial cluster having a type ( b)-@xmath82 ghs .
we define the _ center _ of a building block of @xmath120 having a type ( b)-@xmath99 or ( b)-@xmath82 ghs as the center of the corresponding @xmath99- or @xmath82-facial cluster .
[ definition - bb - type - cd ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] and let @xmath104 be a weak reducing pair .
suppose that the generalized heegaard splitting obtained by weak reduction along @xmath104 is a type ( c ) ghs ( type ( d ) ghs resp . ) .
in this case , we call the weak reducing pair @xmath104 itself `` _ _ a building block of @xmath120 having a type ( c ) ghs _ _ ( _ type ( d ) ghs _ resp . ) '' .
we define the _ center _ of the building block @xmath107 as @xmath104 itself .
note that the embedding of the thick level contained in @xmath82 or @xmath99 does not vary in the relevant compression body up to isotopy if we do weak reduction along a weak reducing pair contained in a fixed building block by lemma [ lemma-2 - 17 ] .
[ lemma - just - bb ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ]
. then every component of @xmath120 is just a building block of @xmath120 .
hence , we can characterize the components of @xmath120 into five types .
moreover , there is a uniquely determined weak reducing pair in each component of @xmath120 , i.e. the `` _ _ center _ _ '' of the component . by theorem [ lemma - just - bb ]
, we can say that a component of @xmath120 has a ghs of either type ( a ) , type ( b)-@xmath99 , type ( b)-@xmath82 , type ( c ) or type ( d ) .
moreover , we define the _ center _ of a component of @xmath120 as the center of the corresponding building block of @xmath120 .
we can refer to figure [ figure1 ] for the shapes of the components of @xmath120 .
[ figure1],width=453 ] the next lemma determines all centers of components of @xmath120 .
[ lemma - character - bb ] assume @xmath0 and @xmath15 as in lemma [ lemma-2 - 14 ] .
a weak reducing pair @xmath104 of @xmath23 is the center of a component of @xmath120 if and only if each of @xmath133 and @xmath123 does not cut off a solid torus from the relevant compression body .
moreover , a compressing disk in a weak reducing pair belongs to the center of a component of @xmath120 if and only if it does not cut off a solid torus from the relevant compression body .
the next theorem means that different components of @xmath120 give different ways to embed the thick levels of the generalized heegaard splittings obtained by weak reductions in the relevant compression bodies .
[ theorem - structure ] let @xmath23 be a weakly reducible , unstabilized , genus three heegaard splitting in an orientable , irreducible @xmath1-manifold @xmath0 .
then there is a function from the components of @xmath120 to the isotopy classes of the generalized heegaard splittings obtained by weak reductions from @xmath23 .
the number of components of the preimage of an isotopy class of this function is the number of ways to embed the thick level contained in @xmath82 into @xmath82 ( or in @xmath99 into @xmath99 ) .
this means that if we consider a generalized heegaard splitting @xmath19 obtained by weak reduction from @xmath23 , then the way to embed the thick level of @xmath19 contained in @xmath82 into @xmath82 determines the way to embed the thick level of @xmath19 contained in @xmath99 into @xmath99 up to isotopy and vise versa .
let @xmath250 be a weakly reducible , unstabilized heegaard splitting of genus three in an irreducible @xmath1-manifold @xmath0 for @xmath251 and @xmath13 an orientation preserving automorphism of @xmath0 that takes @xmath65 into @xmath66 .
let @xmath252 be a compressing disk of @xmath65 .
then we can well - define the map sending the isotopy class @xmath253\in { \mathcal{d}}(f_1)$ ] into @xmath254\in { \mathcal{d}}(f_2)$ ] and we can see that this gives a bijection between the set of vertices of @xmath255 and that of @xmath256 , where we denote this map as @xmath257 by using the same function name @xmath13 ( we will denote this map as @xmath258 rigorously in definition [ def - induce ] ) . the next lemma says that @xmath13 sends the center of a component of @xmath259 into the center of a component of @xmath260 ( the proof is essentially the same as that of lemma 3.1 of @xcite ) .
[ lemma - center - center ] suppose that @xmath0 is an orientable , irreducible @xmath1-manifold and @xmath250 is a weakly reducible , unstabilized , genus three heegaard splitting of @xmath0 for @xmath251 .
let @xmath13 be an orientation preserving automorphism of @xmath0 that takes @xmath65 into @xmath66 .
then @xmath13 sends the center of a component of @xmath259 into the center of a component of @xmath260 .
in this section , we will prove theorem [ theorem - ghs - hs ] .
suppose that there are two generalized heegaard splittings @xmath261 and @xmath262 obtained by weak reductions from weakly reducible , unstabilized , genus three heegaard splittings @xmath263 and @xmath264 of an orientable , irreducible @xmath1-manifold @xmath0 respectively .
assume that there is an orientation preserving automorphism @xmath13 of @xmath0 that takes @xmath261 into @xmath262 , i.e. @xmath13 sends the thick levels of @xmath261 into those of @xmath262 and sends the inner thin level of @xmath261 into that of @xmath262 . in theorem [ lemma - determine - ghss
] , we will prove that we can isotope @xmath13 so that ( i ) @xmath265 and ( ii ) @xmath266 .
[ lemma - determine - ghss ] let @xmath250 be a weakly reducible , unstabilized , genus three heegaard splitting in an orientable , irreducible @xmath1-manifold @xmath0 , @xmath267 a component of @xmath268 , @xmath269 the center of @xmath267 , and @xmath270 the generalized heegaard splitting obtained by weak reduction along @xmath269 from @xmath250 for @xmath251 .
if @xmath13 is an orientation preserving automorphism of @xmath0 sending @xmath261 into @xmath262 , then there is an isotopy @xmath51 such that @xmath49 , @xmath271 , and @xmath272 for @xmath42
. without loss of generality , assume that @xmath13 sends the thick level of @xmath261 contained in @xmath215 into the thick level of @xmath262 contained in @xmath193 .
let @xmath273 be the generalized heegaard splitting @xmath270 , where @xmath274 . in this
setting , @xmath275 by lemma [ lemma - four - ghss ] . by the assumption of @xmath13
, we can see that @xmath276 and @xmath277 for @xmath251 . we will prove that we can isotope @xmath13 so that @xmath278 where the isotopy preserves the thick levels and the inner thin level of @xmath279 during the isotopy .
if we consider the compressing disks @xmath140 and @xmath280 of @xmath281 and @xmath282 , then they are naturally extended to the compressing disks @xmath283 and @xmath284 of @xmath285 and @xmath286 as follows .
if we consider lemma [ lemma-2 - 8 ] , then @xmath287 belongs to the genus two component of @xmath288 , say @xmath289 , and @xmath287 is an essential simple closed curve in @xmath289 ( see ( a ) of figure [ fig - wrc ] ) , where `` the genus two component of @xmath288 '' is the one used when we obtain the thick level @xmath290 as in the last statement of lemma [ lemma - four - ghss ] . and
@xmath280 into the compressing disks of @xmath285 and @xmath286 respectively .
( b ) the ghs [ fig - wrc],width=453 ] here , the region between @xmath289 and @xmath290 is homeomorphic to @xmath291 .
let @xmath292 be a properly embedded incompressible annulus in @xmath291 such that @xmath287 is a component of @xmath293 and the other component of @xmath293 belongs to @xmath290 ( such an annulus can be obtained by projecting @xmath287 into @xmath290 through a given product structure of @xmath291 ) .
moreover , there is a unique properly embedded incompressible annulus in @xmath291 such that it has @xmath287 as a boundary component and the other boundary component belongs to @xmath290 up to isotopy constant on @xmath289 ( see lemma 3.4 of @xcite ) .
hence , the other component of @xmath293 other than @xmath287 is uniquely determined up to isotopy in @xmath290 .
this means that if we define @xmath283 as @xmath294 , then it becomes a compressing disk of @xmath285 and @xmath283 is well - defined up to isotopy in @xmath285 ( if we see ( b ) of figure [ fig - wrc ] or more generally figure 8 , figure 9 , figure 10 and figure 11 of @xcite , then we can see that @xmath283 is contained in @xmath285 ) .
the symmetric argument also holds for @xmath295 by considering the product region between the genus two component of @xmath296 containing @xmath295 , say @xmath297 , and @xmath298 and therefore we get the wanted compressing disk @xmath284 of @xmath286 from @xmath280 .
since @xmath269 is the center of @xmath267 for @xmath251 , we get the following claim . + * claim a * claim a recall that @xmath286 is a genus two compression body with non - empty minus boundary , i.e. there is a unique non - separating disk of @xmath286 if @xmath299 is connected or there is a unique compressing disk of @xmath286 if @xmath299 is disconnected up to isotopy and the uniquely determined disk cuts off @xmath286 into @xmath300 as in the proof of lemma [ lemma - defining ] . hence , it is sufficient to show that @xmath284 is isotopic to such disk in @xmath286 .
if we consider the case when @xmath299 is disconnected , then the proof is trivial by the uniqueness of compressing disk in @xmath286 .
now suppose that @xmath299 is connected .
then we can discard the cases when @xmath270 is a type ( c ) or type ( d ) ghs by lemma [ lemma - four - ghss ] .
if @xmath295 is separating in @xmath301 , then @xmath280 does not cut off a solid torus from @xmath282 by the assumption that @xmath269 is the center of @xmath267 and lemma [ lemma - character - bb ] , i.e. it cuts off @xmath236 from @xmath282 .
but this means that the generalized heegaard splitting obtained by weak reduction along @xmath269 is a type ( b)-@xmath281 ghs or type ( c ) ghs by lemma [ lemma - four - ghss ] , violating the assumption that @xmath302 is connected .
hence , @xmath295 must be non - separating in @xmath301 . here
, we can see that it is also non - separating in @xmath297 because the case when @xmath295 is non - separating in @xmath301 but it is separating in @xmath297 appears only if @xmath270 is of type ( d ) ghs by lemma [ lemma - four - ghss ] .
therefore the canonical projection of @xmath303 into @xmath298 in @xmath304 is also non - separating in @xmath298 , i.e. @xmath284 is a non - separating disk in @xmath286 .
the symmetric argument also holds for @xmath283 in @xmath285 .
this completes the proof of claim a. if we consider the assumption that @xmath283 and @xmath284 are naturally extended from @xmath140 and @xmath280 by attaching uniquely determined annuli to them , then we can assume that @xmath305 and @xmath306 are also product neighborhoods of @xmath140 and @xmath280 in @xmath281 and @xmath282 respectively , say @xmath307 and @xmath308 , by choosing @xmath309 and @xmath310 suitably .
hence , we can consider @xmath309 as a big cylinder and @xmath307 as a vertical small cylinder in the middle of @xmath309 for @xmath251 with respect to a given @xmath311 structure of @xmath309 and the symmetric argument also holds for @xmath310 and @xmath308 for @xmath251 .
( from now on , we will use the term `` _ _ cylinder _ _ '' to denote a @xmath1-manifold homeomorphic to @xmath311 . ) + * claim b1 * _ we can isotope @xmath13 so that ( i ) @xmath312 and @xmath313 , ( ii ) @xmath314 and @xmath315 , and ( iii ) the assumption @xmath266 holds at any time during the isotopy . _
+ claim b1 since @xmath316 by claim a and @xmath13 is a homeomorphism , @xmath317 , where @xmath318 is a @xmath84-handle attached to @xmath319 , i.e. @xmath320 is the cocore disk of the @xmath84-handle .
but lemma [ lemma - defining ] implies that there exists a unique such cocore disk in @xmath321 up to isotopy and therefore @xmath320 is isotopic to @xmath322 in @xmath321 by considering claim a. hence , the existence of the isotopy of @xmath13 satisfying ( i ) is obvious ( see the procedure from ( a ) to ( b ) of figure [ fig - isotopy ] ) .
after the previous isotopy , we can modify the location of the small cylinder @xmath323 in the big cylinder @xmath324 by an isotopy to satisfy ( ii ) ( see the procedure from ( b ) to ( c ) of figure [ fig - isotopy ] ) . since we can assume that @xmath266 during these isotopies , ( iii ) holds .
this completes the proof of claim b1 . and
@xmath314 [ fig - isotopy],width=453 ] note that the isotopy of claim b1 affects not only the image @xmath325 but also @xmath326 near @xmath327 even though both @xmath325 and @xmath326 are preserved setwisely during the isotopy .
but we can assume that it does not affect the image of the inner thin level and therefore this isotopy does not affect @xmath328 .
hence , we get the following claim similarly .
+ * claim b2 * _ without changing the result of claim b1 , we can isotope @xmath13 so that ( i ) @xmath329 and @xmath330 , ( ii ) @xmath331 and @xmath332 , and ( iii ) the assumption @xmath266 holds at any time during the isotopy . _ + the schematic figure describing this situation is figure [ fig - homeo ] .
+ , width=453 ] next , we can observe the follows , where this observation is the crucial idea of the proof of theorem [ lemma - determine - ghss ] . recall that @xmath333 is the center of @xmath267 for @xmath251 , i.e. each of @xmath140 and @xmath280 is either non - separating or cuts off @xmath236 from the relevant compression body by lemma [ lemma - character - bb ] .
( note that we can refer to the top of figure 8 , the top of figure 9 , figure 10 and figure 11 in @xcite for all possible cases . )
1 . if @xmath140 ( @xmath280 resp . )
is non - separating in @xmath281 ( @xmath282 resp . ) , then @xmath334 ( @xmath335 resp . ) is homeomorphic to @xmath311 in @xmath285 ( @xmath286 resp . )
intersecting @xmath301 in @xmath336 such that @xmath337 belongs to the inner thin level @xmath338 , where both disks are the scars of @xmath140 ( @xmath280 resp . ) , and the other levels belong to the interior of @xmath285 ( @xmath286 resp . ) for @xmath251 .
( if we check ( b ) of figure [ fig - wrc ] , then we can see that @xmath339 consists of @xmath304 and @xmath335 consists of a product neiborhood of @xmath280 in @xmath282 . )
moreover , @xmath340 ( @xmath341 resp . )
is also a @xmath311 in @xmath321 ( @xmath342 resp . )
such that the top and bottom levels belong to the inner thin level @xmath343 and the other levels belong to the interior of @xmath321 ( @xmath342 resp . ) by the assumption that @xmath344 ( @xmath345 resp . ) and @xmath13 is a homeomorphism .
[ papa ] + 2 . if @xmath140 ( @xmath280 resp . ) cuts off @xmath236 from @xmath281 ( @xmath282 resp . ) , then @xmath334 ( @xmath335 resp . ) is homeomorphic to @xmath236 in @xmath285 ( @xmath286 resp . ) intersecting @xmath301 in a once - punctured torus such that the top level belongs to @xmath346 ( @xmath347 resp . ) , the bottom level intersects the inner thin level @xmath338 in a disk , where this disk is a scar of @xmath140 ( @xmath280 resp . ) , and the other levels belong to the interior of @xmath285 ( @xmath286 resp . ) for @xmath251 .
( if we check ( b ) of figure [ fig - wrc ] , then we can see that @xmath348 consists of @xmath291 and @xmath334 consists of @xmath236 in @xmath281 . )
moreover , @xmath340 ( @xmath341 resp . )
is also a @xmath236 in @xmath321 ( @xmath342 resp . )
such that the top level belongs to @xmath233 ( @xmath349 resp . )
, the bottom level intersects the inner thin level @xmath343 in a disk , and the other levels belong to the interior of @xmath321 ( @xmath342 resp . ) by the assumption that @xmath344 ( @xmath345 resp . ) and @xmath13 is a homeomorphism .
[ papb ] + now we will visualize each case . for the sake of convenience , we will only consider the disk @xmath283 in @xmath285 and use the symmetric arguments for @xmath284 in @xmath286 .
we will denote @xmath350 as @xmath351 by claim a. let @xmath352 be @xmath353 itself and @xmath354 be the union of the other components of @xmath355 for @xmath251 . 1 .
case : @xmath140 is non - separating in @xmath281.[aaaaa ] + here , we can see that if we drill a hole in @xmath285 through the cylinder @xmath334 and take the closure of the resulting one , then @xmath356 is homeomorphic to @xmath291 , i.e. the core arc of the cylinder is a spine of @xmath285 by definition [ def - spine ] , say @xmath357 , because we can consider the closure of @xmath358 in @xmath285 as @xmath334 itself .
in particular , @xmath359 consists of two components @xmath360 and @xmath361 such that ( i ) @xmath362 and ( ii ) each @xmath363 is a cylinder whose top and bottom levels belong to @xmath364 and the other levels belong to @xmath365 for @xmath251 .
hence , we divide @xmath357 into the three parts , the core arc of @xmath307 and the core arcs of @xmath360 and @xmath361 which are the extended parts from the core arc of @xmath307 down to @xmath353 to complete @xmath357 .
if we consider @xmath366 , then the incompressible annulus @xmath367 is isotopic to vertical one by lemma 3.4 of @xcite by an isotopy constant on @xmath290 . in other words
, we can deform the product structure of @xmath291 so that @xmath367 would be vertical and we can assume that @xmath368 is also vertical .
hence , if we cut @xmath291 along @xmath368 , take the closure of the resulting one ,
i.e. @xmath369 , and move the annuli @xmath370 into vertical ones , then we get new product structure such that @xmath371 is homeomorphic to @xmath372 , where the @xmath84-level is @xmath373 itself .
this suggests a product structure of @xmath374 such that @xmath375 is vertical .
hence , we will represent @xmath375 and @xmath376 as vertical ones for the sake of convenience .
+ + is non - separating in @xmath281 .
[ fig - sit-1],width=453 ] + since @xmath13 is a homeomorphism , @xmath377 is homeomorphic to @xmath378 and the assumption that @xmath344 means that it is homeomorphic to @xmath379 .
moreover , we can see that @xmath380 where @xmath381 is homeomorphic to @xmath311 such that @xmath382 belongs to @xmath383 and @xmath384 belongs to @xmath385 as in the previous observation .
hence , definition [ def - spine ] implies that the core arc of the cylinder @xmath386 is also a spine of @xmath321 , say @xmath387 . here
, we can assume that @xmath388 is a parallel copy of @xmath389 by the assumption @xmath314 .
+ note that the inner thin level @xmath338 is either connected or disconnected even though @xmath140 is non - separating in @xmath281 , i.e. it consists of a torus or two tori ( see figure [ fig - sit-1 ] and figure [ fig - sit-3 ] respectively ) .
+ + is non - separating in @xmath281 .
[ fig - sit-3],width=264 ] 2 .
case : @xmath140 cuts off @xmath236 from @xmath281 . + let @xmath390 be the torus @xmath391
. then we can take @xmath392 such that ( i ) @xmath393 , ( ii ) @xmath394 intersects @xmath395 in a once - punctured torus , and ( iii ) @xmath396 where @xmath397 is a product neighborhood of @xmath140 in @xmath281 containing @xmath307 in the middle and @xmath398 is a compressing disk in @xmath281 isotopic to @xmath140 in @xmath281 .
let us consider the genus two compression body @xmath399 which is a deformation - retraction of @xmath285 .
then @xmath397 is a cylinder connecting the two components of @xmath400 . here , we take the product structure of @xmath351 so that @xmath401 would be horizontal .
if we drill a hole in @xmath402 through the cylinder @xmath397 and take the closure of the resulting one , say @xmath403 , then it is equal to @xmath404 , i.e. @xmath403 is homeomorphic to @xmath291 ( see ( b ) of figure [ fig - wrc ] ) and therefore also homeomorphic to @xmath405 .
this means that the core arc of @xmath397 is a spine of @xmath402 by defnition [ def - spine ] , say @xmath357 .
let the two @xmath236 components of @xmath406 be @xmath407 and @xmath408 , where @xmath409 and @xmath410 .
then @xmath411 consists of two components @xmath360 and @xmath361 such that each @xmath363 is a cylinder whose top and bottom levels belong to @xmath412 for @xmath413 ( see the left of figure [ fig - sit-2 ] ) . here , we can draw @xmath414 and @xmath363 as vertical ones in @xmath415 for @xmath413 by the assumption that @xmath357 is a spine of @xmath402 similarly as the previous case .
+ let us consider the image @xmath416 .
then it is equal to @xmath417 since @xmath344 .
moreover , we can see that @xmath418 is homeomorphic to @xmath419 because @xmath420 .
hence , we can isotope @xmath13 so that @xmath421 , i.e. we get @xmath422 . since @xmath13 is a homeomorphism , @xmath423 is homeomorphic to @xmath424 .
therefore , it is homeomorphic to @xmath425 such that @xmath426 is @xmath427 itself , i.e. the core arc of the cylinder @xmath428 is also a spine of @xmath429 by definition [ def - spine ] , say @xmath430 .
moreover , we can assume that @xmath388 is a parallel copy of @xmath389 .
see the right of figure [ fig - sit-2 ] .
+ cuts off @xmath236 from @xmath281 .
[ fig - sit-2],width=453 ] in any case , ( i ) we have represented @xmath431 for @xmath251 as vertical cylinders in the relevant product structure and ( ii ) we can say that the difference between @xmath432 and @xmath387 comes from the two subarcs @xmath433 and @xmath434 .
+ * claim c * _ we can isotope @xmath13 so that @xmath387 is monotone in the relevant product structure of @xmath435 or @xmath436 . _
+ claim c we will prove that we can isotope @xmath13 so that @xmath437 intersects each level surface of @xmath435 in two disks ( i.e. the core arcs of @xmath437 are monotone ) when @xmath140 is non - separating in @xmath281 and @xmath438 is connected and the other cases are left as exercise . since @xmath312 and @xmath344 , @xmath319 , where @xmath439 .
let @xmath440 .
suppose that there is an ambient isotopy @xmath41 defined on @xmath441 such that 1 .
@xmath41 is the identity on @xmath442 for @xmath42 and 2 .
@xmath443 intersects each level surface of @xmath435 in two disks , then we can extend it to the ambient isotopy @xmath444 defined on @xmath0 such that @xmath445 and @xmath444 is the identity on @xmath446 .
therefore , the argument in definition [ def - isotopy ] induces that @xmath13 can be isotoped so that @xmath437 intersects each level surface of @xmath435 in two disks .
hence , it is sufficient to show the existence of such @xmath41 .
let us define a homeomorphism @xmath447 such that 1 .
@xmath448 for @xmath449 , @xmath450 and a homeomorphism @xmath451 , 2 . where @xmath452 satisfies @xmath453 for @xmath449 and @xmath454 .
then we can see that @xmath455 because ( i ) @xmath456 , ( ii ) each @xmath457 is vertical in @xmath351 for @xmath251 , and ( iii ) @xmath458 by the assumption @xmath314 .
hence , if we consider the composition @xmath459 , then it is an automorphism of @xmath441 such that @xmath460 .
therefore lemma 3.5 of @xcite induces that there is an isotopy @xmath461 defined on @xmath435 , constant on @xmath442 , such that @xmath462 and @xmath463 is a level - preserving homeomorphism .
since @xmath464 intersects each level surface of @xmath435 in two disks , so does @xmath465 because @xmath463 is level - preserving .
hence , if we take the isotopy @xmath466 , then ( i ) @xmath115 and ( ii ) @xmath467 intersects each level surface of @xmath435 in two disks . if we consider the assumption that @xmath461 is constant on @xmath442 and @xmath115 , then we can see that @xmath41 is the identity on @xmath442 for @xmath42 .
this completes the proof of claim c. we can use the symmetric arguments for @xmath335 for @xmath251 and @xmath341 to visualize them and therefore we get the spine @xmath468 of @xmath342 or @xmath469 corresponding to @xmath470 and the spine @xmath471 of @xmath472 or @xmath473 corresponding to @xmath474 respectively , where the cylinder @xmath475 is obtained similarly as @xmath363 for @xmath476 .
moreover , we can assume that @xmath471 is monotone in the relevant product structure of @xmath477 or @xmath478 where @xmath468 is vertical similarly .
since both @xmath432 and @xmath387 are spines dual to the same minimal defining set @xmath479 of @xmath321 or @xmath429 , we can expect that @xmath340 would be isotopic to @xmath480 .
moreover , we can expect that @xmath341 would be isotopic to @xmath481 similarly .
but we can not guarantee that the isotopy sending @xmath340 into @xmath480 might not affect that sending @xmath341 into @xmath481 because they share the common inner thin level .
hence , we will describe the details of these `` _ _ untying isotopies _ _ '' and find the way how to avoid possible interferences in the proof of lemma [ lemma - spines ] .
[ lemma - spines ] we can assume that @xmath482 and @xmath483 simultaneously after a sequence of isotopies of @xmath13 satisfying @xmath266 at any time of the isotopies .
from now on , we will describe the `` _ _ untying isotopies _ _ '' of @xmath340 and @xmath341 rigorously such that the untying isotopies of @xmath340 do not affect those of @xmath341 .
+ * claim d * _ ( i ) we can take the relevant product structures of @xmath484 in @xmath321 and @xmath342 such that the projection images of the top and bottom levels of @xmath324 and @xmath485 as @xmath311 into @xmath343 do not intersect each other and ( ii ) we can isotope @xmath13 so that @xmath486 and @xmath487 satisfying @xmath266 at any time of the isotopies . _
+ claim d first , we will find two disks ( if @xmath488 is connected ) or two sets of two disks ( if @xmath488 is disconnected ) , say @xmath489 and @xmath490 , such that @xmath489 contains the projection image of @xmath491 into @xmath492 , @xmath490 contains the projection image of @xmath493 into @xmath494 , and @xmath495 for the case @xmath496 and @xmath497 , i.e. @xmath262 is a ghs of type ( a ) or type ( d ) . for the other cases ,
we only consider the relevant product structures in @xmath321 and @xmath342 intersecting the inner thin level @xmath488 ( for example , @xmath498 in @xmath321 ) and the details are left as exercise . if we consider the assumptions that @xmath499 is the inner thin level @xmath338 and the observation that each of @xmath334 and @xmath335 intersects @xmath338 in the scars of @xmath140 or @xmath280 respectively , then we can see that @xmath500 for @xmath251 .
recall that @xmath480 and @xmath481 are vertical cylinders in the relevant product structures intersecting @xmath343 .
if we change the product structures of @xmath435 and @xmath477 near @xmath501 and @xmath502 respectively , then we can assume that the projection image of @xmath503 into @xmath504 , say @xmath505 , misses that of @xmath506 into @xmath507 , say @xmath508 , without changing the assumption that @xmath480 and @xmath481 are vertical in the relevant product structures ( see figure [ fig - perturbation ] ) .
moreover , we can assume that this perturbation does not affect the assumption that @xmath387 and @xmath471 are monotone in the relevant product structures in @xmath321 and @xmath342 respectively if we deform the product structures sufficiently near @xmath501 and @xmath502 .
, width=377 ] also we can assume that there is a small neighborhood @xmath509 and @xmath510 of @xmath505 and @xmath508 in @xmath343 respectively such that @xmath511 . here
, we can see that each of @xmath505 and @xmath508 consists of two components .
case : @xmath343 consists of a torus , i.e. @xmath262 is a type ( a ) ghs .
+ we can choose a rectangle @xmath512 in the four - punctured torus @xmath513 such that one edge of @xmath512 belongs to one component of @xmath514 , the opposite edge belongs to the other component of @xmath514 , and the interior of the other two edges belongs to the interior of the four - punctured torus .
moreover , we can find a rectangle @xmath515 in the thrice - punctured torus @xmath516 such that two opposite edges of @xmath515 belong to the two components of @xmath517 and the interior of the other two edges belongs to the interior of the thrice - punctured torus similarly .
this gives two disjoint disks @xmath518 and @xmath519 in @xmath343 .
case : @xmath343 consists of two tori @xmath520 and @xmath521 , i.e. @xmath262 is a type ( d ) ghs .
+ in this case , one component of @xmath505 belongs to @xmath520 and the other component belongs to @xmath521 ( see figure [ fig - sit-3 ] ) .
the symmetric argument also holds for @xmath508
. then we choose @xmath522 as @xmath523 for @xmath413 and @xmath524 as @xmath525 for @xmath413 . in this case , denote @xmath526 and @xmath527 as @xmath489 and @xmath490 respectively . + next , we isotope @xmath13 near @xmath343 so that the disks ( or the disk ) @xmath528 would belong to @xmath489 and the disks ( or the disk ) @xmath529 would belong to @xmath490 satisfying @xmath266 at any time during the isotopy .
this argument can be generalized to the case when @xmath343 is disconnected
. then we get @xmath530 and @xmath487 . after this isotopy , @xmath489 and @xmath490 contain the image of scars @xmath531 and @xmath532 in their interiors as well as the scars @xmath533 and @xmath534 respectively ( see figure [ fig - separating ] ) . and @xmath487 after an isotopy of @xmath13 .
[ fig - separating],width=453 ] note that we can assume that the assumption that @xmath387 and @xmath471 are monotone in the relevant product structures of @xmath321 and @xmath342 are not changed after this step by using a suitable isotopy of @xmath13 .
this completes the proof of claim d. in the following step , we will realize the untying isotopies of @xmath340 and @xmath341 . +
* step a. * we will isotope @xmath340 into @xmath535 without affecting @xmath536 . + * case 1 .
* @xmath338 consists of a torus . + * case 1-a . *
@xmath353 consists of a torus , i.e. a subcase for non - separating @xmath140 .
+ in this case , @xmath537 .
+ * claim e.1-a * _ let @xmath538 be a small product neighborhood of @xmath539 in @xmath342 . after a sequence of isotopies of @xmath13 in @xmath540 which are the identity on @xmath541 , @xmath387 becomes vertical in @xmath435 .
moreover , @xmath266 at any time during the isotopies .
this means that this sequence of isotopies does not affect @xmath341 . _
+ claim e.1-a from now on , we will represent an isotopy of the cylinder @xmath542 by that of @xmath387 for the sake of convenience .
recall that @xmath387 is parallel to @xmath432 in @xmath543 .
but @xmath544 is just a monotone 2-strands in @xmath435 even though each component is unknotted ( see figure [ fig-4-braid ] )
. + and @xmath340 [ fig-4-braid],width=226 ] * step 1 : normalize @xmath387 in @xmath435 .
* in the proof of step 1 , we will denote @xmath489 as @xmath252 for the sake of convenience .
let @xmath545 and @xmath546 be the two strands of @xmath544 such that @xmath547 is the core arc of @xmath548 for @xmath413 .
we isotope @xmath13 near @xmath343 so that the projection of @xmath549 into @xmath343 is equal to @xmath550 and we say @xmath551 for @xmath413 .
then we choose sufficiently small @xmath552 such that @xmath553\cup [ 1-\epsilon,1]))\subset \operatorname{int}(d)\times ( [ 0,\epsilon]\cup [ 1-\epsilon,1])$ ] and reparametrize @xmath435 so that @xmath554
. then we isotope @xmath13 so that each subarc of @xmath555 would be a vertical strand in @xmath556\cup[\frac{3}{4},1])$ ] only affecting @xmath557\cup[\frac{3}{4}-\epsilon',1])$ ] for sufficiently small @xmath558 . after the isotopies and
the reparametrization , @xmath559\cup[\frac{3}{4},1])$ ] is equal to @xmath560\cup[\frac{3}{4},1]))$ ] and we can assume that @xmath387 remains monotone in @xmath561 $ ] .
let @xmath562 be a point in @xmath563 missing @xmath564 .
if we consider @xmath565 in @xmath435 , then it is a genus two handlebody @xmath566 , where we assume a neighborhood @xmath567 in @xmath563 also misses @xmath564 . here , we choose a meridian @xmath568 and a longitude @xmath569 of @xmath563 such that ( i ) @xmath568 intersects @xmath569 transversely in exactly one point @xmath562 , ( ii ) @xmath570 misses @xmath564
. then @xmath571 forms a minimal defining set of @xmath566 .
i.e. @xmath572 is a @xmath1-ball @xmath573 , where @xmath574 is a product neighborhood of @xmath575 in @xmath566 for @xmath251 so that each @xmath576 is vertical in @xmath435 for @xmath42 .
if we isotope @xmath13 , then we can assume that ( i ) @xmath387 misses @xmath577 , ( ii ) @xmath387 intersects @xmath578 transversely , and ( iii ) two different points of the intersection points @xmath579 are positioned at two different levels of @xmath441 without changing the assumption that @xmath387 is monotone in @xmath435 ( see ( a ) of figure [ fig - normalize ] ) .
, width=491 ] here , we assume that the indices of @xmath580 follow the order of levels of @xmath435 .
let @xmath581 ( @xmath582 ) be the set of levels of @xmath441 corresponding to @xmath580 .
assume that ( i ) we ve chosen @xmath583 sufficiently thin so that @xmath584 and ( ii ) whenever @xmath387 intersects @xmath574 for @xmath585 or @xmath77 in a subarc of @xmath387 , it passes through each of @xmath586 and @xmath587 transversely in exactly one point .
let @xmath588 be the level surface @xmath589 and @xmath590 therefore @xmath591 is a disk in the level surface @xmath588 .
let us isotope @xmath13 so that this isotopy forces each component of @xmath592 to belong to the corresponding @xmath588 but monotone elsewhere in @xmath435 .
let @xmath593 be the disk @xmath594 .
if we isotope @xmath13 so that the @xmath1-ball @xmath595 $ ] shrinks into the @xmath1-ball @xmath596 $ ] for very small @xmath558 preserving each level ( i.e. the genus two handlebody @xmath597)$ ] expands into @xmath598 $ ] , see ( a)@xmath599(b ) in figure [ fig - normalize ] ) , then we can assume that 1 .
this isotopy only affects @xmath561 $ ] , 2 .
@xmath387 is horizontal in @xmath600 , and 3 .
@xmath387 belongs to @xmath601 in the complement of @xmath602 in @xmath435 . if we cut off @xmath601 along the level surfaces
@xmath602 , then we can see that ( i ) the closure of each component , say @xmath267 , is a @xmath1-ball , such that @xmath267 intersects @xmath603 if we say @xmath604 and @xmath605 and ( ii ) each of @xmath545 and @xmath546 intersects @xmath606 in a connected arc whose interior belongs to @xmath607 for every @xmath608 . since each component of @xmath609 is monotone in @xmath267
, @xmath610 forms a @xmath77-braid in @xmath606 .
hence , we normalize @xmath611 in @xmath561 $ ] as follows . 1 . [ nor0 ]
let @xmath612 , @xmath613\subset\mathcal{b}_0 $ ] , @xmath614\subset\mathcal{b}_i$ ] for @xmath615 , and @xmath616\subset\mathcal{b}_r$ ] .
+ let @xmath617 , @xmath618 , @xmath619 , @xmath620 , @xmath621 , @xmath622 , and @xmath623 .
[ nor1 ] we isotope @xmath13 so that the @xmath77-braid by the subarcs of @xmath611 in @xmath267 would become a `` _ _ standardly positioned @xmath77-braid _ _ '' with respect to the vertical direction , i.e. ( i ) every non - trivial twist of @xmath611 in @xmath267 belongs to @xmath624 and ( ii ) the image of the projection of the endpoints of this @xmath77-braid in @xmath624 into @xmath343 would be @xmath625 for @xmath608 ( see ( c ) of figure [ fig - normalize ] ) .
3 . cut @xmath435 along the level surfaces @xmath626 and let @xmath627 be the closure of each component .
hence , we get the set of submanifolds @xmath628 , where the index increase from the bottom level to the top level and therefore @xmath629 intersects @xmath492 and @xmath630 intersects @xmath631 .
4 . if @xmath632 is even , then @xmath633 belongs to some @xmath634 .
if @xmath632 is odd , then we can assume that one of @xmath633 is vertical in @xmath627 by adding an additional `` _ _ position - changing half - twist _
_ '' to the top of the corresponding @xmath77-braid in @xmath635 if we need .
( see figure [ fig - position - change ] . if we see ( c ) of figure [ fig - normalize ] , then the thin vertical subarcs of @xmath611 in @xmath636 $ ]
denotes the vertical part of @xmath611 in @xmath627 for odd @xmath632 . ) + , width=377 ] + if the @xmath77-braid in @xmath635 is left - handed ( right - handed resp . )
, then we add a left - handed ( right - handed resp . )
position - changing half - twist for the sake of convenience .
* step 2 : untying @xmath637 and @xmath638 into vertical cylinders . * as we did in step 1 , @xmath639 $ ] is divided into @xmath640-submanifolds @xmath641)$ ] with respect to @xmath611 satisfying the follows .
1 . @xmath617 , @xmath642 , and @xmath643 , 2 .
if @xmath644 intersects @xmath645 $ ] for @xmath646 or @xmath77 , then a subarc of @xmath644 travels horizontally in @xmath647 for some @xmath648 and @xmath649 $ ] is a vertical subarc for @xmath650.[step2a ] 3 .
if @xmath651)=\emptyset$ ] , then @xmath652)$ ] is a standardly positioned @xmath77-braid in @xmath653 $ ] with respect to the vertical direction.[step2b ] + let us consider @xmath654)$ ] which is a @xmath77-braid consisting of subarcs of @xmath611 .
this @xmath77-braid can be written by @xmath655 for @xmath656 , where @xmath91 is a right - handed half - twist between the corresponding subarcs of @xmath611 .
let @xmath657 be a disk in the interior of @xmath252 such that @xmath658 $ ] contains the two cylinders @xmath659)$ ] . here , we can isotope @xmath13 so that this makes this @xmath77-braid into a new @xmath77-braid with the representation @xmath660 such that @xmath661 ( see figure [ fig - untwisting ] ) and therefore we can repeat such isotopy over and over again until @xmath611 becomes vertical strands in @xmath662 $ ] . note that we can assume that this isotopy does not affect the outside of @xmath663)$ ] in @xmath325 , does affect only a small product neighborhood of @xmath657 in @xmath328 , but the compression body @xmath325 is preserved at any time during this isotopy setwisely . -braid
[ fig - untwisting],width=226 ] after this isotopy , @xmath654)$ ] becomes a trivial @xmath77-braid .
therefore , if @xmath664 , then we ve isotoped @xmath13 so that @xmath387 became vertical in @xmath435 and we ve reached the end of the proof of claim e.1-a . hence , we get @xmath665 and therefore @xmath666)\neq \emptyset$ ] for some @xmath585 or @xmath77 , say @xmath545 . in this case
, @xmath546 is vertical in @xmath667 $ ] and a subarc of @xmath545 travels in @xmath668 for some @xmath669 during the time when it leaves @xmath667 $ ] .
if we shrink @xmath637 into sufficiently thinner one in @xmath670 $ ] by an isotopy of @xmath13 and project @xmath671)$ ] into @xmath343 , then we get an annulus , say the `` _ _ shadow _ _ '' , and denote it as @xmath672 ( see figure [ fig - travel ] ) and @xmath673 is a rectangle which divides @xmath252 into two pieces . becomes a vertical cylinder.[fig - travel],width=453 ] if @xmath674 , then we isotope @xmath13 so that @xmath672 would miss @xmath490 ( see the left of figure [ fig - travel ] ) .
choose a small neighborhood of @xmath672 in @xmath343 , say @xmath675 , so that ( i ) @xmath676 would consist of two arcs , ( ii ) @xmath677 , and ( iii ) @xmath678 .
we can give the canonical direction to the core circle of @xmath672 such that it follows the direction where the level of @xmath545 increases .
choose two points @xmath179 and @xmath679 in @xmath680 such that they are contained in different components of @xmath676 .
let @xmath681 where the direction of @xmath682 is the same as the direction of the core circle of @xmath672 .
consider a curve @xmath683 such that @xmath683 starts from @xmath179 , it travels the interior of @xmath675 as much as @xmath684 , turns around along the half of @xmath685 , where @xmath686 means a sufficiently small neighborhood of @xmath687 in @xmath343 , and travels the interior of @xmath675 as much as @xmath682 until it ends at @xmath679 ( see the left of figure [ fig - travel ] ) . here
, @xmath683 meets @xmath688 four times .
if we isotope @xmath13 so that @xmath683 shrinks into a curve @xmath689 contained in the interior of @xmath252 , then we can assume that this isotopy make @xmath690 into a vertical cylinder and it does not affect @xmath691 ( see the right of figure [ fig - travel ] ) .
moreover , we can assume that the compression body @xmath325 is preserved at any time during this isotopy setwisely .
( but it affects the image of @xmath13 in a small product neighborhood of @xmath539 in @xmath342 , say @xmath538 . ) after this isotopy , we can reduce the @xmath640-submanifolds @xmath692)$ ] of @xmath639 $ ] into @xmath693-submanifolds .
+ therefore , if we repeat the arguments in the previous paragraph , then @xmath637 and @xmath638 would become vertical cylinders in @xmath435 and @xmath266 at any time during the isotopies of @xmath13 without affecting @xmath490 .
this completes the proof of claim e.1-a . *
* @xmath353 consists of two tori , i.e. @xmath140 cuts off @xmath236 from @xmath281 . in this case
, @xmath694 consists of two @xmath236s , where @xmath637 belongs to one @xmath236 and @xmath691 belongs to the other .
the relevant product structures are @xmath695 and @xmath498 in @xmath429 ( see the left of figure [ fig-4-braid-1-b ] ) .
and @xmath542 when @xmath696 is disconnected [ fig-4-braid-1-b],width=415 ] therefore , we only need to consider a @xmath84-strand in each of @xmath695 and @xmath498 .
moreover , @xmath697 , i.e. we do nt worry about the possibility that @xmath637 meets @xmath341 in the untying procedure in @xmath695 .
this means that the untying procedure is more easier than case 1-a .
hence , we can isotope @xmath387 so that it would be vertical in @xmath436 without affecting @xmath490 by the similar arguments as in case 1-a .
we can assume that this isotopy preserves @xmath416 setwisely as well as @xmath325 .
+ * case 2 .
* @xmath338 consists of two tori @xmath698 and @xmath699 .
if we consider @xmath700 , then it consists of two @xmath236s , where @xmath637 belongs to one @xmath236 and @xmath638 belongs to the other .
hence , the untying procedure is essentially the same as case 1-b except that both @xmath637 and @xmath638 intersect @xmath343 ( see the right of figure [ fig-4-braid-1-b ] ) .
hence , we can isotope @xmath387 so that it would be vertical in @xmath435 without affecting @xmath524 for @xmath413 by the similar arguments as in case 1-b .
+ after the untying procedure in case 1-a , case 1-b or case 2 , @xmath701 becomes to be apparentely parallel to @xmath682 in @xmath321 .
moreover , we can isotope @xmath13 so that the cylinder @xmath542 would be moved into @xmath702 in @xmath321 without affecting @xmath490 .
this means that we have isotoped @xmath340 into @xmath535 without affecting @xmath536 after claim d. moreover , these isotopies satisfy @xmath266 at any time . + * step b. * after step a , if we use the symmetric arguments in step a , then we can isotope @xmath536 into @xmath481 without affecting @xmath381 at any time
. moreover , this isotopy satisfies @xmath266 at any time .
+ after step a and step b , @xmath13 have been isotoped so that @xmath703 and @xmath704 satisfing @xmath266 at any time during the isotopy .
this completes lemma [ lemma - spines ] . by using the isotopies of lemma [ lemma - spines ] ,
@xmath13 has been isotoped so that it satisfies the following equation .
@xmath705 this completes the proof of theorem [ lemma - determine - ghss ] .
[ definition - hs - ghs ] let @xmath15 be a weakly reducible , unstabilized heegaard surface of genus three in an orientable , irreducible @xmath1-manifold @xmath0 .
let @xmath706 be the set of isotopy classes of the generalized heegaard splittings obtained by weak reductions from @xmath23 .
if there is a generalized heegaard splitting @xmath19 obtained by weak reduction from @xmath23 and its isotopy class is @xmath4\in\mathcal{ghs}_f$ ] , then we call @xmath19 a _ representative of @xmath4 $ ] coming from weak reduction_. we will say two representatives @xmath707 and @xmath708 of @xmath4\in\mathcal{ghs}_f$ ] coming from weak reductions are _ equivalent _ if ( i ) @xmath709 is isotopic to @xmath710 in @xmath82 , ( ii ) @xmath711 is isotopic to @xmath712 in @xmath99 , and ( iii ) @xmath713 is isotopic to @xmath563 in @xmath0 .
suppose that @xmath709 is isotopic to @xmath710 in @xmath82 for two representatives @xmath707 and @xmath708 of some isotopy classes @xmath714 $ ] and @xmath715\in\mathcal{ghs}_f$ ] respectively coming from weak reductions .
if we recall the proof of theorem [ theorem - structure ] in @xcite , then @xmath716 and @xmath717 belong to the same component of @xmath120 and therefore @xmath711 is isotopic to @xmath712 in @xmath99 and the thin level @xmath713 is isotopic to @xmath563 in @xmath0 by the definitions of building blocks and theorem [ lemma - just - bb ] .
moreover , @xmath261 is isotopic to @xmath262 as well as each thick or thin level is isotopic to the relevant thick or thin level , i.e. @xmath714=[\mathbf{h}_2]$ ] in @xmath706 .
this means that @xmath261 is equivalent to @xmath262 .
therefore , _
@xmath261 is equivalent to @xmath262 if and only if at least one thick level of one representative is isotopic to that of the other in the relevant compression body_. obviously , this gives an equivalent relation to the set of all representatives of the elements of @xmath706 coming from weak reductions .
let @xmath718 be the set of all these equivalent classes and we denote the equivalent class of a representative @xmath19 as @xmath719 .
if there is a component of @xmath120 , then every weak reducing pair in the component gives the same equivalent class in @xmath718 after weak reduction by theorem [ theorem - structure ] .
hence , this defines the function @xmath720 .
+ * claim a * _ @xmath721 is bijective . _ + claim a if we consider an element of @xmath718 , then there must be a weak reducing pair in @xmath120 realizing a representative of the equivalent class by weak reduction .
this gives the component of @xmath120 containing the weak reducing pair , i.e. @xmath722 is surjective .
suppose that @xmath723 for some components @xmath724 and @xmath725 of @xmath120 .
this means that every weak reducing pair in @xmath726 gives the same equivalent class in @xmath718 by weak reduction , i.e. this gives a uniquely determined isotopy class of the thick level contained in @xmath82 .
hence , theorem [ theorem - structure ] induces @xmath727 and therefore @xmath721 is injective .
this completes the proof of claim a. by claim a , @xmath721 gives a one - to - one correspondence between the components of @xmath120 and the equivalent classes in @xmath718 . finally , we reach corollary [ corollary - ghs - hs ] .
[ corollary - ghs - hs ] let @xmath23 and @xmath24 be weakly reducible , unstabilized , genus three heegaard splittings in an orientable , irreducible @xmath1-manifold @xmath0 and @xmath13 an orientation - preserving automorphism of @xmath0 .
then @xmath13 sends @xmath15 into @xmath7 up to isotopy if and only if @xmath13 sends a representative of an element of @xmath706 coming from weak reduction into a representative of an element of @xmath728 coming from weak reduction up to isotopy .
@xmath729 suppose that @xmath13 sends @xmath15 into @xmath7 up to isotopy .
that is , we can isotope @xmath13 so that @xmath730 .
let @xmath4 $ ] be an element of @xmath706 .
then there is a weak reducing pair @xmath104 of @xmath23 which gives a representative @xmath731 of @xmath4 $ ] coming from weak reduction .
if we consider the weak reducing pair determined by @xmath732 of @xmath7 , then it gives the generalized heegaard splitting @xmath733 obtained by weak reduction from @xmath24 .
+ * claim a * _ @xmath20 is a representative of an element of @xmath728 coming from weak reduction .
moreover , @xmath734 in @xmath735 . _
+ claim a recall that @xmath730 . without loss of generality , assume that @xmath736 and @xmath737 , i.e. @xmath738 and @xmath739 .
let us consider @xmath740 and observe the compressing disks @xmath741 and @xmath742 .
let @xmath743 be the region in @xmath82 between the genus two component of @xmath173 and @xmath744 where `` the genus two component of @xmath173 '' is the one used when we obtained the thick level @xmath744 .
let @xmath745 be the product neighborhood of @xmath133 in @xmath82 which was used when we compressed @xmath15 along @xmath133 to obtain @xmath173 .
then @xmath743 is homeomorphic to @xmath746 whose @xmath203-level is @xmath744 .
hence , @xmath747 is homeomorphic to @xmath746 whose @xmath203-level is @xmath748 .
moreover , the @xmath84-level of @xmath747 is the genus two component of @xmath749 if we compress @xmath7 along @xmath741 by using @xmath750 as the product neighborhood of @xmath741 in @xmath751 .
therefore , we can easily check the follows . 1 .
@xmath752 is obtained by pushing the genus two component of @xmath753 off into the interior of @xmath751 .
@xmath754 is obtained by pushing the genus two component of @xmath755 off into the interior of @xmath756 similarly .
@xmath757 is the union of components of @xmath758 having scars of both @xmath741 and @xmath742 similarly as @xmath241 because the images of the product neighborhoods of @xmath133 and @xmath123 in @xmath82 and @xmath99 which we used when we compressed @xmath15 along @xmath133 and @xmath123 to obtain @xmath241 of @xmath13 are also product neighborhoods of @xmath741 and @xmath742 in @xmath751 and @xmath756 respectively .
hence , @xmath20 is the generalized heegaard splitting obtained by weak reduction from @xmath24 along the weak reducing pair @xmath759 by lemma [ lemma - four - ghss ] .
this completes the proof of claim a. by claim a , @xmath734 in @xmath735 , i.e. @xmath20 is isotopic to @xmath8 . in other words
, @xmath13 sends @xmath19 into @xmath8 up to isotopy .
+ @xmath760 suppose that @xmath13 sends a representative @xmath731 of an element @xmath4\in \mathcal{ghs}_f$ ] coming from weak reduction into a representative @xmath761 of an element @xmath9\in \mathcal{ghs}_{f'}$ ] coming from weak reduction up to isotopy .
that is , we can isotope @xmath13 so that @xmath762 .
let @xmath107 and @xmath763 be the centers of the components @xmath573 and @xmath764 of @xmath120 and @xmath765 containing the weak reducing pairs @xmath104 and @xmath10 respectively .
then we get two generalized heegaard splittings @xmath766 and @xmath767 obtained by weak reductions from @xmath23 and @xmath24 respectively . here , ( i ) @xmath768 in @xmath718 and ( ii ) @xmath769 in @xmath735 by considering the functions @xmath721 and @xmath770 .
that is , ( i ) induces that @xmath19 is isotopic to @xmath771 by an isotopy @xmath41 such that @xmath39 is the identity and @xmath772 , and therefore @xmath773 is isotopic to @xmath774 by the isotopy @xmath775 . since @xmath8 is isotopic to @xmath776 by ( ii ) , we conclude that @xmath774 is isotopic to @xmath776 .
this means that we can isotope @xmath13 so that @xmath777 by using the argument in definition [ def - isotopy ] .
therefore , theorem [ lemma - determine - ghss ] induces that we can isotope @xmath13 so that @xmath730 .
this completes the proof . in this section
, we will prove theorem [ theorem - main-2 ] .
let @xmath778 be the set of isotopy classes of weakly reducible , unstabilized heegaard surfaces of genus three in @xmath0 .
now we define @xmath779\in\mathcal{f}}\mathcal{ghs}_f$ ] , where we take exactly one representative @xmath15 for each isotopy class @xmath14 $ ] .
suppose that @xmath15 is isotopic to @xmath7 in @xmath0 by an isotopy @xmath41 such that @xmath780 and @xmath781 .
then we get a @xmath84-parameter family of heegaard splittings @xmath782 such that @xmath783 and @xmath784 for @xmath785 , @xmath786 , and @xmath787 .
let @xmath19 be a representative of an element of @xmath788 coming from weak reduction along a weak reducing pair @xmath104 .
if we consider the weak reducing pair @xmath789 of @xmath790 , then it gives the generalized heegaard splitting @xmath791 obtained by weak reduction from @xmath792 . here
, claim a in corollary [ corollary - ghs - hs ] induces that @xmath793 is a representative of an element of @xmath794 coming from weak reduction and @xmath795 in @xmath796 for @xmath797 .
hence , we can see that ( i ) the isotopy @xmath41 sends @xmath19 into @xmath798 and ( ii ) @xmath799 in @xmath735 , i.e. the isotopy class @xmath4 $ ] is the same as @xmath714 $ ] and therefore each element of @xmath706 belongs to @xmath728 . if we consider the isotopy @xmath800 from @xmath7 to @xmath15 , then we can see that each element of @xmath728 belongs to @xmath788 by the symmetric argument , i.e. @xmath801 .
this is why we take only one representative for each element of @xmath778 in the union .
let @xmath25 be the set of isotopy classes of the generalized heegaard splittings consisting of two non - trivial heegaard splittings of genus two .
therefore , every representative of @xmath25 must be of the form @xmath802 , where @xmath803 ( @xmath804 is a torus or two tori ) and the genera of @xmath805 and @xmath806 are both two . if we add the assumption that the minimal genus of heegaard splittings in @xmath0 is three , then we get the following lemma .
[ lemma - ghs - ghss ] let @xmath0 be an orientable , irreducible @xmath1-manifold admitting a weakly reducible , unstabilized heegaard splitting of genus three and assume that the minimal genus of @xmath0 is three
. then @xmath807 . by lemma [ lemma - four - ghss ]
, @xmath808 is obvious .
suppose that @xmath809 is a representative of an element of @xmath25 , where @xmath803 .
then we can express @xmath193 as the union of @xmath194 and a @xmath84-handle attached to @xmath195 and the symmetric argument also holds for @xmath196 since they are genus two compression bodies with non - empty minus boundary .
hence , we obtain a heegaard splitting @xmath23 by the amalgamation of @xmath810 and @xmath811 along @xmath804 with respect to the @xmath84-handle structures of @xmath193 and @xmath196 and a suitable pair of projection functions as in definition [ def - amalgamation ] .
let @xmath252 and @xmath812 be the cocore disks of the @xmath84-handles in the representations of @xmath196 and @xmath193 respectively .
then we can see that @xmath813 is a weak reducing pair of @xmath23 .
moreover , if we observe the amalgamation @xmath15 , then we can see the follows . 1 . if both @xmath233 and @xmath234 are connected ( so @xmath804 consists of a torus ) , then @xmath15 is the one obtained from @xmath804 by attaching two tubes corresponding to the @xmath84-handles of @xmath252 and @xmath812 to @xmath804 .
if @xmath233 is disconnected and @xmath234 is connected ( so @xmath804 consists of a torus ) , then @xmath15 is the one obtained from the union of @xmath804 and a torus @xmath814 parallel to @xmath242 by attaching the tube corresponding to the @xmath84-handle of @xmath252 to @xmath804 and connecting @xmath804 and @xmath814 by the tube corresponding to the @xmath84-handle of @xmath812 ( see ( b ) of figure [ fig - wr - amal-2]).[3333b ] 3 .
if @xmath233 is connected and @xmath234 is disconnected ( so @xmath804 consists of a torus ) , then we get the symmetric result of ( [ 3333b ] ) .
4 . if both @xmath233 and @xmath234 are disconnected and @xmath815 is connected ( so @xmath804 consists of a torus ) , then @xmath15 is the one obtained from the union of @xmath804 , a torus @xmath814 parallel to @xmath242 and a torus @xmath816 parallel to @xmath86 by connecting @xmath804 and @xmath814 by the tube corresponding to the @xmath84-handle of @xmath812 and connecting @xmath804 and @xmath816 by the tube corresponding to the @xmath84-handle of @xmath252 .
if both @xmath233 and @xmath234 are disconnected and @xmath815 is disconnected , i.e. @xmath817 ( so @xmath804 consists of two tori @xmath818 and @xmath819 ) , then @xmath15 is the one obtained from @xmath804 attaching two tubes corresponding to the @xmath84-handles of @xmath252 and @xmath812 where each tube connects @xmath818 and @xmath819 . in all cases , we can see that the genus of @xmath15 is three . here
, we confirm that @xmath23 is unstabilized by the assumption that the minimal genus of @xmath0 is three . by using the above observation ,
if we compress @xmath15 along @xmath252 or @xmath812 and consider the genus two component , then it is isotopic to @xmath805 or @xmath806 respectively and the union of components of @xmath820 having scars of both @xmath252 and @xmath812 is isotopic to @xmath804 ( see ( c ) and ( a ) of figure [ fig - wr - amal-2 ] for type ( b)-@xmath82 ghs and we can draw similar figures for the other types of ghss ) .
( b ) : amalgamation ( b)@xmath599(c ) : weak reduction[fig - wr - amal-2],width=453 ] that is , the generalized heegaard splitting @xmath821 obtained by weak reduction from @xmath23 along the weak reducing pair @xmath813 is isotopic to @xmath19 ( refer to the last statement of lemma [ lemma - four - ghss ] ) .
this means that the isotopy class @xmath4 $ ] belongs to @xmath706 , i.e. @xmath822 .
this completes the proof .
let @xmath823 ( @xmath803 ) be a generalized heegaard splitting whose isotopy class belongs to @xmath25 and @xmath23 be the heegaard splitting obtained by amalgamation from @xmath19 along @xmath804 with respect to suitable @xmath84-handle structures of @xmath193 and @xmath196 and a pair of projection functions @xmath210 and @xmath211 defined on @xmath824s by using the notations in definition [ def - amalgamation ] .
if @xmath19 is isotopic to a generalized heegaard splitting @xmath8 by an isotopy @xmath825 such that @xmath826 and @xmath827 , then @xmath825 gives a @xmath84-parameter family of generalized heegaard splittings @xmath828 for @xmath829 .
if we consider the images of the @xmath84-handles of @xmath193 and @xmath196 in the relevant @xmath84-handle structures and the product structures of @xmath824s determined by the pair @xmath221 of @xmath825 , then there would be the corresponding @xmath84-handles of @xmath830 and @xmath831 and the pair of projection functions @xmath832 and @xmath833 defined on @xmath834s ( see figure [ fig - introduction-2 ] ) . and @xmath835[fig - introduction-2],width=377 ]
hence , we get the @xmath84-parameter family of amalgamations @xmath836 by using these images , where each @xmath837 is obtained from @xmath838 , and we can see that @xmath839 , i.e. it gives an isotopy from @xmath840 to @xmath841 . here , we can see that @xmath841 is obtained by amalgamation from @xmath842 .
this means that the isotopy class of the amalgamation obtained from @xmath19 and that obtained from @xmath8 guaranteed by proposition [ lemma - lackenby ] are the same , i.e. an isotopy class @xmath4\in\widetilde{\mathcal{ghs}}$ ] gives a unique isotopy class @xmath14 $ ] of amalgamation .
let @xmath26}$ ] be the maximal subset of @xmath25 such that every element of @xmath26}$ ] gives the same isotopy class @xmath14 $ ] of amalgamation .
[ def - induce ] let @xmath13 be an orientation - preserving automorphism of an irreducible @xmath1-manifold @xmath0 that takes a weakly reducible , unstabilized heegaard surface @xmath65 of genus three into @xmath66 , and @xmath263 and @xmath264 the relevant heegaard splittings . since we can represent a compressing disk in @xmath215 or @xmath196 as the boundary curve in @xmath65 and @xmath13 is a homeomorphism , @xmath13 would translate the information of the compressing disks of @xmath65 into that of @xmath66 .
let @xmath113 and @xmath114 be compressing disks of @xmath215 .
if @xmath843=[d_2]$ ] in @xmath255 , then there is an isotopy @xmath41 defined on @xmath0 such that ( i ) @xmath39 is the identity , ( ii ) @xmath116 , and ( iii ) @xmath844 for @xmath42 . without loss of generality , assume that @xmath845 and let us consider the images @xmath846 and @xmath847 in @xmath193 .
then @xmath848 is an isotopy sending @xmath846 into @xmath847 and we can see that @xmath849 for @xmath42 , i.e. @xmath850=[f(d_2)]$ ] in @xmath256 . hence , we can well - define the map @xmath851 by @xmath852)=[f(d)]$ ] .
moreover , we can induce the follows easily . 1 .
@xmath258 induces a bijection between the set of vertices of @xmath255 and that of @xmath256 .
2 . @xmath258 sends each @xmath853-simplex in @xmath255 into the corresponding @xmath853-simplex in @xmath256 for @xmath854 .
3 . @xmath258 sends each @xmath853-simplex in @xmath855 into the corresponding @xmath853-simplex in @xmath260 for @xmath854 .
4 . @xmath258 sends each component of @xmath259 into the corresponding component of @xmath260 ( refer to lemma [ lemma - just - bb ] ) .
moreover , if @xmath61 is an orientation - preserving automorphism of @xmath0 that takes the heegaard surface @xmath66 into @xmath856 , then we can see @xmath857)=[g\circ f(d)]=g_\ast([f(d)])=g_\ast(f_\ast([d]))$ ] , i.e. @xmath858 .
in addition , if we define @xmath859 as @xmath860 where @xmath860 is the induced map coming from @xmath861 , then we get @xmath862 and @xmath863 . finally , we reach theorem [ theorem - main - copy-2 ] .
[ theorem - main - copy-2 ] let @xmath0 be an orientable , irreducible @xmath1-manifold having a weakly reducible , genus three heegaard splitting as a minimal genus heegaard splitting .
suppose that there is a correspondence between ( possibly duplicated ) two isotopy classes of @xmath25 by some elements of @xmath12 , say @xmath4\in \widetilde{\mathcal{ghs}}_{[f]}\rightarrow [ \mathbf{h}']\in \widetilde{\mathcal{ghs}}_{[f']}$ ] .
if @xmath2 $ ] , @xmath3\in mod(m)$ ] give the same correspondence , then there exists a representative @xmath5 of the difference @xmath6=[g]\cdot[f]^{-1}$ ] satisfying the follows . for a suitably chosen representative @xmath27 $ ] , 1 .
@xmath5 takes @xmath7 into itself and[thm123 - 1 ] 2 .
@xmath5 sends a uniquely determined weak reducing pair @xmath10 of @xmath7 into itself up to isotopy ( i.e. @xmath28 is isotopic to @xmath29 or @xmath30 in the relevant compression body and @xmath31 is isotopic to the other in the relevant compression body ) , where @xmath10 is determined naturally when we obtain @xmath7 by amalgamation from a representative @xmath8 of @xmath9$].[thm123 - 2 ] moreover , for any orientation - preserving automorphism @xmath11 of @xmath0 satisfying ( [ thm123 - 1 ] ) and ( [ thm123 - 2 ] ) , there exist two elements in @xmath12 giving the correspondence @xmath4\to[\mathbf{h}']$ ] such that @xmath11 belongs to the isotopy class corresponding to the difference between them .
let @xmath19 and @xmath8 be arbitrarily chosen representatives of @xmath4 $ ] and @xmath9 $ ] respectively . here , we can represent each compression body of @xmath19 intersecting @xmath166 as @xmath864 and the symmetric argument also holds for @xmath8 . with respect to the @xmath84-handle structures of these compression bodies and suitable pairs of projection functions on @xmath865s and @xmath866s , we get the weakly reducible , unstabilized heegaard splittings @xmath23 and @xmath24 of genus three obtained by amalgamations from @xmath19 and @xmath8 along @xmath166 and @xmath867 respectively . recall that @xmath19 and @xmath8 are just generalized heegaard splittings such that each consists of two heegaard splittings of genus two and we only know the isotopy classes of the amalgamations are well - defined by proposition [ lemma - lackenby ] .
if we use the proof of lemma [ lemma - ghs - ghss ] , then @xmath19 and @xmath8 are isotopic to the generalized heegaard splittings obtained by weak reductions from @xmath23 and @xmath24 respectively . in other words
, we can isotope @xmath15 and @xmath7 so that @xmath19 and @xmath8 would be the generalized heegaard splittings obtained by weak reductions from @xmath23 and @xmath24 respectively .
let us realize these isotopies .
let @xmath104 and @xmath10 be the weak reducing pairs coming from the cocore disks of the relevant @xmath84-handles used when we obtained the amalgamations @xmath23 and @xmath24 respectively . if we thin the @xmath84-handle parts of @xmath15 and push @xmath15 off slightly to miss the thick levels of @xmath19 if we need , then we can see that @xmath19 itself is a generalized heegaard splitting obtained by weak reduction from @xmath23 along @xmath104 .
( refer to the last statement of lemma [ lemma - four - ghss ] and see figure [ fig - wr - amal-3 ] .
we can draw the similar figures for the other cases among the five cases of amalgamations in the proof of lemma [ lemma - ghs - ghss ] ) .
( a ) : we can isotope @xmath15 so that @xmath868 would be the ghs obtained by weak reductions from @xmath23.[fig - wr - amal-3],width=377 ] the symmetric argument also holds for @xmath7 and @xmath8 by using @xmath10 .
after these isotopies of @xmath15 and @xmath7 , we can define the equivalent classes @xmath719 and @xmath869 in @xmath718 and @xmath735 respectively . from now on , we will use these embeddings of @xmath15 and @xmath7 .
since @xmath133 and @xmath123 comes from the cocore disks of the @xmath84-handles , if any of them is separating in @xmath82 or @xmath99 after the amalgamation , then it cuts off @xmath236 from @xmath82 or @xmath99 ( recall the five cases of amalgamations in the proof of lemma [ lemma - ghs - ghss ] ) .
this means that @xmath104 is the center of the component of @xmath120 which @xmath104 belongs to , say @xmath573 , by lemma [ lemma - character - bb ] .
similarly , @xmath10 is the center of the component of @xmath765 which @xmath10 belongs to , say @xmath764 . by the assumption , @xmath2([\mathbf{h}])=[\mathbf{h}']$ ] and @xmath3([\mathbf{h}])=[\mathbf{h}']$ ] for @xmath2,[g]\in mod(m)$ ] .
hence , there are representatives @xmath13 and @xmath61 of @xmath2 $ ] and @xmath3 $ ] respectively such that @xmath762 and @xmath870 .
therefore , we can isotope @xmath13 and @xmath61 so that ( i ) @xmath730 and @xmath871 and ( ii ) @xmath762 and @xmath870 by theorem [ lemma - determine - ghss ] .
recall that @xmath19 and @xmath8 are generalized heegaard splittings obtained by weak reductions from @xmath23 and @xmath24 along the weak reducing pairs @xmath104 and @xmath10 respectively at this moment .
if we consider claim a in corollary [ corollary - ghs - hs ] , then we can see @xmath872 is the generalized heegaard splitting obtained by weak reduction from @xmath23 along the weak reducing pair determined by @xmath873 , say @xmath874 .
but theorem [ theorem - structure ] means that @xmath874 must belong to @xmath573 since the embeddings of thick levels determined by @xmath874 are isotopic to those determined by @xmath104 in the relevant compression bodies . indeed , @xmath875,[\tilde{w}]\}$ ]
is @xmath876,[w]\}$ ] by lemma [ lemma - center - center ] .
that is , the induced map @xmath877 sends @xmath764 into @xmath573 and @xmath878,[w']\}$ ] into @xmath876,[w]\}$ ] .
similarly , if we consider @xmath870 , then the induced map @xmath879 sends @xmath573 into @xmath880 and @xmath876,[w]\}$ ] into @xmath878,[w']\}$ ] .
let us consider the difference @xmath6=[g]\cdot [ f]^{-1}$ ] .
then @xmath881 is a representative of @xmath6 $ ] such that @xmath882 .
moreover , the induced map @xmath883 sends @xmath764 into @xmath764 and @xmath878,[w']\}$ ] into @xmath878,[w']\}$ ] itself by the previous observations .
this completes the proof of the existence of the representative @xmath5 . since ( i ) the weak reducing pair @xmath10 is the center of @xmath764 which is unique in @xmath764 by definition and ( ii ) the component @xmath764 is uniquely determined by the equivalent class @xmath884 as the preimage of the bijection @xmath770 , the weak reducing pair @xmath10 is uniquely determined
this completes the proof of the first statement . from now on
, we will prove the last statement .
consider an orientation - preserving automorphism @xmath11 of @xmath0 such that ( i ) @xmath11 takes @xmath7 into itself and ( ii ) @xmath11 sends @xmath10 into itself up to isotopy .
this means that if we consider the embeddings of thick levels of the generalized heegaard splitting @xmath885 obtained by weak reduction from @xmath24 along the weak reducing pair determined by @xmath886 , then they are isotopic to those obtained by weak reduction from @xmath24 along @xmath10 in the relevant compression bodies , i.e. @xmath887 in @xmath735 .
moreover , we can see that @xmath888 in @xmath735 by claim a of corollary [ corollary - ghs - hs ] .
i.e. @xmath889 .
therefore , we can isotope @xmath11 so that @xmath890 .
since there is at least one correspondence between @xmath4 $ ] and @xmath9 $ ] by an element @xmath2\in mod(m)$ ] , choose a representative @xmath72 of @xmath2 $ ] such that @xmath891 .
let @xmath892 .
then we can see that ( i ) @xmath893 sends @xmath19 into @xmath8 and ( ii ) @xmath894 .
hence , @xmath11 is a representative of the difference @xmath895\cdot[f']^{-1}$ ] between two elements @xmath895 $ ] , @xmath896\in mod ( m)$ ] giving the correspondence @xmath4\to[\mathbf{h}']$ ] .
this completes the proof of the last statement .
this completes the proof .
this research was supported by bk21 plus snu mathematical sciences division . | let @xmath0 be an orientable , irreducible @xmath1-manifold admitting a weakly reducible genus three heegaard splitting as a minimal genus heegaard splitting . in this article , we prove that if @xmath2 $ ] , @xmath3\in mod(m)$ ] give the same correspondence between two isotopy classes of generalized heegaard splittings consisting of two heegaard splittings of genus two , say @xmath4\to[\mathbf{h}']$ ] , then there exists a representative @xmath5 of the difference @xmath6=[g]\cdot[f]^{-1}$ ] such that ( i ) @xmath5 preserves a suitably chosen embedding of the heegaard surface @xmath7 obtained by amalgamation from @xmath8 which is a representative of @xmath9 $ ] and ( ii ) @xmath5 sends a uniquely determined weak reducing pair @xmath10 of @xmath7 into itself up to isotopy .
moreover , for every orientation - preserving automorphism @xmath11 satisfying the previous conditions ( i ) and ( ii ) , there exist two elements of @xmath12 giving correspondence @xmath4\to[\mathbf{h}']$ ] such that @xmath11 belongs to the isotopy class of the difference between them . |
mass outflow , seen as blueshifted absorption in uv and x - ray spectra , is an important component of active galactic nuclei ( agns ; see recent review in * ? ? ?
this intrinsic absorption " is ubiquitous in nearby agns , appearing in over half of seyfert 1 galaxies having high - quality uv spectra obtained with the _ hubble space telescope _ ( _ hst _ ; * ? ? ? * ) and the _ far ultraviolet spectroscopic explorer _
( _ fuse _ ; * ? ? ? * ) .
spectra from the _ advanced satellite for cosmology and astrophysics ( asca ) _ identified x - ray warm absorbers " , seen as absorption edges , in a similar percentage of objects @xcite .
large total ejected masses have been inferred for these outflows , exceeding the accretion rate of the central black hole in some cases , indicating mass outflow plays an important role in the overall energetics in agns ( e.g. * ? ? ?
recent studies have recognized and explored the potential effect of outflows on all scales of the agn environment , from feeding the central supermassive black hole in agns @xcite , to influencing the evolution of the host galaxy @xcite and the metallicity of the intergalactic medium @xcite . measured ionic column densities provide the basis for interpretation of the physical nature of agn outflows .
detailed uv spectral studies over the past decade have shown measurements of these crucial parameters are often not straightforward .
analyses of absorption doublets and multiplets has revealed the absorbers typically only partially occult the background emission sources . without proper treatment of this effect , the column densities could be severely in error ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
additional complications that could affect column density measurements are different covering factors for different background emission sources @xcite , velocity - dependent covering factors ( e.g. * ? ? ?
* ) , and inhomogeneous distributions of absorbing material @xcite .
many recent investigations of agn outflows have focused on intensive multiwavelength observations of seyfert 1 galaxies .
seyferts are well suited for these studies because they include the brightest agns in the uv and x - ray .
the x - ray spectra contain the imprint of the bulk of the outflow s mass , which can now be deblended into individual absorption lines with the high - resolution capabilities of the _ chandra x - ray observatory ( cxo ) _ and _ xmm - newton space observatory _ , allowing detailed study .
the high quality uv spectra available with _ hst _ and _ fuse _ provide a complimentary , precise probe of the complex absorption troughs .
due to the relatively narrow absorption in seyfert outflows , the important uv doublets and multiplets are typically unblended , allowing measurements of these key diagnostic lines .
we have undertaken an intensive multiwavelength observing campaign with _
hst_/stis , _ fuse _ , and _ cxo _ to study the intrinsic absorption in the seyfert 1 galaxy mrk 279 .
mrk 279 was selected for this study because of its uv and x - ray brightness and the rich absorption spectrum in both bands , including unblended and well - resolved uv doublets ( see * ? ? ?
* hereafter sk04 ) .
additionally , it has minimal contamination by galactic absorption and a relatively weak contribution from a narrow emission line region ( nlr ) , both of which can complicate measurements of the absorption properties . as part of a series of papers devoted to this campaign
, we present here a detailed study of the uv absorption in the combined stis and _ fuse _ spectra .
we develop a new approach for measuring the covering factors and column densities in the absorbers , making full use of the high quality far - uv spectrum .
these measurements provide the foundation for subsequent analysis and interpretation of the mass outflow in mrk 279 , and provide novel geometric constraints . in parallel papers , we present analysis of the x - ray spectrum @xcite , inhomogeneous models of the uv absorption @xcite , and density diagnostics based on ov k - shell x - ray lines @xcite . in future papers ,
we will present photoionization models of the uv and x - ray absorption and analysis of absorption variability . in the next section ,
we describe the _ hst _ and _ fuse _ observations and present an overview of the absorption spectrum ; in 3 , we review the standard doublet technique for measuring intrinsic absorption and , together with an appendix , discuss important limitations of this method ; the formalism of our fitting method and results for mrk 279 are described in 4 ; in 5 , the fits are interpreted , and implications for physical constraints on the outflow are explored ; finally , a summary is presented in 6 .
the nucleus of mrk 279 was observed for a total of 41 ks ( 16 orbits ) with the space telescope imaging spectrograph ( stis ) on board _ hst _ between 2003 may 13 18 and for 91 ks with _ fuse _ between 2003 may 12 14 .
the stis observation used the e140 m grating , which covers 1150 1730 , and was obtained through the [email protected] @xmath2 [email protected] aperture .
the spectrum was processed with calstis v2.16 , which removes the background light from each echelle order using the scattered light model from @xcite .
low residual fluxes in the cores of saturated galactic lines indicate accurate removal of scattered light : typical fluxes in the cores are @xmath3 2.5% of the local unabsorbed continuum flux levels , and mean fluxes averaged over the absorption cores are @xmath4 3% of the noise in the troughs .
the final spectrum was sampled in 0.012 0.017 bins , thereby preserving the @xmath5 6.5 km s@xmath6 kinematic resolution of stis / e140 m .
we found that the standard pipeline processing did not yield a fully calibrated spectrum due to two effects : a ) the echelle ripple structure , due to the characteristic efficiency of the detector along each order , is not completely removed ( see * ? ? ?
* ) , and b ) the sensitivity of the mama detectors has degraded with time , and the change has not been incorporated in the pipeline for the echelle gratings . in order to correct for these effects in the mrk 279 spectrum , we used multiple spectra of the white dwarf spectrophotometric standard , bd+28 4122 , one of which was taken close in time to our observation .
first , a composite stellar spectrum of bd+28 4122 composed of fos and stis data @xcite was used to flux - calibrate a 1997 stis spectrum of bd+28 4122 that does not exhibit the echelle ripple structure .
this flux - calibrated spectrum was then used to correct a stis spectrum of bd+28 4122 , taken on 3 may 2003 with the same grating and aperture as our mrk 279 observation , and which does show the same ripple structure seen in the mrk 279 spectrum .
these two corrections were performed by dividing the fiducial spectrum for each step by the comparison spectrum , fitting a polynomial to the result , and then multiplying the comparison spectrum by that polynomial .
we used the polynomials from the second step described above to correct the spectrum of mrk 279 . to obtain a smooth correction for each order in the mrk 279 spectrum , we applied averages of the polynomials corresponding to the four adjacent orders .
although we were able to remove most of the echelle ripple structure in this way , some lower amplitude residual curvature remains in some orders .
however , the intrinsic absorption features measured in this study are well - corrected .
lrr 1 & 85 & 40 + 2 & -265 & 50 + 2a & -290 & 30 + 2b & -325 & 30 + 2c & -355 & 65 + 3 & -390 & 20 + 4 & -460 & 20 + 4a & -490 & 65 + 5 & -550 & 30 + the _ fuse _ spectrum , obtained through the 30@xmath7 30@xmath1 aperture , covers 905 1187 .
the spectrum was processed with the current standard calibration pipeline , calfuse v2.2.3 .
the eight individual spectra obtained with _
, from the combination of four mirror / grating channels and two detectors , were coadded for all exposures .
mean residual fluxes measured in the cores of saturated galactic lines are consistent with zero within the noise ( i.e. , standard deviation of the fluxes ) in the troughs of these lines , indicating accurate background removal .
the spectrum was resampled into @xmath5 0.02 0.03 bins to increase the signal - to - noise ratio ( s / n ) and preserve the full resolution of _ fuse _ , which is nominally @xmath5 20 km s@xmath6 .
to place the _ fuse _ and stis spectra on the proper wavelength scale , we followed the procedure described in sk04 .
the centroids of prominent , unblended galactic interstellar absorption lines were measured and used as fiducials in comparing to the galactic 21 cm hi line in the line - of - sight to mrk 279 @xcite .
the lines measured in the stis spectrum are consistent with the 21 cm hi line within the measurement uncertainties and thus required no correction .
the _ fuse _ spectrum showed substantial shifts relative to the adopted standard .
due to non - linear offsets in the wavelength scale , local shifts were measured and applied individually to each spectral region containing intrinsic absorption features . to normalize the absorption , we fit the total intrinsic ( i.e. , unabsorbed ) agn emission in the _ fuse _ and stis spectra over each intrinsic absorption feature .
this was done empirically by fitting cubic splines to unabsorbed spectral regions adjacent to the features , at intervals of @xmath5 5 .
we also derived models for the individual contributions of the different emission sources ( continuum and emission lines ) , since they are required for our analysis .
for the continuum source , we fit a single power law ( @xmath8 ) to the observed flux at two widely separated wavelengths that are relatively uncontaminated with absorption or line emission features , @xmath9 955 and 1500 .
after first correcting for galactic extinction ( using the extinction law of * ? ? ?
* with @xmath10 ) , we find a best - fit spectral index @xmath11 1.6 .
this power law model matches the few other line - free regions of the uv spectrum , i.e. @xmath12 1150 1180 , 1330 , and 1390 , to within a few percent .
thus , for the emission line model , we simply subtracted the continuum power law model from the empirical fit to the total emission . the full far - uv spectrum from our _ fuse _ and stis observations is shown in figure 1 .
the active nucleus in mrk 279 was in a relatively high flux state during this epoch ; the uv continuum flux was similar to a 2000 january _ fuse _ observation , and @xmath13 7 times stronger than in _ fuse _ and stis spectra obtained in 2002 may .
full treatment of these earlier observations is given in sk04 .
qualitatively , the intrinsic absorption spectrum in our new observations is similar to the earlier epochs ( although some important variations were detected that will be the subject of a later study ) . here
, we give a brief overview , and refer the reader to sk04 for a more thorough phenomenological discussion of the absorption .
absorption from a range of ions is detected around the systemic velocity of the host galaxy , between @xmath14600 to @xmath15150 km s@xmath6 ; we adopt the redshift for mrk 279 from sk04 , @xmath16 0.0305 @xmath17 0.0003 .
normalized absorption profiles for some of the prominent lines are shown in figure 2 .
the absorption is seen to be resolved into multiple distinct kinematic components at the resolution of stis e140 m and _ fuse _ , revealing striking differences in the kinematic structure of different ions .
low - ionization species appear in several narrow components ( see siiii in figure 2 , but also siii , cii , ciii , and niii in sk04 figures 7 14 ) . however , the more highly ionized ovi , nv , and civ doublets , which are the primary uv signatures of intrinsic absorption in agns , are much broader and have different centroid velocities .
the lyman lines exhibit the kinematic structure of the low - ionization components , but also appear in the lower outflow velocity region coinciding with the high - ionization lines , @xmath18300 to @xmath19200 km s@xmath6 .
profile is not plotted at @xmath20400 km s@xmath6 due to contamination with galactic absorption at these velocities .
[ fig2],width=321 ] we adopt the component numbering system from sk04 , which was based on the kinematic structure in ly@xmath21 . in figure 2 ,
dotted vertical lines mark the centroids of the components in sk04 that exhibit narrow absorption structure in low - ionization species in the current spectrum but which have no corresponding structure in the high - ionization cno doublets .
centroids of the components seen in the high - ionization lines are identified with dashed lines ; we have added component 2c to the sk04 system based on structure in the civ and nv profiles .
measured centroid radial velocities and widths of the components are listed in table 1 .
the differences in ionization and kinematic structure between these two groups of components strongly suggests they are physically distinct . based on their ionization , narrow widths , distinct centroid velocities , and ( in component 4 ) low density implied by the stringent upper limit on the cii column density in the excited fine - structure level ,
sk04 concluded at least some of the low - ionization components are not associated with the agn outflow .
instead , they posited they are associated with gas located at relatively large distances from the nucleus - perhaps from an interaction with the companion galaxy mcg@xmath1512 - 13 - 024 , high - velocity clouds associated with the host galaxy of mrk 279 , or , in the case of component 1 , in the interstellar medium of the host galaxy . in this study , we restrict our attention to the _ bona fide _ intrinsic absorption , i.e. , that presumed to be directly associated with an outflow from the agn .
we take this to include all absorption from the broad ovi , nv , and civ doublets ; figure 2 shows that any absorption associated with the narrow low - ionization components will at most only effect the outer wings of the outflow components in these lines .
conversely , figure 2 shows lyman line absorption from the low - ionization components is strong and heavily blended with the intrinsic absorption components ; thus , we limit our analysis of hi to the uncontaminated region , @xmath18300 to @xmath19200 km s@xmath6 .
we present here a brief review of the standard technique for measuring uv absorption in agn outflows and describe some limitations of this method to highlight the motivation for our method of analysis . in earlier studies of intrinsic absorption , the red members of doublet pairs were often found to be deeper than expected relative to the blue lines , based on their intrinsic 2:1 optical depth ratios . in many cases this was interpreted as due to partial coverage of the background nuclear emission by the absorbing gas , e.g. , @xcite , @xcite , @xcite ; ( other possibilities are scattering from an extended region and emission from an extended source unrelated to the central engine of the agn ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
if partial coverage is not accounted for , the absorption ionic column densities can be severely underestimated , which will dramatically affect the interpretation of the outflow .
the expression for the observed absorption that includes the effects of line - of - sight covering factor ( @xmath22 ) and optical depth ( @xmath23 ) is : @xmath24 where @xmath25 is the normalized flux , and all quantities are written as a function of radial velocity , @xmath26 . since the optical depths of the uv doublet pairs are in the simple 2:1 ratio , equation 1 can be solved for the covering factor and optical depths of each doublet @xcite .
the resulting expressions , which we will refer to as the doublet solution , are : @xmath27 @xmath28 where @xmath29 and @xmath30 subscripts identify the red and blue members of the doublet , and the equation for @xmath23 was derived by @xcite .
these expressions can be evaluated for unblended ( i.e. , sufficiently narrow ) absorption doublets with members that are individually resolved , and derived as a function of radial velocity . while this has provided a revolutionary advance in the study of intrinsic agn absorption ,
there are some key limitations to this method , described below .
an implicit assumption in the doublet solution is that the absorption is imprinted on a uniform , homogeneous background emission source , since it allows for the solution of only a single @xmath22 and @xmath23 .
however , the agn emission is comprised of multiple , physically distinct sources , i.e. , a continuum source and emission line regions ( including multiple kinematic components ) , which have different sizes , morphologies , and flux distributions .
thus , in cases where the absorber only partially occults the total background emission , the distinct sources would be expected to have different line - of - sight covering factors , in general .
this possibility was first explored by ganguly et al .
( 1999 ) for the continuum source and broad emission line region ( blr ) , and was demonstrated in the intrinsic absorption systems by ganguly et al . ,
@xcite , @xcite , and sk04 . to account for multiple discrete background emission sources , equation 1 can be expanded to give the normalized flux of the j@xmath31 line @xmath32,\ ] ] where the i@xmath31 individual emission source contributes a fraction @xmath33 $ ] to the total intrinsic flux and has covering factor @xmath34 .
the _ effective _ covering factor for each line is the weighted combination of individual covering factors , and can be written : @xmath35 these are expansions of the expressions given in @xcite to include an arbitrary number of emission sources .
the multi - component nature of the background emission has several important implications for the analysis of agn outflows : @xmath36 from equation 5 it can be seen that lines of the same ion could have different effective covering factors , which may introduce an error into the doublet equation .
this happens when the underlying emission fluxes differ from the spectral position of one line to the other and is illustrated in the nv doublet absorption shown in figure 3 . here , the emission line flux underlying the blue member is @xmath13 15% greater than under the red line , while the continuum flux under the two lines is identical .
the magnitude of this error depends on the slopes of the flux distributions between the doublet lines and differences in individual covering factors of the distinct emission sources .
@xcite showed this effect is typically small when considering the continuum / blr distinction , due to the gradual slope of the blr ; however , if the doublet lines are near saturation , it could have a very large effect since small flux differences correspond to large optical depth differences in these cases . also , an underlying narrow emission line component could have a pronounced effect on the solution @xcite .
- nv ( top panel ) and ly@xmath21 - ovi ( bottom panel ) , illustrating complexities in treating covering factors in intrinsic absorption measurements .
the continuum flux level , plotted as dashed lines , is essentially identical for each pair of lines , while the emission - line fluxes underlying each line differs greatly .
this has important implications for the relative _ effective _ covering factors .
additionally , the nature of the blr emission underlying these lines is complex : ly@xmath21 absorption lies on the high - velocity blue wing of ovi , while nv absorption has a contribution from the red wing of the ly@xmath0 blr.[fig3],width=321 ] @xmath36 without separation of the covering factors of the distinct sources , covering factors derived from a doublet pair can not be applied to measure column densities of other lines .
this is evident in figure 3 ; clearly the effective covering factor derived for nv is not applicable to ly@xmath0 if the individual continuum and emission - line covering factors differ since ly@xmath0 has much more underlying line flux .
similarly , ly@xmath21 , plotted in the bottom panel in figure 3 , will not generally have the same effective covering factor as ly@xmath0 , due to the different emission line contributions under each line .
however , if the individual covering factors are known , effective covering factors can be constructed for any line using equation 5 .
this is important for measuring singlet lines or contaminated multiplets that have no independent measure of the covering factor .
@xmath36 finally , the doublet solution misses potentially important , unique constraints on the absorption and emission geometry .
for example , combined with estimates of the sizes of the individual sources derived from other techniques , the individual covering factors constrain the size of the absorber , e.g. @xcite , and the relative location of the different emission components and absorber as projected on the plane of the sky .
the individual covering factors can also serve as a unique probe of more detailed geometry of the background emission .
consider for example the ovi
ly@xmath21 spectrum shown in figure 3 .
the ovi doublet absorbs its own emission line flux at the blueshifted velocity of the outflow ( @xmath37600 to @xmath19 200 km s@xmath6 ) , while the ly@xmath21 absorber sits primarily on the high - velocity blue wing of the ovi blr at @xmath372200 km s@xmath6 .
the ly@xmath21 line emission is relatively weak .
similarly , there is a contribution from the extreme red wing of the ly@xmath0 blr profile under the nv absorption .
constraints on the emission line covering factors for these lines could be used to probe the kinematic - geometric structure of the blr ; the absorber can thus serve as a filter to view and explore the background agn sources .
another limitation is that the doublet method always gives a solution , but it is often difficult to gauge its accuracy due to the non - linear dependency of the solution on measurement errors . to explore this ,
we have generated synthetic absorption profiles that include random fluctuations simulating spectral noise , and calculated @xmath22 and @xmath23 using the doublet equations .
illustrative results are shown in an appendix , whereas a complete quantitative treatment will be presented in a later paper .
we find there are systematic errors in the solutions that can give misleading results .
these errors are not random about the actual value , but rather systematically underestimate the actual covering factor , as seen in the appendix ; the discrepancy in the solution increases with weaker absorption doublets and decreased s / n .
additionally , @xcite demonstrated that the finite instrumental line spread function can lead to further systematic errors in the doublet solution .
motivated by the limitations of the traditional doublet method described above , we introduce here a different approach for measuring intrinsic absorption .
the underlying principle is to increase the number of lines that are simultaneously fit in order to ( a ) explore additional parameters contributing to the formation of observed absorption troughs and ( b ) overconstrain the set of equations .
this allows for the treatment of more complex , physically realistic scenarios of the absorption - emission geometry . by minimizing errors to simultaneous fits of multiple lines , noise in the spectrum
will generally be smoothed out , in contrast to the erratic behavior of the doublet solution demonstrated in 3.2 and the appendix .
our fitting algorithm employs the levenberg - marquardt non - linear least - squares minimization technique to solve equation 4 for specified absorption parameters ( @xmath34 , @xmath38)craigm / idl / idl.html , which is based on the minpack-1 optimization software of j. mor available at www.netlib.org ] .
it is similar in principle to that used in sk04 to analyze the lyman lines in earlier spectra of mrk 279 . given a total of n observed absorption lines ( @xmath39 ) as constraints , up to n@xmath191 parameters can be modeled .
no a priori assumptions are made about the kinematic distribution of the covering factors and optical depths of the absorbing material ( e.g. , gaussian ) indeed , one goal is to solve for the velocity - dependent absorption parameters to constrain the kinematic - geometric structure of the mass outflow .
thus , we derive fits to the absorption equations for each velocity bin .
this also avoids errors in the solutions resulting from averaging over variable profiles .
the algorithm minimizes the @xmath40 function , with each data point appropriately weighted by the 1 @xmath41 errors , which are a combination of spectral noise and estimated uncertainties in fitting the intrinsic underlying fluxes . for the latter ,
the continuum flux uncertainties were determined from the residuals between the power - law model and the line - free regions of the spectrum .
we estimated uncertainties in the emission - line fluxes by testing different empirical fits over the absorption features , finding the range that gave what we deemed reasonable line profile shapes .
the key requirement in employing this technique is to link multiple absorption lines for simultaneous fitting .
there are two general ways to do this : @xmath36 _ lines from the same energy level : _ the most straightforward way is to fit all available lines arising from the same ionic energy level , thereby eliminating uncertainties in ionic abundances or level populations .
if the relative underlying fluxes from distinct emission sources differs between the lines , the individual covering factors of those emission sources can be derived . in the subsequent analysis
, we fit all of the uncontaminated lyman series lines in mrk 279 .
these lines are ideal because they span a very large range in optical depth and have significantly different amounts of underlying emission - line flux ( see * ? ? ?
* ) . additionally , the full set of lines is accessible in low redshift agns with combined _ fuse _ and stis spectra .
we note the feii uv multiplets , which appear in a small fraction of agn absorbers @xcite , are another promising set of lines for this analysis .
@xmath36 _ global fitting approach : _ the second approach involves linking lines from different ions ( or any group of lines arising from different levels ) by placing physically motivated constraints on their absorption parameters . in our analysis of mrk 279
below , we fit the six combined lines of the ovi , nv , and civ doublets by assuming they share the same covering factors .
another potential application of this method is to link the absorption for a given line in spectra from different epochs via assumptions about the relative values of the absorption parameters between epochs ( e.g. , assuming the covering factors did not change ) .
the validity of the assumptions used to link the equations can then be tested by the result of the fit . for the intrinsic absorption in mrk 279 ,
we independently fit the two groups of lines described above : the lyman series lines and the combined cno doublets , i.e. global fit . for each set of lines , we tested two different models of the absorption covering factor . in model
a , a single covering factor was assumed to describe all lines , i.e. , no distinction was made between the different emission sources . in model b ,
independent covering factors for the continuum source and emission lines , @xmath42 and @xmath43 , were assumed . in this case , the general expressions in equations 4 and 5 reduce to those in @xcite . for the lyman lines ,
the solvable range is limited to @xmath19300 @xmath44200 km s@xmath6 , due to blending with the narrow , low - ionization components in the high velocity region of the outflow ( see 2.2 ) .
this is due to the failure of equation 4 where multiple absorption components with different covering factors contribute in the same velocity bin ; there is no straightforward way to disentangle how the different absorbers overlap as projected against the background emission sources .
figure 2 shows ly@xmath0 , ly@xmath21 , and ly@xmath45 give the best constraints for the lyman series analysis , exhibiting clean well - defined absorption profiles at relatively high s / n .
weak absorption in ly@xmath46 is also present in component 2 , while ly@xmath47 is contaminated with galactic h@xmath48 absorption and thus omitted from the fitting .
ly@xmath49 is not detected in components 2 2a within the limits of the spectral noise , thus to reduce the effect of noise on the solution , we set the normalized flux to unity for this line .
all lines of higher order than ly@xmath49 were omitted from the analysis since they exhibit no intrinsic absorption and provide no additional constraints .
thus , there are five lines as constraints to fit the two and three free parameters for the lyman solution in model a and b , respectively . for the global fitting of the ovi , nv , and civ doublets , the absorption equations were linked by assuming their covering factors are equal ( separately for the continuum and emission lines in model b ) .
the optical depth of each ion is a free parameter .
as described in 2.2 and seen in figure 2 , these lines are not strongly contaminated by the narrow , low - ionization absorption that dominates the lyman lines ; at most , there is only weak contamination in the wings of the broad , intrinsic absorption features by these low - ionization systems .
minimization for the single covering factor geometric model ( model a ) .
top panels show @xmath22 solutions , solved independently for the lyman series ( left panels ) and combined cno doublets ( right ) .
bottom panels show the hi column density and civ ( dashed histogram ) , nv ( dotted ) , and ovi ( solid ) optical depth solutions .
the contaminated , and thus unreliable , region of hi absorption is plotted with dotted histograms . [ fig4a],width=321 ] , width=321 ] , width=321 ] , width=321 ] , width=321 ] , width=321 ] best - fits to the covering factors and optical depths ( or equivalently column densities ) for both line groups are shown in figure 4 ( model a ) and figure 5 ( model b ) . in all cases ,
the parameter space search for the covering factors was restricted to the physically meaningful range , 0 @xmath50 1 .
the plotted error bars represent the formal 1 @xmath41 statistical errors in the best - fit parameters .
specifically , these correspond to values giving @xmath511 , and were computed from the diagonal elements of the covariance matrix for the optimal fit @xcite . for cases where the covering factor solution is a boundary value ( 0,1 ) , e.g. much of the @xmath42 solution for the cno doublets in model b ,
no covariance matrix elements are computed for that parameter since it is not a minimum in the solution . for these cases , we estimated uncertainties by deriving solutions for models keeping the parameter fixed , and finding the value giving @xmath511 from the best - fit solution at the boundary value . to ensure the computations did not erroneously stop at local minima ,
we generated solutions using different starting points for the parameter search space ; identical results were found in all cases . the fitted profiles , derived by inserting the best - fit solutions into equation 4 , for both models ( model a in blue , model b in red )
are compared with the observed normalized absorption profiles in figure 6 .
there are some possible errors in this fitting that should be mentioned .
first , the emission lines are treated as all arising from a single component .
line emission from distinct kinematic components ,
i.e. broad , intermediate , narrow line regions , that are covered by different fractions by the absorbers could introduce an error into the solution ( although , the narrow line region in mrk 279 is relatively weak ) .
also , there are cases where absorption features sit on the blr emission from different lines and thus sample different velocities of the blr profile ( i.e. , ly@xmath21 and nv described in 3.1 and shown in figure 3 ) .
this could introduce an error into the solution if there are spatial inhomogeneities in the blr gas as a function of velocity .
second , the area on the sky sampled by the _ fuse _ aperture is four orders of magnitude greater than the stis aperture .
while this has no consequences for the continuum and blr emission , which are unresolved and much smaller than the 0.@xmath12 stis aperture , any extended emission might effect the solutions .
this could include scattering of nuclear emission by an extended scatterer ( e.g. * ? ? ? * ) or extended ovi nlr emission .
finally , we have assumed the absorption optical depth of each line ( @xmath38 ) is uniform across the lateral extent of the emission sources , and thus the absorber is completely homogeneous ( see * ? ? ?
models departing from this assumption are presented in @xcite .
figure 6 shows both geometric models are able to match the intrinsic absorption features well at most outflow velocities , indicating the solutions for these models and parameters are degenerate over much of the profiles .
there are some regions fit somewhat better by model b , particularly civ in the low - velocity outflow component . however , there is additional , stronger evidence that supports the two covering factor model for the outflow , as outlined below .
@xmath36 _ consistent covering factor solutions from the independent lyman line and global cno doublet fitting methods : _ figure 5 shows that both the lyman and global cno doublet best - fit solutions are consistent with full coverage of the continuum source over most of the velocity range with a valid lyman solution .
further , the independent solutions for the two groups of lines are plotted together in figure 7 for a direct comparison ; the emission line covering factor profiles computed in model b are shown in the top panel and the single covering factors derived in model a are in the bottom panel .
the hi and cno @xmath43 fits are nearly identical over most of the core region of the absorption , with @xmath52 0.7 ( although in the wing of the absorption , @xmath53250 km s@xmath6 , the solutions diverge somewhat ) .
in contrast , the lyman method solution in model a is systematically less than the global doublet solution by about 0.1 0.15 at all velocities . ,
width=321 ] @xmath36 _ consistency with the emission source sizes : _ the continuum and emission line covering factor solutions in model b are also physically consistent with our understanding of the sizes of those emission sources .
reverberation mapping studies of agns have shown the blr is substantially larger than the ionizing continuum source .
these studies measure the blr to be several to tens of light - days across ( e.g. * ? ? ?
* ; * ? ? ?
* ) , while the uv continuum source is likely at least an order of magnitude smaller in size and possibly much smaller ( e.g. * ? ? ?
* ; * ? ? ?
thus , the solution of a fully covered continuum source and partial coverage of the emission lines is consistent with the nuclear emission geometry .
the fact that this result was arrived at separately with the independent lyman and global cno fits is compelling .
@xmath36 _ variability in ly@xmath0 absorption : _ in figure 8 , the normalized ly@xmath0 profile from our observation is compared with the stis spectrum obtained a year earlier , showing the absorption was shallower in the previous epoch by @xmath5 0.15 in normalized flux units .
this variability is not due to a lower h i column density ; the strength of ly@xmath21 in a contemporaneous _ fuse _ observation indicates ly@xmath0 was saturated in the earlier epoch ( sk04 ) , as it is in the 2003 spectrum , hence the profiles in both epochs simply delineate the unocculted flux .
thus , a change in covering factor must be responsible for the observed variability .
noting that the emission line - to - continuum flux ratio was greater in the 2002 spectrum than in 2003 , we tested if the difference in covering factor of the respective sources could explain the variability . to this end , we constructed a synthetic absorption profile for ly@xmath0 in the 2002 epoch based on the results from our analysis of the 2003 spectrum .
specifically , we set @xmath541 at all velocities , and solved for @xmath43 in the 2003 ly@xmath0 profile .
the observed continuum and emission line fluxes in the 2002 spectrum were then weighted by these covering factors according to equation 4 .
the resulting model profile , shown in the bottom panel of figure 8 , matches the observed absorption very well .
this provides a natural explanation for the variability in the ly@xmath0 absorption profile : the covering factors of each source individually remain the same , but a change in the _ effective _ covering factor occurs due to different relative strengths of the distinct background emission components . absorption .
the top panel shows ly@xmath0 absorption in components 22a was shallower in a may 2002 stis spectrum ( solid histogram ) than in may 2003 ( dashed ) . the bottom panel shows a model of the 2002 profile ( dotted ) , constructed with individual covering factors derived from the 2003 spectrum ( figure 5 ) , matches the observed profile well
this indicates the variation in absorption depth is likely due to a change in _ effective _ covering factor resulting from different relative strengths of the continuum and emission - line fluxes in the two epochs.[fig8],width=321 ] the covering factor and optical depth solutions for civ , nv , and ovi from the two-@xmath22 global fit ( model b ) are compared with the single doublet solutions ( equations 2 and 3 ) in figure 9 .
the global fit effective covering factors are weighted combinations of the individual covering factors shown in figure 5 , derived using equation 5 . in figure 9 ,
solid red lines show @xmath22 profiles for the short - wavelength members of each doublet and dotted lines show results for the long - wavelength lines ; they are nearly identical in all cases , indicating negligible errors in the doublet solution due to different contributions from underlying line emission ( see first bullet in 3.1 ) .
, width=321 ] , width=321 ] figure 9 shows the solutions from the two methods differ significantly in some regions , with important implications for the kinematic - geometric and -ionization structure in the outflow .
for example , the doublet covering factor solution , plotted in black , exhibits much more velocity - dependence than the global fit .
this is seen , for example , in the wings of civ and nv in component 4a and the red wing of component 2 , where @xmath22 decreases strongly in the wings , while the optical depth profiles show little or no relation to the absorption trough structure when compared with figure 2 .
further , the kinematic structure in these ions is much different from the ovi solution . in these cases ,
the doublet solution implies the observed absorption profiles are determined almost entirely by velocity - dependent covering factor .
additionally , in some regions , particularly in civ , the doublet solutions for @xmath22 and @xmath23 exhibit sharp fluctuations over small velocity intervals that do not coincide with structure in the observed absorption features , nor with the solutions from the other ions ( e.g. , components 2a 2c , and 4a ) .
comparison with the observed profiles in figure 2 shows these effects in the doublet solution are strongly correlated with the absorption strength .
smaller derived covering factors are associated with weaker absorption , both in the wings of individual kinematic components and in overall absorption features , such as civ components 2c and 2a . as discussed in the appendix and 3.2 ,
synthetic absorption profiles show the doublet solution suffers from systematic errors consistent with this general trend
spectral noise causes an underestimate of the covering factor .
the simulations we present in the appendix indicate the solutions become unreliable for cases where @xmath55 1 in the blue doublet member , and increasingly so for lower @xmath23 .
this , combined with the large , seemingly random fluctuations seen in certain regions suggests some features in the doublet solutions in figure 9 are artifacts due to noise and not real features in the outflow .
however , one puzzling aspect of the solutions , in comparison to the results in the appendix , is that there are not more negative values for the covering factors .
the simulations predict that for sufficiently low optical depth , there is a high probability that @xmath560 .
this should be especially apparent for intermediate values of @xmath23 , such as the middle two models presented in the appendix , where dramatic fluctuations between large positive and negative values are expected .
it is unclear why this is not more prevalent in the solutions in figure 9 . in table 2 , we compare integrated column densities measured from the different solutions : the two geometric coverage models presented in 4.2 , plus the traditional single doublet solution for the cno doublets .
the integrated column densities for the doublet solution are larger than the global fits in all cases . in component 2@xmath152a ,
the doublet solution is 60% greater than the 2-@xmath22 model for civ , while in component 4a , it is 45% and 70% greater for civ and nv , respectively .
comparing the two global fit models , the cno column densities from model b are greater than the single @xmath22 model by 0 30% , while the hi column density is 50% less . here
we explore constraints on the nuclear absorption and emission geometry available from the covering factor solutions derived in 4.2 .
full analysis of the physical conditions in the outflow , using photoionization modeling of the combined uv and x - ray absorption , will be presented in a future study . as discussed in 5.1 ,
we adopt the 2-@xmath22 global fit ( model b ) for the covering factors and column densities .
the covering factor solutions in figure 5 are consistent with the uv continuum source being fully in our sight - line through the absorber , for all kinematic components , while the emission lines are only partially so .
this constrains the relative line - of - sight geometry of the absorbers and nuclear emission sources and places a lower limit on the transverse size of the outflow .
monitoring of time lags between blr and continuum variations in mrk 279 by @xcite gives the size of the blr , @xmath57 12 light days .
thus , the projected size of the uv absorbers on the plane of the sky , against the agn emission , is at least @xmath58 , or @xmath1310 light days in the cores of the absorption components , and possibly decreasing in the wings .
lcccccccc c iv & 1.6 & 0.9 & 2.1 & 1.1 & 3.3 & 1.6 & 1.5 & 0.9 +
n v & @xmath596.6 & 2.1 & @xmath596.6 & 2.1 & @xmath598.1 & 3.6 & 4.7 & 2.0 + o vi & @xmath5914 & @xmath5913 & @xmath5914 & @xmath5913 & @xmath5914 & @xmath5913 & 9.8 & 9.5 + h i & 7.5 & & 5.0 & & & & 4.4 & + an additional probe of the nuclear geometry is possible because the ly@xmath21 absorption lies on the high - velocity blue wing of the ovi blr profile , at @xmath372200 km s@xmath6 , as discussed in 3.1 and illustrated in figure 3 .
thus , it samples different blr kinematics than most other absorption lines , which absorb blr velocities coinciding with the absorption outflow velocities ( @xmath37300 km s@xmath6 ) .
therefore , comparison of the ly@xmath21 emission - line covering factor and those associated with other lines serves as a probe of the kinematic - spatial structure of the blr .
the potential affect of complex velocity - dependent structure in the blr on the absorption covering factors was explored in @xcite . since we assumed a priori that all lines share the same individual covering factors in the lyman series fit in 4.2
, a legitimate comparison requires obtaining independent constraints on the covering factor of the ovi blr blue wing by the ly@xmath21 absorber .
this comes straightforwardly from the observed ly@xmath45 absorption line , since ly@xmath45 has no underlying line emission . using the result that @xmath541 , the intrinsic optical depth ratio for ly@xmath21 : ly@xmath45 , and the observed ly@xmath45 profile , the absorption profile for ly@xmath21 can be derived as a function of @xmath43 from equation 4 .
some illustrative results are shown in figure 10 .
this shows , for example , that both an unocculted and fully occulted high velocity ovi blr are ruled out by the data .
we find that values ranging from 0.5 @xmath60 0.8 are required to fit the majority of the ly@xmath21 absorption in components 2 2a .
this is similar to the emission line covering factor derived from the global fit to the cno doublets ( figure 5b ) .
additionally , the ly@xmath0 @xmath43 profile can be derived independently for comparison with ly@xmath21 , in a similar manner as for ly@xmath21 ( i.e. , using @xmath541 , the observed ly@xmath45 profile , and the ly@xmath0 : ly@xmath45 @xmath23 ratio ) .
the result is identical to the solution to the combined lyman lines in components 2 2a shown in figure 5a within uncertainties .
therefore , the absorption covering factor of the ovi blr emission at @xmath372200 km s@xmath6 is similar to @xmath43 at lower emission - line velocities ( by the cno doublets and ly@xmath0 ) .
these results may provide constraints for models of the blr , e.g. , testing disk vs spherical geometries and outflow vs rotational kinematics for the blr .
we have presented a study of the intrinsic uv absorption in the seyfert 1 galaxy mrk 279 from an analysis of combined long observations with _ hst_/stis and _ fuse_. these spectra were obtained simultaneously in may 2003 as part of an intensive multiwavelength observing campaign .
we present a review of the standard technique for measuring intrinsic uv absorption parameters based on individual doublet pairs , showing some key limitations of this method : 1 ) it can not treat multiple background emission sources .
this introduces a potential error in the solution and misses important geometric constraints on the outflow .
2 ) using synthetic absorption profiles , we show it systematically underestimates the covering factor ( and overestimates @xmath23 ) in response to spectral noise .
the discrepancy in the solution is shown to be strongly dependent on absorption strength . to measure the uv absorption parameters in mrk 279 , we independently fit two groups of lines : the lyman series lines and the combined cno lithium - like doublets .
the doublet fitting involved a global fitting approach , which assumes the same covering factors apply to all ions . by increasing the number of lines that are
simultaneously fit , more complex and physically realistic models of the absorption - emission geometry can be explored .
solutions for two different geometrical models , one assuming a single covering factor for all background emission and the other separate covering factors for the continuum and emission lines , both give good statistical fits to the observed absorption
. however , several lines of evidence support the model with two covering factors : 1 ) the independently fit lyman lines and cno doublets give similar solutions to the covering factors of both emission sources ; 2 ) the fits are consistent with absorbers that fully occult the continuum source and partially cover the emission lines , consistent with the relative sizes of the emission sources ; and 3 ) observed variability in the ly@xmath0 absorption depth can be explained naturally by this model as a change in effective covering factor resulting from a change in the relative strengths of the emission components .
comparison of the traditional solutions based on individual doublets and the global - fit solutions shows the former exhibits much stronger velocity dependence .
this is seen as decreases in covering factor in the wings of individual kinematic components , and as peculiar fluctuations in both @xmath23 and @xmath22 in other regions of relatively weak absorption . in light of the systematic errors shown to be inherent in the individual doublet solution ,
we conclude some of these effects are likely artifacts of the solution and should be interpreted with caution .
the covering factor solutions from our global fit constrains the relative line - of - sight geometry of the absorbers and nuclear emission sources .
the derived emission line covering factor , combined with the size of the blr , constrains the projected size of the absorber to be @xmath61 10 light days .
we utilize the coverage of the high velocity ovi blr by the ly@xmath21 absorber to explore kinematic structure in the blr ; we find no evidence for dependence of the absorber s blr covering factor on the blr velocity .
support for this work was provided by nasa through grants hst - ar-9536 , hst - go-9688 , and nags 12867 and through _ chandra _ grant 04700532 .
we thank d. lindler and j. valenti for their assistance in correcting the stis spectrum and c. markwardt for making his software publicly available .
we also thank the referee p. hall for comments that helped clarify and improve this study .
here , we address how the expressions for covering factor and optical depth from the doublet solution ( equations 2 and 3 ) depend on noise in the spectrum .
we have generated synthetic absorption profiles for doublet pairs that include random fluctuations to simulate spectral noise .
we derive the @xmath22
@xmath23 solutions for these synthetic profiles to determine any trends in the solution as a function of noise level and absorption strength .
our synthetic profiles were derived in the following way .
the optical depth profiles were assumed to be gaussian , parameterized by the width ( @xmath41 ) and peak optical depth in the core of the blue line ( @xmath62 ) .
the true " normalized absorption profiles for the lines are then derived from equation 1 , with the covering factors set at a constant value across the profile for each doublet pair , and @xmath63/2 at all velocities .
thus , we have assumed in effect a single background emission source to avoid the complications described in 3.1 .
for comparison with our study of mrk 279 , we have set the velocity resolution of our synthetic profiles to that of the stis e140 m grating , and the absorption width ( @xmath41 = 50 km s@xmath6 ) to be approximately consistent with intrinsic absorption components 22c . to simulate spectral noise
, we generated a normally distributed random number associated with each velocity bin in the synthetic spectra .
the noise level was normalized by selecting the desired s / n in the unabsorbed continuum and then weighting the noise by flux level , s / n@xmath64 , according to poisson statistics .
we generated profiles for a range of s /
n , @xmath22 ( real ) , and @xmath62 . here , we give some brief illustrative results , while reserving a full analysis for a later study .
the left panels in figure 11 show synthetic profiles for doublet pairs with @xmath65 2 , 1 , 0.5 , and 0.25 . in all cases ,
@xmath660.8 and the noise level was normalized to be 3% in the continuum .
the middle two panels show the corresponding covering factor and optical depth profiles derived directly from equations 2 and 3 , with the actual values marked with dashed lines for comparison .
errors in the doublet solution are immediately apparent .
the covering factors are systematically underestimated , and the magnitude of error is strongly dependent on absorption strength .
this is seen both in the lower covering factors derived in the cores of features with lower @xmath62 ( means and standard deviations measured over the central 150 km s@xmath6 are printed in each plot ) , and in the decrease in @xmath22 computed in the wings of each profile .
these systematic errors are due to non - linear effects in the doublet equations .
this is seen most clearly by comparing the numerator , @xmath67 , and denominator , @xmath68 , in the expression for covering factor ( equation 2 ) .
the right panels of figure 11 show the values of @xmath69 and @xmath70 ( in red ) , which would be identical in each velocity bin for infinite s / n . due to the forms of @xmath71 and @xmath69 , these quantities have very different dependences on noise ; @xmath72 becomes much smaller than @xmath73 for weak absorption and , at sufficiently small @xmath23 , @xmath71 is less than the noise level of @xmath69 .
as @xmath71 decreases relative to the noise in @xmath69 for weaker absorption , the probability that @xmath74 becomes vanishingly small , and the average value of @xmath75 becomes increasingly small .
these errors could have pronounced effects on the interpretation of the outflow .
each solution that underestimates @xmath22 overestimates @xmath23 .
thus , ionic column densities are systematically overestimated , with increasing relative discrepancies in weaker doublets , leading to errors in determining the ionization structure and total gas in the absorber via photoionization models .
additionally , the errors in covering factor solutions will effect geometric inferences .
for example , due to the high - ionization state of agn outflows , lower - ionization species appearing in uv spectra are generally weaker .
thus , the increasing discrepancy in weaker absorption doublets may lead to the misinterpretation of ionic - dependent covering factors .
also , weaker absorption in the wings of an absorption feature could lead to apparent velocity - dependent covering factors that are instead due to optical depth variations , or at least exaggerate the effect .
arav , n. , et al .
2004 , , in press arav , n. , korista , k. t. , & de kool , m. 2002 , , 566 , 699 arav , n. , becker , r. h. , laurent - muehleisen , s. a. , gregg , m. d. , white , r. l. , brotherton , m. s. , & de kool , m. 1999 , apj , 524 , 566 blandford , r , d. , & begelman , m. c. 2004 , mnras , 349 , 68 blandford , r , d. , & begelman , m. c. 1999 , mnras , 303 , l1 barlow , t. a. , & sargent , w. l. w. 1997 , , 113 , 136 bevington , p. r. 1969 , data reduction and error analysis for the physical sciences , pp 242 - 245 bohlin , r. c. , dickinson , m. e. , & calzetti , d. 2001 , , 122 , 2118 cardelli , j. a. , clayton , g. c. , & mathis , j. s. 1989 , , 345 , 245 cavaliere , a. , lapi , a. , & menci , n. , 2002 , apj , 581 , l1 cohen , m. h. , ogle , p. m. , tran , h. d. , vermeulen , r. c. , miller , j. s. , goodrich , r. w. & martel , a. r. 1995 , apj , 448 , l77 costantini , e. , et al .
2004 , , in press crenshaw , d. m. , kraemer , s. b. , & george , i. m. 2003 , aara , 41 , 117 crenshaw , d. m. , kraemer , s. b. , boggess , a. , maran , s. p. , mushotzky , r. f. , & wu , c .- c . 1999 , , 516 , 750 de kool , m. , korista , k. t. , & arav , n. 2002 , apj , 580 , 54 de kool , m. , arav , n. , becker , r. , laurent - muehleisen , s. a. , white , r. l. , price , t. , gregg , m. d. 2001 , apj , 548 , 609 gabel , j. r. , kraemer , s. b. , & crenshaw , d. m. 2004 , in agn physics with the sloan digital sky survey , eds . g. t. richards & p. b. hall ( san francisco : asp ) , p239 gabel , j. r. , et al . 2003 , apj , 583 , 178 ganguly , r. , eracleous , m. , charlton , j. c. , & churchill , c. w. 1999 , , 117 , 2594 george , i. m. , turner , t. j. , netzer , h. , nandra , k. , mushotzky , r. f. , & yaqoob , t. 1998 , , 114 , 73 goodrich , r. w. , & miller , j. s. 1995 , , 448 , l73 hall , p. b. , hutsemekers , d. , anderson , s. f. , brinkmann , j. , fan , x. , schneider , d. p. , york , d. g. 2003 , , 593 , 189 hamann , f. , barlow , t. a. , junkkarinen , v. , & burbidge , e. m. 1997 , , 478 , 80 heap , s. r. & brown , t. m. 1997 , in the 1997 hst calibration workshop with a new generation of instruments , ed . s. casertano ( baltimore : stsci ) , 114 kaastra , j. s. , et al .
2004 , , in press kaspi , s. , smith , p. s. , netzer , h. , maoz , d. , jannuzi , b. t. , & giveon , u. 2000 , apj , 533 , 631 kriss , g. a. 2002 , in asp conf .
255 , mass outflow in active galactic nuclei : new perspectives , ed .
d. m. crenshaw , s. b. kraemer , & i. m. george ( san francisco : asp ) , 69 kraemer , s. b. , et al .
2001 , , 551 , 671 kraemer , s. b. , crenshaw , d. m. , george , i. m. , netzer , h. , turner , t. j. , & gabel , j. r. 2002 , apj , 577 , 98 laor , a. & netzer , h. 1989 , mnras , 238 , 897 lindler , d. , & bowers , c. 2000 , baas , 197 , 1202 maoz et al .
1990 , apj , 351 , 75 peterson , b. m. , wanders , i. , bertram , r. , hunley , j. f. , pogge , r. w. , wagner , r. m. 1998 , apj , 501 , 82 proga , d. 2003 , apj , 585 , 406 reynolds , c. s. 1997 , mnras , 286 , 513 scannapieco , e. , & peng ho , s. , 2004 , apj , submitted , astro - ph/0401087 al .
2004 , apjs , 152 , 1 ( sk04 ) silk , j. , & rees , m. j. , 1998 , a&a 331 , l1s srianand , r. , & shankaranarayanan , s. , 1999 , , 518 , 672 wakker , b. p. , kalberla , p. m. w. , van woerden , h. , de boer , k. s. , & putman , m. e. 2001 , apjs , 136 , 537 wampler , e. j. , bergeron , j. , & petitjean , p. 1993 , a&a , 273 , 15 | we present an analysis of the intrinsic uv absorption in the seyfert 1 galaxy mrk 279 based on simultaneous long observations with the _ hubble space telescope _ ( 41 ks ) and the _ far ultraviolet spectroscopic explorer _ ( 91 ks ) . to extract the line - of - sight covering factors and ionic column densities , we separately fit two groups of absorption lines : the lyman series and the cno lithium - like doublets . for the cno doublets
we assume that all three ions share the same covering factors .
the fitting method applied here overcomes some limitations of the traditional method using individual doublet pairs ; it allows for the treatment of more complex , physically realistic scenarios for the absorption - emission geometry and eliminates systematic errors that we show are introduced by spectral noise .
we derive velocity - dependent solutions based on two models of geometrical covering a single covering factor for all background emission sources , and separate covering factors for the continuum and emission lines .
although both models give good statistical fits to the observed absorption , we favor the model with two covering factors because : ( a ) the best - fit covering factors for both emission sources are similar for the independent lyman series and cno doublet fits ; ( b ) the fits are consistent with full coverage of the continuum source and partial coverage of the emission lines by the absorbers , as expected from the relative sizes of the nuclear emission components ; and ( c ) it provides a natural explanation for variability in the ly@xmath0 absorption detected in an earlier epoch .
we also explore physical and geometrical constraints on the outflow from these results . |
in human societies , cooperation is essential for the maintenance of public goods .
however , the collapse of cooperation happens often in many public goods dilemmas which we nowadays face , like protecting the global climate or avoiding overfishing of our oceans @xcite . for avoiding the tragedy of the commons @xcite
, we often rely on institutions to enforce public cooperation and acceptable behavior .
although institutionalized punishment appears to be more common than institutionalized reward , both concepts are in use throughout the world .
recently , ample research efforts have been devoted to the study of the emergence of institutions and their effectiveness in promoting prosocial behavior @xcite .
it has been shown , for example , that institutional rewarding promotes the evolution of cooperation in the liner public goods game @xcite , the nonlinear public goods games @xcite , and in structured populations in general @xcite .
however , institutional punishment is less costly and thus more effective to warrant a given level of public cooperation , especially if participation in the public goods game is optional @xcite . besides the obvious
stick versus carrot dilemma @xcite , the question emerges how to best make use of the available resources , which inevitably are finite @xcite . in particular , we wish to make optimal use of different forms of reciprocity to promote human cooperation @xcite .
one plausible approach appears to be allocating the resources depending on the properties of the interaction network that describes the connections among us .
surprisingly , few studies have thus far considered the problem of the optimal allocation of incentives for maximizing public cooperation .
traditionally , all groups and all individuals are considered equal , and depending on their strategies thus deserved of the same reward or punishment @xcite .
this simple assumption , however , does not agree with the fact that in social networks individuals have different roles , which depend significantly on the degree of the node that they occupy . indeed , the prominent role of heterogeneous interaction networks for the successful evolution of cooperation is firmly established and well known @xcite , and the reasonable assumption is that the incentives should likely also be distributed accordingly for optimal evolutionary outcomes . in the framework of evolutionary graph theory
the interaction groups are diverse , and naturally thus the provided incentives within each group should also be different .
the number of links an individual player has is traditionally assumed to be a good proxy for that player s influence and importance . in this sense , it is interesting and highly relevant to determine how to distribute the incentives in the light of this heterogeneity .
here we consider the spatial public goods game @xcite with institutional reciprocity on a scale - free network @xcite , where the assumption is that the incentives at disposal for rewarding cooperators and punishing defectors are limited .
we assume that the budget available to each group is proportional to its size , and that the distribution of the incentives depends on the number of individual links within the group .
our aim is to arrive at a thorough understanding of how the incentives should be best distributed to maximize public cooperation . in what follows , we present the results obtained with the model described in the methods section , to where we refer for details .
as we will show , if the enhancement factor @xmath0 is small , the level of cooperation can be maximized simply by adopting the simplest `` equal distribution '' scheme . if the value of @xmath0 is large , however , it is best to reward high - degree nodes more than low - degree nodes . unlike for institutionalized rewards ,
the optimal distribution of resources within the framework of institutional punishment depends on whether absolute or degree - normalized payoffs are used .
high - degree nodes should be punished more lenient than low - degree nodes if degree - normalized payoffs apply , while high - degree nodes should be punished stronger than low - degree nodes if absolute payoffs count .
we perform monte carlo simulations of the public goods game described in the methods section , whereby we consider separately institutional rewarding with absolute payoffs and institutional punishment with absolute payoffs , as well as institutional rewarding with degree - normalized payoffs and institutional punishment with degree - normalized payoffs . as the key parameters we consider the enhancement factor @xmath0 , the average amount of available incentives @xmath1 , and the distribution strength of incentives @xmath2 ( see methods for details ) .
we determine the stationary fractions of cooperators in the stationary state on networks comprising @xmath3 to @xmath4 nodes .
the final results are averaged over @xmath5 independent initial conditions to further enhance accuracy .
figure [ fig1 ] shows the stationary fraction of cooperators for two different values of the enhancement factor @xmath0 . when the enhancement factor is small [ fig .
[ fig1](a ) and ( c ) ] , defectors always dominate if @xmath6 , and this regardless of the value of @xmath2 . for intermediate values of @xmath1 ,
the cooperation level can be maximized at an intermediate value of the distribution strength @xmath2 , which ought to be close to zero .
this indicates that an equal distribution of positive incentives , regardless of the degree of players within the group , is the optimal distribution scheme for public cooperation . for high values of @xmath1 , the cooperation level increases with increasing the value of @xmath2 .
if the enhancement factor @xmath0 increases [ fig .
[ fig1](b ) and ( d ) ] , defectors still dominate for small values of @xmath1 and regardless of the value of @xmath2 .
however , the nonmonotonous dependence of the fraction of cooperators on the distribution strength @xmath2 disappears for intermediate values of @xmath1 . instead , the highest cooperation level is attainable for large values of @xmath2 . intuitively , it is possible to understand that when the enhancement factor is small , a modest positive incentive is not enough to reverse the doom of cooperators , no matter which distribution scheme is used @xcite .
conversely , if the incentives are large and targeted preferentially towards influential players , they can have a high payoff even if the part stemming from the public goods game is small . in agreement with the traditional argument of network reciprocity @xcite , only cooperators are able to forge a long - term advantage out of this favorable situation and build sizable cooperative groups .
thus , for high - enough values of @xmath2 , which favor the distribution of rewards towards high - degree nodes , the evolution of public cooperation is successful . to gain an understanding of the optimal intermediate value of the average amount of available incentives @xmath1 requires more effort
first , we show in fig .
[ fig2 ] the payoff differences between cooperators and defectors as well as the fraction of cooperators as a function of degree @xmath7 during different typical evolutionary stages of the game .
we observe that for @xmath8 , which implies an `` equal distribution '' scheme irrespective of the degree of players , the payoff of cooperators is higher than that of the corresponding defectors , and this regardless of @xmath7 .
thus , cooperators can successfully occupy all nodes of the networks .
in contrast , for negative values of @xmath2 , the payoff of cooperators with high or middle degree is less than that of the corresponding defectors , while cooperators with low degree have a higher mean payoff than defectors with small degree . because there are interconnections among different types of nodes , and because the fermi strategy updating rule is adopted , cooperators can coexist with defectors at equilibrium . but defectors can occupy most of the nodes in the network , since low - degree cooperators do not obtain a high enough payoff to spread their strategy across the network . for positive values of @xmath2 , the payoff of cooperators with high and middle degree is larger than that of the corresponding defectors , while cooperators with low degree have a lower mean payoff than defectors with low degree .
in addition , high - degree cooperators obtain a sufficiently high payoff through institutional rewarding that enables them to spread the cooperative strategy also towards some of the auxiliary low - degree nodes .
accordingly , cooperative behavior prevails over defection , but the stationary state is still a mixed @xmath9 phase cooperators are unable to dominate completely . to further corroborate our arguments ,
we show in fig .
[ fig3 ] the payoff difference between cooperators and defectors as well as the fraction of cooperators in dependence on degree @xmath7 , as obtained for a two times larger value of @xmath0 than used in fig .
[ fig2 ] . in comparison
, it can be observed that for @xmath10 the results remain unchanged . for @xmath11 ,
on the other hand , the process of evolution is different from what we have presented in fig .
[ fig2 ] . during the early stages of evolution [ fig .
[ fig3 ] ( a ) and ( b ) ] , cooperators with low degree can have a lower mean payoff than low - degree defectors , while cooperators with middle and high degree can have a higher mean payoff than the corresponding defectors .
further in time , cooperators succeed in occupying all high - degree nodes [ fig .
[ fig3 ] ( g ) ] , and even low - degree cooperators have a payoff comparable to that of low - degree defectors [ fig . [ fig3 ] ( c ) ] .
cooperators can eventually invade the whole network [ fig .
[ fig3 ] ( h ) ] , thus giving rise to the absorbing @xmath12 phase at @xmath13 , which emerges for sufficiently large values of @xmath2 and intermediate values of @xmath1 .
figure [ fig4 ] shows that when the enhancement factor is small , cooperators are unable to survive for small values of @xmath1 , and this irrespective of the value of @xmath2 . for intermediate values of @xmath1 ,
the highest cooperation level is attained at an intermediate value of the distribution strength @xmath2 , which is almost equal to zero , like by the consideration of institutional reward in the preceding subsection .
this indicates that , for small enhancement factors , in case of institutional punishment too an `` equal distribution '' scheme works best for the evolution of public cooperation .
if @xmath1 is large if resources for punishment abound cooperators can always dominate , regardless of the value of @xmath2 . for a two times larger value or @xmath0 cooperators are favored even more , so that the nonmonotonous dependence of the cooperation level on @xmath2 at intermediate values of @xmath1 vanishes .
instead , the fraction of cooperators simply increases with increasing values of @xmath2 .
thus , the more the high - degree defectors are punished , the better for the evolution of public cooperation .
it is understandable that low values of @xmath0 , paired with modest resources for punishing defectors , lead to the dominance of defectors , regardless of the value of @xmath2 .
conversely , if the resources abound , defectors are punished severely and cooperators dominate . in this limit example the distribution of fines between low , middle and high degree nodes does not play an important role . if , however , the combination of values of @xmath0 and @xmath1 just barely , or not at all , support the survivability of cooperators , then the value of @xmath2 , and thus the particular distribution of incentives ( in this case fines ) , plays a significant role . with the aim to explain this nontrivial dependence on @xmath2
, we show in fig .
[ fig5 ] the payoff difference between cooperators and defectors as well as the fraction of cooperators as a function of degree @xmath7 during different stages of the evolutionary process .
it can be observed that for @xmath8 cooperators can always have a higher payoff than the defectors with the same corresponding degree [ fig .
[ fig5 ] ( a)-(c ) ] .
cooperators can thus rise to complete dominance . while for @xmath14 , however , low - degree cooperators can have a higher mean payoff than low - degree defectors , while cooperators with middle or high degree ca nt match the corresponding defectors during the early stages of the evolution [ fig .
[ fig5 ] ( a ) ] .
defectors can therefore , over time , occupy the high - degree and middle - degree nodes [ fig .
[ fig5 ] ( f ) ] .
this invasion can decrease the fraction of cooperators on low - degree nodes @xcite .
accordingly , the payoff difference between cooperators and defectors on these nodes continues to be negative , although low - degree defectors receive the negative incentives .
ultimately cooperators therefore die out .
for @xmath11 , on the other hand , defectors with higher degree are punished preferentially they receive a bigger share of fines from the available fond than low - degree defectors . due to the small enhancement factor and the institutional punishment ,
both high - degree cooperators and high - degree defectors have negative payoffs .
in fact , low - degree players can have a higher payoff than high - degree players , despite of the fact that we use absolute payoffs in this particular case . either way
, defectors can easily invade low - degree nodes [ fig .
[ fig5](f ) ] , and they can spread further towards middle and high degree nodes , although at the beginning of evolution cooperators have a higher payoff than defectors on these nodes .
the ultimate consequence is that defectors dominate completely [ fig .
[ fig5 ] ( h ) ] .
it remains of interest to explain why the dependence on @xmath2 disappears at intermediate values of @xmath1 for @xmath13 .
for this purpose , we show in fig .
[ fig6 ] the same quantities as in fig .
[ fig5 ] , from where it follows that the results do not change for @xmath10
. however , for @xmath11 the differences are clearly inferable . for @xmath13
the high - degree cooperators can obtain a positive payoff , and naturally they then have a higher payoff than high - degree defectors , because the latter receive ample negative incentives from the institutional punishment pool [ fig .
[ fig6 ] ( a)-(c ) ] .
cooperators can therefore occupy high - degree nodes and from there spread across the whole population [ fig .
[ fig6 ] ( g ) ] , and this the more effectively the higher the value of @xmath2 .
from here onwards we turn to considering degree - normalized payoffs , which can have important negative consequences for the evolution of public cooperation in heterogeneous environments if compared to absolute payoffs @xcite . as shown in fig .
[ fig7 ] , the fraction of cooperators unsurprisingly increases with increasing @xmath1 for various distribution strength @xmath2 ( top panels ) .
furthermore , at small values of @xmath0 , for small values of @xmath1 defectors always dominate , regardless of the value of @xmath2 , while for intermediate values of @xmath1 cooperators recover gradually as @xmath2 increases . for high @xmath1 values
there exists an intermediate close - to - zero value of @xmath2 that maximizes the stationary fraction of cooperators [ fig .
[ fig7](a ) and ( d ) ] . when the enhancement factor is larger , the extent of the parameter region where the nonmonotonous phenomenon can be observed decreased . instead , for high values of @xmath1 , the cooperation level increases with increasing values of @xmath2 and the area of complete cooperator dominance increases as well [ fig . [ fig7](e ) and
( f ) ] . if comparing the results presented in fig .
[ fig7 ] with those presented in fig .
[ fig1 ] , we find that the nonmonotonous phenomenon still exists , and it can appear even at larger values of @xmath0 because of the consideration of degree - normalized payoffs . in general
, however , the explanation of these results and the evolutionary mechanisms behind are the same as those described when considering institutional rewarding with absolute payoffs .
lastly , we consider institutional punishment with degree - normalized payoffs . from the results presented in fig . [ fig8 ]
it follows that the stationary fraction of cooperators increases with increasing values of @xmath1 , and this regardless of the value of @xmath2 ( top panels ) .
when the enhancement factor is small , we can see that the nonmonotonous dependence of the fraction of cooperators on @xmath2 exists at intermediate values of the average incentive @xmath1 [ fig .
[ fig8](d ) ] .
when the enhancement factor increases , this phenomenon still exists , but the extent of the parameter region where the nonmonotonous dependence can be observed decreases [ fig . [ fig8](e ) ] .
surprisingly , when the enhancement factor increases further , the nonmonotonous dependence disappears .
instead , in a narrow region of intermediate @xmath1 values , the fraction of cooperators decreases with increasing values of @xmath2 . at the same time , the extent of the full cooperation area increases while the full defection region decreases in the considered @xmath15 parameter space [ fig .
[ fig8](f ) ] .
while the underlying mechanism for the nonmonotonous dependence on @xmath2 is qualitatively identical to that reported before when considering institutional punishment with absolute payoffs , the decrease of the level of cooperation at intermediate values of @xmath1 and @xmath16 as @xmath2 increases requires special attention .
we note that when the value of @xmath0 is large , low - degree cooperators can still have a positive payoff . for negative @xmath2 , these low - degree cooperators can even have the highest payoffs because of the consideration of degree - normalized payoffs and sufficiently high values of @xmath1 to weigh heavily on the defectors .
therefore , cooperators dominate on all low - degree nodes and from there spread further across the whole network and rise to dominance .
this atypical spreading is a unique consequence of the consideration of the optimal distribution of negative incentives from the punishment pool , and it highlights the importance of the parameter @xmath2 . for positive values of @xmath2
namely , because most of the negative incentives are then assigned to high - degree defectors and there are only a few of those in the entire population @xcite , the majority of low - degree defectors is not punished at all .
the previously described spreading of cooperators from the low - degree nodes outwards is therefore impaired , which ultimately results in an overall lower stationary fraction of cooperators . instead of cooperation , for larger values of @xmath2
the low - degree nodes `` emit '' defection throughout the population .
to summarize , we have studied how to best distribute limited institutional incentives in order to maximize public cooperation on scale - free networks .
we have considered both institutional rewarding of cooperators and institutional punishment of defectors , and we have also distinguished between absolute and degree - normalized payoffs . our key assumptions was that , since in heterogeneous environments players have a different number of partners , the incentives ought to be distributed by taking this into account
. this would be in agreement with the established importance of degree heterogeneity for cooperation in evolutionary games @xcite .
traditionally , however , previous research has considered the limited budged be distributed equally among all the potential recipients of the incentives , irrespective of the players status and influence within the network .
accordingly , how to distribute the incentives to optimize public cooperation was an important open problem . we have found interesting solutions on how to optimally distribute the incentives based on each player s social influence level , the proxy for which are the number of social ties the players have within the interaction network . we have shown that sharing the incentives equally among all regardless of status is optimal only if the social dilemma is strong and the propensity to contribute to the common pool is thus weak , and if in addition the available amount of incentives is intermediate . this result is valid for both institutional punishment and institutional rewarding , and it does not depend on whether absolute or degree - normalized payoffs count towards evolutionary fitness . however , if the environment already favors cooperative behavior when the public goods game is characterized with a high enhancement factor then it is best to reward influential players more than low - degree players , and this regardless of whether absolute or degree - normalized payoffs apply . for institutional punishment , on the other hand ,
the solution of the optimization problem depends on whether absolute or degree - normalized payoffs are used .
we have shown that degree - normalized payoffs require high - degree nodes be punished more lenient than low - degree nodes , while if absolute payoffs count , then high - degree nodes should be punished stronger than low - degree nodes .
in general , rewarding influential cooperators strongly and punishing auxiliary defectors leniently appears to be optimal for the successful evolution of public cooperation . in terms of solving actual common goods problems
, our work might have merit in situations with strong diversity in roles and group sizes .
one representative example of such a situation is climate change governance , where existing research has shown that local institutions are an effective way to promote the emergence of widespread cooperation @xcite .
since our results are derived not only from local institutions , but take into account also the heterogeneous interaction environment , they could offer further advice on how to arrive at globally acceptable climate policies @xcite . while the evolution of institutions remains a puzzle @xcite , their importance for enforcing socially acceptable behavior in human societies
can hardly be overstated .
although institutionalized punishment appears to be prevailing , recent research concerning the effectiveness of punishment , for example related to antisocial punishment @xcite , reciprocity @xcite , and reward @xcite , is questioning the aptness of sanctioning for elevating collaborative efforts and raising social welfare . indeed , although the majority of previous studies addressing the `` stick versus carrot '' dilemma concluded that punishment is more effective than reward in sustaining public cooperation @xcite , evidence suggesting that rewards may be as effective as punishment and lead to higher total earnings without potential damage to reputation or fear from retaliation is mounting @xcite .
in particular , rand and nowak @xcite argue convincingly that healthy levels of cooperation are likelier to be achieved through less destructive means .
we hope that our study will prove to be inspirational for further research aimed at discerning the importance of positive and negative reciprocity for human cooperation , as well as for looking closely at their correlated effects @xcite .
we consider the evolutionary public goods game on the barabsi - albert scale - free network @xcite .
each player @xmath17 occupies one node of the network , and it can choose between cooperation @xmath18 ) and defection ( @xmath19 ) as the two competing strategies .
to each public goods game cooperators contribute the cost @xmath20 , while defectors contribute nothing .
the payoff of player @xmath17 who is member in the group @xmath21 , which is centered on player @xmath22 , depends on the size of the group @xmath23 ( here @xmath24 is also the degree of node @xmath22 ) , on the number of cooperators @xmath25 in the group , and on the enhancement factor @xmath0 .
in addition to the payoffs stemming from the public goods game , each group receives institutional incentives @xmath26 to be used either for rewarding cooperators or for punishing defectors , where @xmath1 is the average amount of available incentives .
when the incentives are used for rewarding , a cooperator @xmath22 with degree @xmath24 that is member in the group @xmath27 thus receives the payoff @xmath28 while a defector in the same group receives @xmath29 where @xmath2 is the distribution strength . according to the definition of the payoffs , for @xmath11 high - degree nodes
obtain larger rewards than low - degree nodes , while for @xmath14 low - degree nodes receive a larger share from the incentive pool .
if the incentives are used for punishing defectors rather than rewarding cooperators , then a cooperator @xmath22 with degree @xmath24 that is member in the group @xmath27 receives @xmath30 while a defector in the same group receives @xmath31 as by institutionalized rewarding , here to @xmath11 implies high - degree nodes are punished stronger than low - degree nodes , and vice versa for @xmath14 .
each player @xmath17 participates in @xmath32 public goods games , which are staged in groups that are centered on player @xmath17 itself and on its @xmath33 neighbors , respectively .
the total payoff player @xmath17 obtains is thus @xmath34 . after playing the games ,
a player is allowed to learn from one of its randomly chosen neighbors @xmath22 and update its strategy accordingly .
the probability of strategy change is given by the fermi function @xcite @xmath35}},\ ] ] if we assume that absolute payoff are considered .
however , previous research has emphasized also the importance of degree - normalized payoffs @xcite , in which case the probability of strategy change is @xmath36}}.\ ] ] we consider both absolute ( eq . [ eq.1 ] ) and degree - normalized ( eq . [ eq.2 ] ) payoffs to be representative for the evolutionary fitness of individual players . especially for institutional punishment ,
the solution of the considered optimization problem depends significantly on this difference . without losing generality we set the uncertainly in the strategy adoption process to @xmath37 @xcite , so that it is very likely that the better performing players will be imitated , although it is also possible that players will occasionally learn from those performing worse .
this research was supported by the national natural science foundation of china ( grant 11161011 ) and the slovenian research agency ( grants j1 - 4055 and p5 - 0027 ) .
x. c. would like to thank ulf dieckmann for helpful discussion . for different values of the distribution strength @xmath2 .
the enhancement factor is @xmath39 in ( a ) and @xmath13 in ( b ) .
bottom row depicts the contour plot of the fraction of cooperators as a function of @xmath2 and @xmath1 , as obtained for the enhancement factor @xmath39 in ( c ) and @xmath13 in ( d).,width=377 ] for three typical values of @xmath2 .
institutional rewarding and absolute payoffs apply .
the insets of ( a ) and ( b ) show the mean payoff difference between cooperators and defectors for low - degree and middle - degree nodes during the early stages of evolution . during the evolutionary process ,
if the enhancement factor is small , cooperators always have a higher mean payoff than defectors at an intermediate value of @xmath2 .
parameter values are @xmath39 and @xmath40.,width=604 ] for different values of the distribution strength @xmath2 .
the enhancement factor is @xmath39 in ( a ) and @xmath13 in ( b ) .
bottom row depicts the contour plot of the fraction of cooperators as a function of @xmath2 and @xmath1 , as obtained for the enhancement factor @xmath39 in ( c ) and @xmath13 in ( d).,width=377 ] for three typical values of @xmath2 . institutional punishment and absolute payoffs apply .
the insets of ( a ) and ( b ) show the mean payoff difference between cooperators and defectors for low - degree and middle - degree nodes during the early stages of evolution . during the evolutionary process ,
if the enhancement factor is small , cooperators always have a higher mean payoff than defectors at an intermediate value of @xmath1 .
parameter values are @xmath39 and @xmath40.,width=604 ] for different values of the distribution strength @xmath2 .
the enhancement factor is @xmath39 in ( a ) , @xmath13 in ( b ) , and @xmath16 in ( c ) .
bottom row depicts the contour plot of the fraction of cooperators as a function of @xmath2 and @xmath1 , as obtained for the enhancement factor @xmath39 in ( d ) , @xmath13 in ( e ) , and @xmath16 in ( f).,width=453 ] for different values of the distribution strength @xmath2 .
the enhancement factor is @xmath39 in ( a ) , @xmath13 in ( b ) , and @xmath16 in ( c ) .
bottom row depicts the contour plot of the fraction of cooperators as a function of @xmath2 and @xmath1 , as obtained for the enhancement factor @xmath39 in ( d ) , @xmath13 in ( e ) , and @xmath16 in ( f).,width=453 ] | in the framework of evolutionary games with institutional reciprocity , limited incentives are at disposal for rewarding cooperators and punishing defectors . in the simplest case
, it can be assumed that , depending on their strategies , all players receive equal incentives from the common pool .
the question arises , however , what is the optimal distribution of institutional incentives ?
how should we best reward and punish individuals for cooperation to thrive ?
we study this problem for the public goods game on a scale - free network .
we show that if the synergetic effects of group interactions are weak , the level of cooperation in the population can be maximized simply by adopting the simplest `` equal distribution '' scheme . if synergetic effects are strong , however , it is best to reward high - degree nodes more than low - degree nodes .
these distribution schemes for institutional rewards are independent of payoff normalization . for institutional punishment , however , the same optimization problem is more complex , and its solution depends on whether absolute or degree - normalized payoffs are used .
we find that degree - normalized payoffs require high - degree nodes be punished more lenient than low - degree nodes .
conversely , if absolute payoffs count , then high - degree nodes should be punished stronger than low - degree nodes . |
ions and electrons trapped in the earth s magnetic field may affect our technology and our daily lives in significant ways .
energetic plasma particles may penetrate satellites and disable them temporarily or permanently .
they can also pose serious health hazards for astronauts in space .
spectacles like the aurora are created by particles that enter the earth s atmosphere at polar regions ; on the other hand , aircraft personnel and frequent flyers may accumulate a significant dose of radiation due to the same particles.@xcite all of these effects are enhanced at periods of solar maximum , the next one being expected to happen between 2012 and 2014 .
occasional extreme solar events may induce currents in the ionosphere , which in turn induce significant currents on power lines , causing power outages.@xcite such events can also disrupt communications , radio and gps .
thus , understanding and predicting the processes in the earth s magnetosphere have practical importance .
this paper aims to outline one of these processes , charged - particle motion and associated adiabatic invariants , for physics students and instructors who wish to use it in lectures .
the emphasis is on numerical computation and visualization of trajectories . for a more comprehensive discussion
advanced texts on plasma physics@xcite can be consulted .
other authors have also suggested using topics from plasma research to enhance undergraduate curriculum .
lopez@xcite provides examples of how space physics can be incorporated in undergraduate electromagnetism courses , and mcguire@xcite shows how computer algebra systems can be used to follow particle trajectories in electric dipole and ( separately ) in magnetic dipole fields .
photographs of plasma experiments , such as those provided by huggins and lelek@xcite and by the ucla plasma lab web site@xcite are also helpful for understanding space plasma behavior . a schematic view of the earth s magnetosphere .
the solar wind comes from the left .
( courtesy of kenneth r. lang,@xcite reproduced with permission . ) ] figure [ fig : magnetosphere ] shows a schematic description of the earth s magnetosphere , which is the region in space where the magnetic field of the earth is dominant .
charged particles trapped in the magnetosphere form the radiation belts , the plasmasphere , and current systems such as the ring current , tail current , and field - aligned currents .
the earth radius @xmath0 ( 6378.137 km ) is a natural length scale for the magnetosphere . near the earth ,
up to 3 - 4@xmath0 , the field can be very well approximated with the field of a dipole .
however , at larger distances , the effects of the solar wind cause significant deviations from the dipole .
the solar wind is a stream of plasma carrying magnetic field from the sun .
when the solar wind encounters the earth s magnetosphere , the two systems do not mix .
this is because of the `` frozen - in flux '' condition@xcite which dictates that plasma particles stay attached to magnetic field lines , except at special locations such as polar cusps .
the solar wind influences the magnetosphere by applying mechanical and magnetic pressure on it , compressing it earthward on the side facing the sun ( the `` dayside '' ) .
this compression is stronger when the sun is more active . on the opposite side ( the `` nightside '' ) , the field
is extended over a very large distance , forming the magnetotail .
wolf@xcite provides a review of the complex and time - dependent interactions between magnetic fields , induced electric fields and plasma populations .
the van allen radiation belts form a particularly significant plasma population due to their high energy and their proximity to earth .
they can be found from 1000 km above the ground up to a distance of 6@xmath0 .
these belts are composed of electrons with energies up to several mevs and of protons with energies up to several hundred mevs .
the dynamics of these particles is the main focus of this paper .
this paper is organized as follows : section [ sec : nleqs ] introduces the relativistic equation of motion for a particle in an electric and magnetic field and describes the cyclotron , bounce and drift motions .
it also shows some typical particle trajectories under the dipolar magnetic field , approximating the earth s field .
section [ sec : firstinv ] introduces the concept of adiabatic invariants and derives the first adiabatic invariant associated with the particle motion .
section [ sec : gceqs ] gives the approximate equations of motion for the guiding center of a particle , obtained by averaging out the cyclotron motion .
section [ sec : secondthirdinv ] presents and derives the second and third invariants associated with the bounce and drift motions , respectively . section [ sec : exercises ] lists some exercises building on the concepts described in the paper .
two of these exercises describe non - dipole fields that are used for modeling different regions of the magnetosphere .
the motion of a particle with charge @xmath1 and mass @xmath2 in an electric field @xmath3 and magnetic field @xmath4 is described by the newton - lorentz equation : @xmath5 here @xmath6 is the relativistic factor and @xmath7 is the particle speed .
suppose that @xmath8 .
then , because of the cross product , the acceleration of the particle is perpendicular to the velocity at all times , so the speed of the particle ( and the factor @xmath9 ) remains constant .
further suppose that the magnetic field is uniform .
then , particles move on helices parallel to the field vector .
the circular part of this motion is called the `` cyclotron motion '' or the `` gyromotion '' .
the `` cyclotron frequency '' @xmath10 and the `` cyclotron radius '' @xmath11 are respectively given by @xmath12 where @xmath13 is the uniform field strength and @xmath14 is the component of the velocity perpendicular to the field vector .
if there are not any other forces , the parallel component of the velocity remains constant .
the combined motion traces a helix.@xcite if the electric field is not zero , we can write it as @xmath15 , where @xmath16 is parallel to @xmath4 , and @xmath17 is perpendicular to it . if @xmath18 , particles accelerate with @xmath19 along the field line and they are rapidly removed from the region . therefore , the existence of a trapped plasma population implies that the parallel electric field must be negligible .
the perpendicular component of the electric field will move particles with an overall drift velocity , known as the e - cross - b - drift , which is perpendicular to both field vectors : @xmath20 the particle will move with with the velocity @xmath21 , plus the cyclotron motion described above .
the drift velocity @xmath21 is independent of particle mass and charge .
therefore , in an inertial frame moving with @xmath21 , the e - cross - b - drift will vanish for all types of particles . for the remainder of this paper we take @xmath22 ,
that is , the acceleration due to the electric field is not taken into consideration .
this is not because electric fields are unimportant ; on the contrary , they play an important role in the complex dynamics of plasmas . the first reason for leaving out electric field effects is described above : if the field is uniform and constant , we can transform to another frame that cancels it .
even if the field is nonuniform and time - dependent , electric drifts can be vectorially added to magnetic drifts in order to obtain the overall drift .
drift velocities due to different fields are independent .
the second reason is the need for simplicity ; a static magnetic field provides sufficient real - life context for the discussion of guiding - center and adiabatic invariant concepts in general - purpose lectures .
the final reason is that this paper focuses on the region occupied by radiation belts , and in this region the magnetic term of ( [ eq : nleqs ] ) is the dominant force.@xcite now we consider motion under the influence of a magnetic dipole .
the field @xmath23 of a magnetic dipole with moment vector @xmath24 at location @xmath25 is given by:@xcite @xmath26,\ ] ] where @xmath27 , @xmath28 and @xmath29 .
for earth , we take @xmath30 , antiparallel to the @xmath31-axis , because the magnetic north pole is near the geographic south pole.@xcite at the magnetic equator ( @xmath32 , @xmath33 ) the field strength is measured to be @xmath34 .
substitution shows that @xmath35 .
then in cartesian coordinates , the field is given by : @xmath36.\ ] ] ( color online ) trajectories of two 10mev protons in the earth s dipole field .
the dipole moment is in the @xmath37 direction .
both panels show the same trajectories from different viewing angles.,title="fig : " ] ( color online ) trajectories of two 10mev protons in the earth s dipole field .
the dipole moment is in the @xmath37 direction .
both panels show the same trajectories from different viewing angles.,title="fig : " ] figure [ fig : proton_full ] shows trajectories of two protons with 10mev kinetic energy , a typical energy for radiation belts.@xcite the trajectories are calculated with the scipy module using the python language.@xcite one proton is started at @xmath38 and the other at @xmath39 .
both start with an equatorial pitch angle ( angle between the velocity and field vectors ) @xmath40 so that @xmath41 , @xmath42 and @xmath43 .
both are followed for 120 seconds .
the motion is again basically helical , but the nonuniformity of the field introduces two additional modes of motion on large spatial and temporal scales .
these are called `` the bounce motion '' and `` the drift motion '' .
the bounce motion proceeds along the field line that goes through the helix ( the `` guiding line '' ) .
the motion slows down as it moves toward locations with a stronger magnetic field , reflecting back at `` mirror points '' .
the bounce motion is much slower than the cyclotron motion .
the drift motion takes the particle across field lines ( perpendicular to the bounce motion ) . in general ,
drift motion is faster at larger distances , as observed in figure [ fig : proton_full ] .
particles in dipole - like fields are trapped on closed `` drift shells '' as long as they are not disturbed by collisions or interactions with em waves .
the drift motion is much slower than the bounce motion . under a dipolar field ,
the bounce motion period @xmath44 and the drift motion period @xmath45 are approximately given as:@xcite @xmath46 \\
\label{eq : driftperiod}\tau_{\rm d } & \approx & \frac{2\pi qb_0r_{\rm e}^3}{mv^2}\frac{1}{r_0}\left[1-\frac{1}{3}(\sin \alpha_{\rm eq})^{0.62}\right],\end{aligned}\ ] ] where @xmath47 is the equatorial distance to the guiding line and @xmath48 is the equatorial pitch angle .
both approximations have an error of about 0.5% .
if the parameters of an oscillating system are varied very slowly compared to the frequency of oscillations , the system possesses an `` adiabatic invariant '' , a quantity that remains approximately constant . in the hamiltonian formalism
, the adiabatic invariant is the same as the action variable:@xcite @xmath49 where @xmath1 , @xmath50 are canonical variables and the integral is evaluated over one cycle of motion satisfying @xmath51 .
the integral should be evaluated at `` frozen time '' , that is , the slowly varying parameter is considered constant during the integration cycle .
there are three separate periodic motions of a charged particle in a dipole - like magnetic field .
this means there are three adiabatic invariants for the particle s motion .
the canonical momentum for a charged particle in a magnetic field with vector potential @xmath52 is @xmath53 . to obtain the first adiabatic invariant @xmath54
, we integrate the canonical momentum over one cycle of the cyclotron orbit : @xmath55 here @xmath56 is the line element of the particle trajectory . even though the path does not exactly close , we evaluate the integral as if it does .
stokes theorem states that : @xmath57 where the right - hand side integral is taken over the surface bounded by the closed path of the left - hand side integral .
using stokes theorem with @xmath58 , the integral @xmath54 takes the form : @xmath59 the second equation follows from substituting @xmath11 from ( [ eq : cycrad ] ) .
it is assumed that the cycle is sufficiently small so that @xmath4 is considered uniform over the area bounded by the gyromotion .
this assumption is essential for the existence of adiabatic invariants .
the area element @xmath60 is antiparallel to @xmath4 because of the sense of gyration of the particles ; hence the negative sign of the second term .
customarily , one takes @xmath61 as the first invariant , which differs from @xmath54 only in some constant factors .
the parameter @xmath62 is called the `` magnetic moment '' because it is equal to the magnetic moment of the current generated by the particle moving on this circular path .
the adiabatic invariance of @xmath62 explains the existence of mirror points .
as the particle moves along the field line toward points with larger @xmath63 , the perpendicular speed @xmath14 must increase in order to keep @xmath62 constant .
however , @xmath14 can be at most @xmath7 , the constant total speed
. the particle stops at the point where @xmath64 and falls back ( see section [ sec : gceqs ] ) .
the left panel of figure [ fig : firstinv ] shows the magnetic moment @xmath62 for the two protons shown in figure [ fig : proton_full ] .
the values are oscillating with the local cyclotron frequency because instantaneous values of @xmath14 and @xmath63 are used .
the actual adiabatic invariant is the average of these oscillations and it is constant in time .
the top right panel shows the oscillations of @xmath62 vs. time for a shorter interval .
comparison with the @xmath31 vs. time plot below , it can be seen that these oscillations are correlated with the bounce motion .
oscillations of @xmath62 have a small amplitude near the mirror point because there the parallel motion slows down and the overall motion becomes more adiabatic . near
the equatorial plane ( @xmath65 ) parallel motion is fastest , the motion is less adiabatic , and @xmath62 oscillates with a larger amplitude .
the proper way of calculating the first invariant would remove all oscillations : after following the full trajectory , find the times @xmath66 where @xmath67 ( or @xmath68 ) by searching along discrete path points and by interpolating .
the difference between any successive time points is half a gyroperiod .
then take the average of @xmath62 over that time interval using the path points . repeating this procedure for all time intervals
, we get a constant set of @xmath62 values ( apart from numerical errors ) .
the `` guiding center '' is the geometric center of the cyclotron motion .
if the magnetic field is uniform the guiding center moves with constant velocity parallel to the field line . in nonuniform field geometries , there is a sideways drift in addition to the motion along the field lines , as seen in the dipole example in figure [ fig : proton_full ] .
calculation of the guiding center motion requires that the motion is helical in the smallest scale , and that the field does not change significantly within a cyclotron radius .
this condition can be expressed as @xmath69 the magnetic moment is an adiabatic invariant under this condition .
northrop@xcite and walt@xcite give detailed derivations of the equations of guiding - center motion .
in order to derive the acceleration of the guiding center , the particle position @xmath25 is substituted with : @xmath70 where @xmath71 is the position of the guiding center .
the vector @xmath72 lies on the plane perpendicular to the field , oscillates with the cyclotron frequency , and its length is equal to the cyclotron radius . assuming that the cyclotron radius is much smaller than the length scale of the field
, we can expand @xmath73 around @xmath71 to first order in a taylor series : @xmath74 this expansion is substituted into the newton - lorenz equation ( [ eq : nleqs ] ) and the equation is averaged over a cycle , eliminating rapidly oscillating terms containing @xmath72 and its derivatives .
the resulting acceleration of the guiding center is given by : @xmath75 taking the dot product of both sides of the equation with @xmath76 , the local magnetic field direction , will yield the equation of motion along the field line .
the first term becomes identically zero because it is perpendicular to the field vector .
then : @xmath77 where @xmath78 is the speed along the field line . replacing @xmath79 and defining @xmath80 as the distance along the field line
, this equation can be written as : @xmath81 here @xmath82 is the field strength along the field line .
the factors @xmath62 and @xmath9 can be taken inside the derivative because they are constants .
this expression shows that the quantity @xmath83 acts like a potential energy in the parallel direction .
the negative sign indicates that the parallel motion is accelerated toward regions with smaller field strength .
the motion of the guiding center perpendicular to field lines can be determined by taking the cross product of eq .
( [ eq : gcaccel ] ) with the field direction vector .
the resulting equation is then iterated to obtain an approximate solution for the drift velocity across field lines .
@xmath84 this drift velocity is actually the sum of two separate drift velocities : the gradient drift that arises from the nonuniformity of the magnetic field , and the curvature drift that occurs because the field lines are curved . for an example of pure gradient drift motion ,
see exercise [ ex : graddrift ] in section [ sec : exercises ] .
equation ( [ eq : driftvelocity ] ) shows that electrons and ions drift in opposite directions .
this creates a net current around the earth , called `` the ring current '' .
gradient and curvature drifts are the only drifts seen in static magnetic fields .
external electric fields , external forces such as gravity , and time dependent fields create additional drift velocities.@xcite combining these , we obtain the following equations of motion for the guiding center : @xmath85 these equations are more complicated than the simple newton - lorentz equation , and they require computing @xmath86 and @xmath87 at each integration step .
still , they have the advantage that we can follow the overall motion with relatively large time steps because we do not need to resolve the cyclotron motion .
this reduces the cumulative error , as well as the total computation time .
( color online ) guiding - center trajectories for the particles shown in figure [ fig : proton_full ] . the cyclotron motion is averaged out.,title="fig : " ] ( color online ) guiding - center trajectories for the particles shown in figure [ fig : proton_full ] .
the cyclotron motion is averaged out.,title="fig : " ] figure [ fig : proton_gc ] shows the solution of the guiding - center equations for the same protons shown in figure [ fig : proton_full ] under a dipolar magnetic field ( python source code provided in supplement).@xcite it should be noted that the guiding - center equations are approximate because only terms first order in cyclotron radius are used in their derivation . for particles with larger cyclotron radii ( higher kinetic energies ) , there may be a noticeable difference between guiding - center and full - particle trajectories ( see exercise [ ex : gcandfull ] in section [ sec : exercises ] ) .
the second adiabatic invariant is associated with the bounce motion , and it is calculated by integrating the canonical momentum over a path along the guiding field line : @xmath88 where @xmath89 is the line element along the field line .
the adiabatic integrals are evaluated in a `` frozen '' system : it is assumed that the drift is stopped , so the motion moves back and forth along a single guiding field line .
using stokes theorem , the second term can be converted to an integral over a surface bounded by the bounce path @xmath90 which is zero because the bounce motion goes along the same path in both parts of the cycle so that the enclosed area vanishes .
then , the second adiabatic invariant can be written as @xmath91 where @xmath80 is the path length along the field line , and @xmath92 , @xmath93 are locations of the mirror points where the particle comes back . at the mirror point
the parallel speed @xmath78 vanishes so that @xmath94 . from the invariance of the magnetic moment @xmath62 it follows that @xmath95 where @xmath96 is the field strength at the mirror point .
substituting @xmath97 and solving for @xmath78 gives @xmath98 the integral @xmath99 is an adiabatic invariant in general ( even if there are electric fields or slow time - dependent fields ) .
if the speed is constant , @xmath100 can be used as an adiabatic invariant .
the integral @xmath100 depends only on the magnetic field , not on the particle velocity , so it can be used to compute the drift path using the field geometry only ( see exercises [ ex : evaluatei]-[ex : driftpath ] in section [ sec : exercises ] ) .
the second invariant values in time , calculated using the guiding - center trajectories in fig .
[ fig : proton_gc ] .
lower curve is for the proton starting at 2 @xmath0 distance , upper curve for the proton starting at 4 @xmath0 . ]
figure [ fig : secinv ] shows that the value of the second invariant @xmath100 , evaluated using the guiding - center trajectories shown in figure [ fig : proton_gc ] , stays constant in time .
the integral is evaluated not using the definition of @xmath100 in eq .
( [ eq : secinv ] ) , but using the dynamical form @xmath101 where the integral is evaluated over a half period .
the limits of the integrals are determined by interpolation between two points where the parallel speed changes sign .
the values do not oscillate because the adiabatic invariant is calculated as an average over a cycle .
the drift path is found by averaging the bounce motion . in a dipolar field
all drift paths are circular due to the symmetry of the field .
the third invariant , associated with the drift motion , is defined as an integral along the drift path : @xmath102 where @xmath56 is a line element on the drift path .
this can be written as @xmath103 in the first term @xmath104 , the drift speed , is the magnitude of the expression given in eq .
( [ eq : driftvelocity ] ) .
the second term is obtained by using stokes theorem as above .
an order of - magnitude comparison shows that the first term of @xmath105 can be neglected because it is much smaller than the second term : from eq .
( [ eq : driftvelocity ] ) the order of magnitude of the drift speed can be written as @xmath106 where @xmath63 is the typical field strength at the drift path and @xmath107 is the typical distance from the origin .
similarly , from eq .
( [ eq : cycrad ] ) , the cyclotron radius has the order of magnitude @xmath108 then , the order - of - magnitude ratio of the terms in eq .
( [ eq : thirdinv2 ] ) is @xmath109 according to the adiabaticity condition eq .
( [ eq : adbcondition ] ) , @xmath110 must be very small .
therefore the first term of eq .
( [ eq : thirdinv2 ] ) is ignored and we have @xmath111 where @xmath112 is the magnetic flux through the drift path .
the third adiabatic invariant is useful as a conservation law when the magnetosphere changes slowly , i.e. , over longer time scales compared to the drift period .
the use of three invariants gives more accurate results for the motion of particles over long periods .
numerical solution of equations of motion are less accurate because of accumulated numerical errors .
roederer@xcite discusses in detail how drift shells can be constructed geometrically using the invariants ( see exercise [ ex : driftpath ] in sec .
[ sec : exercises ] ) .
furthermore , as three invariants uniquely specify a drift shell , the invariants themselves can be used as dynamical variables when investigating the diffusion of trapped particles.@xcite
this section lists some further programming exercises with varying difficulty . the code given in the supplement@xcite
can be modified to solve some of the exercises . 1
. * uniform magnetic field . * follow charged particles under a uniform magnetic field @xmath113 where @xmath114 .
verify that the particles follow helices with cyclotron radius and frequency as given in eqs .
( [ eq : cycfreq ] , [ eq : cycrad ] ) .
experiment with particles with different mass and charge values .
2 . [ ex : graddrift ] * gradient drift . * consider a magnetic field given as @xmath115 .
the field has a gradient in the @xmath116-direction , but no curvature .
set @xmath117 and @xmath118 .
follow the trajectory of a particle with mass @xmath119 and @xmath120 initialized with velocity @xmath121 at the origin .
note that the sideways drift arises from the fact that the cyclotron radius is smaller at stronger fields .
* equatorial particles . *
consider a particle in a dipolar magnetic field , located at the equatorial plane ( @xmath65 ) with zero parallel speed . as the field strength is minimum at the equator with respect to the field line
, there is no parallel acceleration and the particle stays on the equatorial plane at all times . using the dipole model , follow an equatorial particle and verify that the center of the motion stays on a contour of constant @xmath63 , as implied by the conservation of the first adiabatic invariant .
4 . * explore the drift motion .
* run the programs in the supplement to trace protons and electrons using the dipole model .
initialize particles with different energies , starting positions and pitch angles .
verify that electrons and protons drift in opposite directions , and electrons have a much smaller cyclotron radius than protons with the same kinetic energy .
estimate the periods of bounce and drift motions and compare them with eqs .
( [ eq : bounceperiod ] , [ eq : driftperiod ] ) . 5 .
[ ex : gcandfull ] * accuracy of the guiding - center approximation . * simulate the full particle and guiding center trajectories with the same initial conditions and plot them together .
shift the initial position of the particle properly so that the guiding center runs through the middle of the helix .
+ repeat with protons with 1kev , 10kev , 100kev and 1mev kinetic energies . at higher energies
, the guiding - center trajectory lags behind the full particle because the omitted high - order terms become more significant as the cyclotron radius increases .
* different numerical methods . *
solve the full particle and guiding center equations using different numerical schemes,@xcite such as verlet , euler - cromer , runge - kutta and bulirsch - stoer .
verify the accuracy of the solution by checking the conservation of kinetic energy and adiabatic invariants .
. * field line tracing . * plot the magnetic dipole field line starting at position @xmath122 . for any vector field @xmath123
, a field line can be traced by solving the differential equation @xmath124 where @xmath80 is the arclength along the field line .
[ ex : evaluatei ] * compute @xmath100 . *
compute the second invariant @xmath100 ( eq . [ eq : secinv ] ) under a dipolar field for a guiding center starting at position @xmath125 and an equatorial pitch angle @xmath48 .
the integral should be taken along a field line , which can be traced as described above . from the first adiabatic invariant one finds @xmath126 and the limits of the integral
are found by solving @xmath127 .
[ ex : ialongdrift ] * second invariant along the drift path . *
produce a guiding - center trajectory under the dipole field . by interpolation ,
determine the points @xmath128 where the trajectory crosses the @xmath65 plane .
compute the second invariant @xmath100 at these equatorial points and plot .
verify that the values are constant as shown in fig .
[ fig : secinv ] . 10 .
[ ex : driftpath ] * drift path tracing using the second invariant . *
pick a starting location @xmath129 and mirror field @xmath96 , and evaluate the second invariant @xmath130 as described above . compute the gradient @xmath131 numerically using central differences : @xmath132\\ \partial_y i & \approx & \frac{1}{2\delta}\left[i(x_0,y_0+\delta , b _ { \rm m } ) - i(x_0,y_0-\delta , b _ { \rm m})\right],\end{aligned}\ ] ] where @xmath133 is a small number ( e.g. 0.01@xmath0 ) .
+ the second invariant is constant along the drift path , so for a finite step @xmath134 , it holds that @xmath135 .
use this relation to trace successive steps along the drift shell .
this method is more accurate than following a particle or a guiding center
* * the double - dipole model.**@xcite the dipole ceases to be a good approximation for the magnetic field of the earth as we go farther in space .
the double - dipole model , although unrealistic , introduces a day - night asymmetry that vaguely mimics the deformation of the magnetosphere by the solar wind .
it can be used to capture some basic features of particle dynamics in the outer magnetosphere , if only qualitatively .
+ the model has one dipole ( earth ) at the origin , pointing in the negative @xmath31-direction , and an image dipole at @xmath136 . if both dipoles are identical , the magnetic field is given by : @xmath137 where @xmath138 is given by eq .
( [ eq : dipolefield ] ) .
+ the domains of each dipole are separated by the plane @xmath139 .
this plane simulates the magnetopause , the boundary between the magnetosphere and the solar wind . for slightly better realism
, the image dipole can be multiplied by a factor larger than 1 so that the magnetopause becomes curved .
also , the two dipoles can be tilted by equal and opposite angles with respect to the sun - earth line ( @xmath116-axis ) , to simulate the fact that the dipole moment of the earth is tilted .
1 . starting at various latitudes , plot the magnetic field lines of eq .
( [ eq : ddfield ] ) on the @xmath140 plane .
observe the compression of field lines on the dayside and extension on the nightside .
note that no field line crosses the @xmath139 plane .
multiply the image dipole term by 1.5 and repeat .
2 . follow several guiding - center trajectories starting position @xmath116 between @xmath141 and @xmath142 , @xmath33 , and pitch angles between @xmath143 and @xmath144 ( smaller pitch angle creates a longer bounce motion ) . with small pitch angles ,
the particle should come closer to earth on the day side .
repeat with a pitch angle of @xmath145 .
now the particle goes away from the earth on the dayside .
explain these observations using the conservation of first and second adiabatic invariants .
3 . the double - dipole field can break the second invariant for some trajectories .
the reason of this breaking is that the field strength has a local maximum on the dayside around the equatorial plane .
particles with sufficiently small mirror fields are diverted to one side of the equatorial plane because they can not overcome this field maximum , as seen in fig .
[ fig : protondsb ] . + a proton with 200kev kinetic energy , initialized on the right edge at @xmath146 with @xmath144 pitch angle in a double - dipole field . ]
+ start an electron guiding center at position @xmath147 , @xmath148 with kinetic energy @xmath149 with an equatorial pitch angle @xmath150 and follow its guiding center for 1000 seconds with time step 0.01 . on the dayside
the trajectory will temporarily move above or below the equatorial plane . using the method used in fig .
[ fig : secinv ] , plot the second invariant @xmath100 versus time . the second invariant will be constant between breaking points , but its value will differ from the initial value .
the reason is that near the breaking points bounce motion slows down and the adiabaticity condition does not hold .
however , the first invariant is not broken .
+ this phenomenon , named drift - shell bifurcation,@xcite can be one of the causes of particle diffusion in the magnetosphere
. * magnetotail current sheet . * on the tail region of the magnetosphere , magnetic field lines are heavily stretched , and a sheet of current is flowing through them.@xcite the field in the magnetotail can be represented by the simple form : @xmath151 in this problem we set @xmath152 , @xmath153 and @xmath154 .
the field lines trace parabolas on the @xmath140 plane , which can be seen by integrating the equation @xmath155 .
the parameter @xmath156 is the scale of the current sheet thickness .
the truncation of the field at @xmath157 simulates the finite size of the tail region .
+ the field vector points to opposite directions on both sides of the equatorial plane @xmath65 .
when a charged particle is released from above , it moves toward the weaker region near @xmath158 where the adiabaticity condition does not hold .
the helix becomes a `` serpentine orbit '' that moves in and out of the equatorial plane .
the chaotic dynamics of these orbits is extensively studied.@xcite + ( color online ) types of orbits created by a particle with mass @xmath159 and charge @xmath160 near a current sheet .
( a ) speiser orbits of transient particles , ( b ) cucumber orbits of quasitrapped particles and ( c ) ring orbits of trapped particles .
note the different scales of axes.,title="fig : " ] ( color online ) types of orbits created by a particle with mass @xmath159 and charge @xmath160 near a current sheet .
( a ) speiser orbits of transient particles , ( b ) cucumber orbits of quasitrapped particles and ( c ) ring orbits of trapped particles .
note the different scales of axes.,title="fig : " ] + figure [ fig : speiser ] shows the three types of orbits that can exist in such a model.@xcite `` speiser orbits '' approach the equatorial plane and later go beyond @xmath157 and leave the tail region .
`` cucumber orbits '' alternate between helical and serpentine orbits . these do not form closed orbits because of the breaking of the first invariant at the equatorial plane .
`` ring orbits '' alternate between oppositely - directed fields ; they do not have a helical section
evaluate @xmath161 for @xmath162 and determine the direction of gradient - curvature drift .
2 . by trial and error , find initial conditions that create the types of orbits shown in fig .
[ fig : speiser ] .
space plasmas provide many case studies which , after proper simplification , can be used in the undergraduate physics curriculum .
we have presented one such case , the basic theory of charged - particle motion under the dipole .
this paper focuses on visualization and concrete computation , with the hope that students will modify or rewrite the code to run their own numerical experiments on particle motion in magnetic fields . in my opinion ,
numerical simulations provide at least two important pedagogical benefits : first , even if the required analytical tools are beyond the students level , they can use simulations to obtain a qualitative understanding .
second , the process of coding the simulation forces students to understand the problem at a basic and operational level . the main body of this article or the exercises can be incorporated in lectures , or they can be given as advanced assignments to interested students .
a natural place for this subject is a course on electromagnetism and/or plasma physics .
when the basics are introduced , the instructor can discuss related subjects such as plasma confinement , radiation belts , or space weather . in advanced mechanics courses , adiabatic invariants
are usually presented with an abstract formalism .
charged particle motion provides a natural and concrete case where adiabatic invariants are relevant and indispensable .
the subject can also be incorporated in courses on computational physics .
accuracy and stability of different numerical integration schemes may be presented using charged particle motion .
the widely separated time scales of the motion would be a challenge for most of the schemes .
electric potential differences across the magnetosphere are of the order of 100kv which , over a distance of about @xmath165 , yield an electric field strength of the order of @xmath166 .
the magnetic field of the earth has strength @xmath167 on the magnetic equatorial plane , where @xmath168 is the distance measured in @xmath0 .
a typical radiation - belt proton with 10mev energy has speed @xmath169 .
then , in si units , @xmath170
. therefore the electric drift can be neglected near the earth where @xmath171 .
alternatively one can say that low - energy particles that could be affected by electric drifts have already drifted away , leaving behind the trapped high - energy particles .
we are using the geocentric solar ecliptic ( gse ) coordinate system .
the earth is at the origin , the @xmath116-axis points to the sun , the @xmath31-axis is set so that the dipole moment vector is on the @xmath172-plane , and the @xmath173-axis is perpendicular to both axes .
see supplementary material at @xmath163https://sites.google.com / site / mkaanozturk / programs@xmath164 for the python source code that solves the newton - lorentz and the guiding - center equations and displays them .
w. j. hughes , `` the magnetopause , magnetotail and magnetic reconnection '' in _ introduction to space physics _ , edited by margaret g. kivelson and cristopher t. russell ( cambridge university press , 1995 ) , pp .
227 - 287 .
jrg bchner , lev m. zelenyi , `` regular and chaotic charged particle motion in magnetotaillike field reversals 1 .
basic theory of trapped motion '' , journal of geophysical research , * 94 * , a9 , 11821 - 11842 ( 1989 ) . | i outline the theory of relativistic charged - particle motion in the magnetosphere in a way suitable for undergraduate courses .
i discuss particle and guiding center motion , derive the three adiabatic invariants associated with them , and present particle trajectories in a dipolar field .
i provide twelve computational exercises that can be used as classroom assignments or for self - study .
two of the exercises , drift - shell bifurcation and speiser orbits , are adapted from active magnetospheric research .
the python code provided in the supplement can be used to replicate the trajectories and can be easily extended for different field geometries . |
in bosonic systems where two - body scattering is resonant , efimov predicted in the early 70 s the existence of shallow trimers with a spectrum characterized by a discrete scaling symmetry and an accumulation point at zero energy @xcite .
the first clear evidence of such states occurred in ultra - cold atomic systems @xcite , where fine control of the resonant behavior through a feshbach resonance ( fr ) mechanism is possible @xcite .
the few - body problem which was originally deeply linked to nuclear physics is now a hot topic in ultra - cold physics @xcite and plays a significant role also in the search for highly correlated many - body states @xcite . while experiments unambiguously confirm general properties of efimov physics @xcite , deviations from universal predictions
are observed supporting the interest in using detailed approaches like the one in ref .
@xcite or including a microscopic description of the fr mechanism with a coupling between atoms ( in the open channel ) and molecules ( in the closed channel ) as in refs .
@xcite . in the latter references , the finite width of the magnetic fr and the background scattering length ( denoted @xmath0 and @xmath1 , respectively ) appear as key parameters for the low energy scattering amplitude which plays a central role in three - body properties @xcite . at the two - body level ,
two - channel models permit the expression of the scattering length ( denoted @xmath2 ) as a function of the external magnetic field @xmath3 in the vicinity of a given resonance located at @xmath4 with : @xmath5 the magnetic width can also be characterized by a length @xmath6 hereafter referred as the width radius and defined by @xmath7 , where @xmath8 is the difference in magnetic moment for an atomic pair in the open- and in the closed - channel @xcite . in the limit of an asymptotically narrow resonance ,
the width radius takes a large value as compared to the range of the interparticles forces ( denoted by @xmath9 ) and defines a low energy scale . in this peculiar regime
it is possible to determine efimov physics and deviations from universality ( in the intermediate detuning regime ) both in terms of @xmath2 and @xmath10 only @xcite .
recently , it has been shown that the superposition of an external electric field permits to tune the background scattering length @xmath1 and thus the width radius @xmath6 @xcite which makes possible study of low energy three - body properties as a function of these quantities .
following this idea , a zero - range model generalizing the universal theory of broad resonances and the effective range approach of ultra - narrow resonances is solved in what follows .
this modeling is especially pertinent in situations where each of the three parameters @xmath11 of the fr are large ( in absolute value ) with respect to the range @xmath12 which is of the order of the van der waals radius given by @xmath13 ( where @xmath14 is the van der waals coefficient of the interatomic potential ) . in this model ,
the _ so - called _ three - body parameter ( denoted @xmath15 ) which characterizes the high momentum of the three - body wave function , is introduced by means of a nodal condition .
while keeping the simplicity of the zero - range approach , this filtering method of physical solutions permits avoiding the thomas collapse @xcite . in the limit of broad resonances , the nodal condition implemented in the original skorniakov ter - martirosian ( stm ) equation ( eq.(12 ) in ref .
@xcite ) through a simple subtraction scheme permits recovery of the results of the universal theory. in the limit of asymptotically narrow resonances @xmath16 , a decoupling occurs between shallow states of the effective range theory @xcite and deeper states depending on @xmath15 . between the ultra - narrow and the infinitely broad resonance limits , the efimov spectrum exhibits a crossover involving these two families of states . in the regime of large positive background scattering length and width radius ,
the model supports a shallow dimer state .
in this situation , efimov trimers of lowest energy are quasi - bound states and manifest themselves as asymmetric resonance peaks in atom - dimer scattering which are characteristic of fano resonances @xcite .
this result confirms the scenario introduced in nuclear physics for the compound system made of two neutrons interacting with one @xmath17c nucleus @xcite .
the zero - range approach used hereafter is based on the expression of the low energy two - body @xmath18-wave scattering phase shift ( denoted @xmath19 ) or equivalently of the scattering amplitude ( denoted by @xmath20 ) , obtained from the limit of zero potential range ( @xmath21 ) of realistic models @xcite or directly from a zero - range diagrammatic approach ( see eq .
( 2 ) in ref .
@xcite ) . in this limit ,
the scattering phase shift ( related to the scattering amplitude by @xmath22 ) at collisional momentum @xmath23 is : @xmath24 equation can also be obtained in the low momentum limit ( @xmath25 ) of the models in refs .
@xcite , and is thus the basic object which describes the interparticle interaction for low energy process @xcite . for a vanishing background scattering length , eq .
coincides with the effective range approximation used in refs .
systems studied hereafter are translation invariant , hence the zero - range approach is expressed directly in the momentum representation in the same manner as in refs . @xcite . in the center of mass frame
, the wave function of the two interacting particles of mass @xmath26 at colliding energy @xmath27 satisfies @xmath28 where @xmath23 is the relative momentum .
the zero - range interaction appears as a source term in the right hand side of eq . and the source amplitude ( denoted @xmath29 )
is obtained from a contact condition in the configuration space which reduces to an integral equation in the momentum space @xmath30 where the operator @xmath31 extracts the regular part of the integral in the limit @xmath32 . in the regime
where @xmath10 is vanishingly small , the condition expressed in eq .
is equivalent to the bethe - peierls contact condition introduced in pioneering works on the deuteron @xcite .
the formalism is now applied to three identical bosons of mass @xmath26 at negative energy @xmath33 in the center of mass frame .
the wave function of the three particles labeled @xmath34 ( @xmath35 ) with respective momentum @xmath36 factorizes as @xmath37 and @xmath38 where @xmath39 is a positive wavenumber defined by @xmath40 and @xmath41 is the source amplitude associated with the pair @xmath42 ( @xmath43 are distinct labels ) .
the contact condition in eq . applied for the pair @xmath44 with the colliding energy @xmath45 gives the integral equation satisfied by the source amplitude : @xmath46 where @xmath47 is the imaginary colliding momentum .
equation is a generalization of the stm equation ( corresponding to the case @xmath48 ) .
it was first derived by massignan and stoof ( eqs .
( 1 ) and ( 2 ) in ref .
@xcite ) and follows from the zero - range limit of the generalized stm equation in ref . @xcite . from these two references , @xmath49 can be interpreted as a dressed ( atom @xmath50 closed channel molecule ) wave function ( the molecule in the closed channel is not described by the zero range approach ) . for @xmath51 and @xmath52 (
_ i.e. _ the fr is asymptotically narrow ) , eq
. permits extraction of all the low energy three - body properties and it has been solved in refs . @xcite . for @xmath53 or @xmath54 , eq .
does not constitute a well defined problem in the @xmath18-wave sector of the source amplitude .
this can be shown along the same lines as in ref .
@xcite by considering the high momentum behavior of the eigenfunctions in the @xmath18-wave sector . in this limit ,
the scattering amplitude coincides with the unitary expression , @xmath55 and eq . supports a pair of power - law solutions of the form @xmath56 where @xmath57 .
danilov showed that the stm model is self - adjoint if one filters eigenfunctions by fixing the linear combination of the two conjugate solutions as @xmath58 with an asymptotic phase shift .
however the spectrum obtained from this filtering procedure is not bounded from below @xcite : this is the thomas collapse @xcite analog to the falling of a particle in a @xmath59 attractive potential @xcite . in the present work , instead of
the danilov s phase - shift filtering , a nodal condition is used which permits getting a finite minimal energy in the spectrum . this method is based on the expression of the source amplitude at unitarity ( _ i.e _ @xmath60 and @xmath61 ) , which is given , up to a normalizing constant by @xcite : @xmath62 .
\label{eq : exact - unitary}\ ] ] equation supports the discrete scaling symmetry found by efimov and the binding wavenumbers verify : @xmath63 where @xmath64 is an integer and @xmath15 is the three - body parameter . at a large momentum ,
one obtains : @xmath65 , \label{eq : def_kappa*}\ ] ] which gives the relation between the three - body parameter and the danilov s phase shift . for @xmath66 , the spectrum of the zero - range ( or universal ) theory is invariant in the transformation @xmath67 showing that @xmath68 is not unique
however , for interatomic forces of finite range @xmath12 , the binding wavenumber @xmath69 in eq . is bounded from below by @xmath70 . in what follows ,
@xmath68 is chosen to be of the order of @xmath71 meaning that the scaling law in eq .
is supposed to be valid for @xmath72 .
unphysical solutions of eq . are characterized by a large binding wave number @xmath73 and are also given by eq .
for the zero - range theory at unitarity . by noting that as @xmath23 increase from @xmath74 , the first node in the expression of eq
. is found at a momentum of the order of @xmath39 , it is possible to filter out the unphysical eigenfunctions of the stm equation by imposing the nodal condition : @xmath75 that way the spectrum obtained from the stm equation at unitarity has a minimum energy @xmath76 ( with a relative error given by @xmath77 ) and for @xmath78 , it coincides asymptotically ( _ i.e. _ for large @xmath79 ) with the efimov spectrum of the universal theory @xcite . in the regime
where @xmath80 and/or @xmath81 , the three - body parameter @xmath68 characterizes the high - momentum behavior of the eigenstates ( _ i.e. _ the danilov s phase - shift ) given by eq . which is a supplementary and necessary contact condition at the three - body level .
in this regime of parameters , at least if a trimer exists , the binding wave numbers no longer verify the scaling law eq . .
formally , the fact that @xmath68 is a parameter which is independent of the low energy two - body properties ( and thus of the width radius @xmath10 ) can be seen as resulting from the high momentum behavior of the solution of the generalized stm equation ( for finite @xmath82 ) which is of the same type as the universal solutions . in the specific limit where @xmath83 and @xmath84 in eq . , eigenfunctions are non zero in the region of momentum less than the order of @xmath85 . in this limit
, results are thus independent of @xmath68 and the spectrum is characterized by another momentum hereafter denoted @xmath86 @xcite and also called three - body parameter in the latter references . in the general case where the parameters in eq .
are finite , the nodal condition in eq .
can again be used to prevent any thomas collapse without modifying eq . .
in numerical computations there always exists a large but finite high momentum cut - off ( denoted hereafter @xmath87 ) in the evaluation of the right hand side of eq .
and one needs a stable procedure to implement the nodal condition eq . .
this can be achieved by subtracting term by term eq . from its own expression at @xmath88 where eq .
is also used .
one obtains the regularized integral equation : @xmath89 \phi(u ) , \label{eq : sko_regular}\end{gathered}\ ] ] where the kernel is given by : @xmath90 one can verify in numerical solutions of eq . , that results are invariant in a change of the uv cut - off @xmath87 of the integral term for @xmath91 . as a benchmark ,
the source amplitude @xmath92 has been computed at unitarity for the eight first trimers with the uv cut - off @xmath93 and are displayed in fig .
( [ fig : function_univ ] ) .
this figure illustrates the high degree of convergence of the method even for the first excited states ( which is a consequence of the relatively large value of the scale factor @xmath94 ) .
moreover , the binding wavenumbers of the excited trimers differ from the exact values of eq . by a relative error of less than @xmath95 . while not equivalent , the present regularization procedure of the stm equation as some similarity to that used in the effective - field - theory approach to the three - boson problem @xcite where an integral counter - term is also introduced to cancel the @xmath87 dependence .
the present nodal condition has the advantage of making clear that the integral counter - term is just a way to achieve a filtering of the `` physical '' solutions which are eigenfunctions of the generalized stm equation eq . .
in refs .
@xcite a subtraction scheme has been also introduced for the atom - dimer scattering equation [ see eq . ] in the framework of the effective field theory . in the latter references ,
the renormalization is done at zero energy with the atom - dimer scattering length as the input parameter and this regularizing method does not refers to a nodal condition .
interestingly , this subtraction method was used in ref .
@xcite to evaluate linear effective range corrections to the universal predictions @xcite near resonance @xmath96 in the regime where the effective range and @xmath12 are of the same order . in ref .
@xcite another regularizing technique was used : a supplementary effective range term @xmath97 was introduced in the two - body phase shift of eq . with a fitting parameter @xmath98 of the order of @xmath10 , thus modifying the two - body low energy properties in the regime of a large width radius @xcite .
the present model can be used to exemplify the crossover in the efimov spectrum as one varies the two parameters @xmath99 going from an infinitely broad fr @xmath100 to an asymptotically narrow fr @xmath101 . to focus the discussion on this issue , in what follows the model
is solved at resonance @xmath102 . for a negative background scattering length @xmath103 , there is no dimer , and the results are displayed in fig .
( [ fig : spec_rstar_mabg ] ) .
this figure shows the crossover between two types of efimov spectra , each supporting the discrete symmetry of scaling factor @xmath104 but having different ground states .
the first one is the spectrum of the effective range approach @xcite characterized by a ground state binding wave number @xmath105 where @xmath106 @xcite .
the second is the spectrum of a broad resonance where @xmath107 . for a vanishingly small background scattering length @xmath108 and a large width radius , the spectrum obtained from eq .
coincides with the first one . as @xmath109 grows at a fixed and large value of the width radius [ fig .
( [ fig : spec_rstar_mabg]-a ) ] , deeper broad - type trimers ( @xmath110 ) appear in the spectrum and narrow - type states evolve gradually towards the broad type states .
interestingly by noticing that the high momentum behavior of the scattering amplitude is given by : @xmath111 one can deduce from the universal theory the threshold of apparition of the deepest trimer by using the formal substitution @xmath112 : @xmath113 ( value of @xmath114 in ref .
second and third thresholds for the apparition of the two others deep trimers in fig .
( [ fig : spec_rstar_mabg]-a ) are deduced from the first threshold value by the discrete symmetry of scaling factor @xmath115 . as shown in figs .
( [ fig : function_cross]-[fig : function_cross2_log ] ) broad type states are located in a momentum region @xmath116 ( this is a consequence of eq . ) and narrow type states are located in the region @xmath117
. therefore , they experience a very small coupling and this explains the tiny avoided crossings in fig .
( [ fig : spec_rstar_mabg]-a ) . in the case of a positive background scattering length @xmath118
, the model supports a dimer with a binding wave number @xmath119 . for @xmath120 ,
one finds eigenstates of the broad type. for @xmath121 there is an atom - dimer continuum , and in what follows eq .
( [ eq : skorniakov ] ) is solved for a @xmath18-wave atom - dimer scattering process . for an incoming atom - dimer plane wave of momentum @xmath122 ,
the wavenumber @xmath39 is given by @xmath123 and the @xmath18-wave scattering ansatz is : @xmath124 one recognizes in eq .
the @xmath18-wave atom - dimer scattering amplitude defined by @xmath125 . injecting eq . in eq .
, one obtains : @xmath126 + { \cal p } \int_0^\infty\!\ ! \frac{du}{\pi } \ , \frac{{\mathcal k}_q(k , u ) } { u^2-k_0 ^ 2 } g(u)=\\ \frac{3 g(k)}{8(q_{\rm dim}+q_{\rm col } ) } \left[1+\frac{(1+\frac{q_{\rm col}}{q_{\rm dim}})(1 -\frac{1}{a q_{\rm dim}})}{r^\star a_{\rm bg}(1-\frac{a_{\rm bg}}{a } ) q_{\rm col}^2 - 1 } \right ] .
\label{eq : diff}\end{gathered}\ ] ] in eq .
, the integral is performed in the sense of cauchy principal value and the filtered equation is obtained by making the substitution @xmath127 . as shown in fig .
( [ fig : quasi_efi ] ) , the atom - dimer scattering amplitude obtained after regularization of eq .
supports a series of peaks as a function of the wavenumber @xmath39 at momenta @xmath128 related by the efimov scaling factor @xmath104 . in the regime where the peaks are well separated ,
their profile are typical of fano resonances wich can be parameterized by the following expression @xcite : @xmath129 where @xmath130 is the profile parameter which is related to the asymmetry of the peak , @xmath131 is the reduced energy , @xmath132 is the binding wave number of the excited efimov state , and @xmath133 is the width of the resonance .
interestingly , for a fixed value of @xmath134 the peaks computed in fig .
( [ fig : quasi_efi ] ) are characterized approximately by the same ratio @xmath135 . as expected , in the limit where @xmath136 excited trimers are decoupled from the high momentum scale @xmath68 .
first , the peaks are located at the positions predicted by the effective range approach and second , their width goes to 0 as @xmath10 increases .
this can be interpreted from the existence of a large scale separation between quasi - bound trimers extending over the region @xmath137 and the dimer of momentum @xmath138 .
hence the production of long - lived efimov s trimers seems to be possible in this regime .
if one takes into account only spontaneous decay toward one atom and one dimer , the trimer s lifetime is given by : @xmath139 .
as an example , the peak in fig .
( [ fig : quasi_efi]-b ) at @xmath140 was computed for @xmath141 , and one finds @xmath142 . in this case
the lifetime of the quasi - bound state is limited by inelastic processes with other components in the system . in the opposite limit where @xmath143 [ see fig .
( [ fig : quasi_efi]-c ) ] , resonances are large meaning that the three - body quasi - bound state are not long - lived and peaks are located at the positions predicted by the universal theory . for negative background scattering length @xmath144 ,
the decay rate of narrow type trimers involves deep dimer state which are not described in the present model . in the regime
where @xmath145 , the eigenfunction @xmath146 experience a @xmath147 decreasing law in the region @xmath148 where @xmath149 [ see fig . ] .
this suggests that the coupling with deep dimers ( not supported by the model ) is small .
one thus expects also large lifetimes in this limit . a possible way to observe long - lived trimers
is given by the radio - frequency association technique which has been already implemented in fermionic systems @xcite .
for finite background scattering length , the present zero - range model is justified in the limit where @xmath150 . when this last condition is not satisfied , more realistic two - channel model as in refs .
@xcite should be used in order to evaluate the lifetime of the trimers .
work in this direction is in progress .
in this paper , a zero range model is introduced for solving the three - boson problem in the regime of resonant @xmath18-wave binary interactions . at the two body level
, the model takes into account the scattering length @xmath151 , the width radius @xmath6 ( related to the width of a magnetic feshbach resonance ) , and the background ( or off - resonant ) scattering length @xmath1 .
effective range @xcite and bethe peierls @xcite approaches can thus be obtained as specific limits of this formalism . for three interacting bosons ,
a generalized skorniakov ter - martirosian equation is derived .
the danilov s filtering condition that fixes the value of the three - body parameter @xmath15 is imposed through a nodal condition .
the trimers spectrum is then solved at resonance ( @xmath61 ) . for negative background scattering length @xmath144 , as one varies @xmath10 or @xmath82 for a fixed value of the three - body parameter ( @xmath15 ) , the spectrum exhibits a crossover between efimov states of narrow resonances ( characterized by an effective three - body parameter @xmath152 defined by @xmath10 only , with @xmath106 ) and efimov states of broad resonances with binding wave numbers deduced from @xmath15 by the scaling factor @xmath104 . for a positive background scattering length @xmath118 , a dimer exists . for energies higher than the dimer s energy ,
efimov states are quasi - bound and manifest themselves as fano resonances . for sufficiently large values of @xmath10
these fano resonances are narrow and the peaks are located at the positions predicted by the effective range theory . in the latter regime ,
quasi - bound efimov trimers thus appear as long lived molecules .
more generally , this result opens the issue of the stability of resonant many - boson systems in the regime of narrow resonances .
y. castin , m. jona - lasinio , c. mora and f. werner are acknowledged for discussions .
the laboratoire de physique thorique de la matire condense is umr 7600 of the cnrs and its cold atoms group is associated with the ifraf . | a filtering method is introduced for solving the zero - range three - boson problem .
this scheme permits to solve the original skorniakov ter - martirosian integral equation for an arbitrary large ultraviolet cut - off and avoiding the thomas collapse of the three particles .
the method is applied to a more general zero - range model including a finite background two - body scattering length and the effective range .
a cross - over in the efimov spectrum is found in such systems and a specific regime emerges where efimov states are long - lived . |
inflationary cosmology@xcite is the most strong candidate to explain the isotropic , homogeneous and flatness nature of the universe on cosmological scales .
this fact is supported by experimental evidence@xcite . in particular , the model of fresh inflation@xcite attempts to build a bridge between the standard and warm inflationary models and hence can be viewed as an unification of both , chaotic and warm inflation scenarios . in this model
the universe begins from an unstable primordial matter field perturbation with energy density nearly @xmath1 and chaotic initial conditions to later describing a second order phase transition with heating and particles production .
hence , radiation energy density grows during fresh inflation because the yukawa interaction between the inflaton field and other fields in a thermal bath lead to dissipation which is responsible for the slow rolling of the inflaton field though the minimum energetic configuration .
hence , slow - roll conditions are physically justified and there are not a requirement of a nearly flat potential in fresh inflation . as a consequence of the strong dissipation produced by the yukawa interaction @xmath2 ) , there is no oscillation of the inflaton field around the minimum of the effective potential@xcite .
a very interesting issue to study is the natural exit of fresh inflation to a radiation dominated epoch . in this work
we examine this topic in the framework of a scalar field which is coupled to gravity though a coupling @xmath3 . in this context
one can see that for @xmath0 radiation dominance becomes possible when the scalar field @xmath4 approaches to the minimum energetic configuration .
the system is considered though a lagrangian density of an inflaton field which is coupled to the scalar curvature @xmath5 , plus a self - interaction lagrangian @xmath6 , which describes the self - interaction of the inflaton field@xcite @xmath7,\ ] ] where @xmath8 is the gravitational constant , @xmath9 is the planckian mass . during fresh inflation @xmath10 , but at the end of fresh inflation it can hold to values close to @xmath0 . in this paper
we are interested to study this particular case .
we shall consider a 4d friedmann - robertson - walker ( frw ) metric with a line element given by @xmath11 , such that @xmath12 .
the diagonal einstein equations are given by @xmath13 , \label{3}\\ 3 h^2 + 2 \dot h & = & -8\pi g \left [ \left(\frac{1}{2 } - 2\xi\right ) \dot\phi^2 - v(\phi ) + \frac{\rho_r}{3}\right ] , \label{4}\end{aligned}\ ] ] where @xmath14 is the hubble parameter and @xmath15 is the scale factor of the universe which is considered with zero spatial curvature . if @xmath16@xcite or @xmath17@xcite , has been proposed in the literature
] describes a yukawa interaction between the inflaton field and the thermal bath , the equations of motion for @xmath4 and @xmath18 , hold @xmath19 the parameter @xmath20 , such that @xmath21 give us the equation of state of the universe , or @xmath22 with a scalar curvature @xmath23 . in the early stages of inflation
this parameter is very small : @xmath24 , but at the end of inflation it reaches values close to @xmath25 , so that , for couplings close @xmath26 , inflation ends in a radiation dominated era with @xmath27 . from the inequalities in eq .
( [ 7 ] ) we obtain @xmath28 + \rho_r \left(\frac{4}{3 } - f\right ) - f v(\phi)=0 , \label{8 } \\ & & h = \frac{2}{3 \int f dt } , \label{9}\end{aligned}\ ] ] where , because @xmath29 , one obtains @xmath30 by replacing eq .
( [ 10 ] ) in eq .
( [ 8 ] ) and later in eq .
( [ 4 ] ) , we obtain respectively the eduations for the radiation energy density and the potential @xmath31 @xmath32 , \\
v(\phi ) & = & \frac{3(1 - 8\pi \xi g \phi^2)}{8\pi g } \left [ \left(\frac{4 - 3f}{4}\right ) h^2 + \frac{3\pi}{2 } \frac{g f^2}{(1 - 8\pi \xi g \phi^2 ) } \left(\frac{h^2}{h'}\right)^2 ( 1 + 8\xi)\right ] . \end{aligned}\ ] ] fresh inflation is described by a global group @xmath33 that involve a single @xmath34-vector multiplet of scalar fields @xmath35@xcite , where the effective inflaton field is describes by the norm of their scalar components : @xmath36 .
the effective potential is described by @xmath37 , such that @xmath38 where the effective squared mass @xmath39 , @xmath40 and @xmath34 is the number of created particles due to the interaction of @xmath4 with the particles in the thermal bath @xmath41 during the fresh inflationary epoch @xmath42 , so that @xmath18 increases with the time .
therefore , the number of particles created is increased dramatically because @xmath43 , but the temperature @xmath44 increases with time .
this means that the ratio @xmath45 will be increased as fresh inflation evolves .
the dynamics of the background scalar field @xmath47 and the fluctuations @xmath46 are described respectively by the equations @xmath48
\delta\phi = 0 , \\end{aligned}\ ] ] such that the redefined fluctuations @xmath49 is can be written as a fourier expansion @xmath50,\ ] ] where @xmath51 and @xmath52 are the annihilation and creation operators that comply with the commutation rules : @xmath53 = \delta^{(3)}\left(\vec{k}- \vec{k'}\right ) , \qquad \left[a^{\dagger}_k , a^{\dagger}_{k ' } \right]=\left[a_k , a_{k ' } \right]=0.\ ] ] the dynamics of the time dependent modes @xmath54 being given by @xmath55\right]\right ] \right\ } \chi_k(t)=0.\ ] ] during inflation the infrared ( ir ) sector includes long wavelength modes with @xmath56 which are unstable .
since the decay width of the produced particles grows with time , when the inflaton approaches the minimum of the potential , there is no oscillation of @xmath4 around the minimum energetic configuration due to at the end of fresh inflation dissipation is dominant with respect to rate of expansion of the universe : ( @xmath2 ) .
hence , the reheating period avoids fresh inflation ) . at the end of fresh inflation
one would expect that the universe becomes radiation dominated . to examine this possibility we consider the case with @xmath57 , such that the inflaton field is coupled to gravity with @xmath0
. this will be possible at the end of fresh inflation when @xmath58 and @xmath59 .
hence the equation ( [ 7 ] ) holds @xmath60 for a number of degrees of freedom for relativistic particles of the order of @xmath61 one obtains that the temperature reaches values like @xmath62 which is a value very close to the gut - temperature value and corresponds to a radiation energy density @xmath63 .
the effective friction parameter at this moment will be @xmath64 , which can take the value @xmath65 after it , the radiation energy density begins to decrease as @xmath66 and so that @xmath67 decreases in a such manner that the temperature decreases as @xmath68 , which is that one expects in a radiation dominated universe . at this epoch
the effective equation of motion ( [ mod ] ) for the modes is @xmath69 when we have made use of the fact that @xmath70 .
the general solution for this equation is @xmath71 + c_2 \ , \textsc{w}\left[\frac{k^2 t^2_0}{2 m_p } , \frac{1}{2 } , 2 m_p t\right],\ ] ] where ( @xmath72 ) are the whittaker functions . in the uv sector of the spectrum
the modes are stable . however , for @xmath73 the this modes are unstable and the dominant part of the solution ( [ sol ] ) is @xmath74 .\ ] ] however , notice that the modes @xmath75 increase at early radiation times , but after it decay to zero with time without crossing the hubble horizon .
it can be seen in the figure ( [ fig1 ] ) for the values @xmath76 .
reheating is not a minor phase at the end of standard inflation .
standard inflation may cause the monopole and domain wall nonthermal symmetry restoration@xcite , which could give a very inhomogeneous universe that disagree with observation .
this last possibility is actually rather difficult in simple models of reheating with only two fields@xcite .
however , this situation can be solved in the case of multiple fields , relevant for gut models@xcite .
this is the case here studied , where the number of created particles can be justified by means of the self - interaction of the inflaton . as in the case of warm inflation@xcite ,
the main difference between fresh inflation and standard inflation resides in that here there is not oscillation of the inflaton field around the minimum of the potential , due to dissipation is too large at the end of inflation . here
, particles creation becomes during inflation , beginning it at zero temperature . in this work
we have examined the natural exit from a fresh inflationary scenario to a radiation dominated universe when the inflaton field is coupled to gravity . in the particular case where the coupling constant is @xmath0 it is possible to see that this transition is very natural at the end of inflation in such that manner that the parameter @xmath25 when the inflaton field reaches its minimum energetic configuration . at this moment
all the modes of the inflaton field fluctuations are stable and the background temperature is close to the gut one .
however only oscillate the modes with wavelength below the planckian scale .
the modes which are in the classical limit are all stable but they increase to thereafter go down to zero [ see figure ( [ fig1 ] ) ] .
this is because at this time the inflaton field is strongly coupled to gravity and the modes decay very faster than during inflation due to they are super - damped by friction .
m. b. acknowledge conicet and unmdp ( argentina ) for financial support .
99 a. starobinsky , phys . lett . * b91*:99 ( 1980 ) .
+ a. guth , phys .
* d23*:347 ( 1981 ) .
a. albrecht and p. j. steinhardt , phys .
rev . lett . * 48*:1220 ( 1982 ) ; + a. linde , phys . lett . * b116*:335 ( 1982 ) .
r. l. smoot _
et al . _ ,
j. * 396*:l1 ( 1992 ) .
m. bellini , phys . rev . * d63 * : 123510(2001 ) ; + m. bellini , phys . rev .
* d64 * : 123508(2001 ) ; + m. bellini , phys.rev . *
d67 * : 027303(2003 ) .
a. berera , phys .
* 75 * : 3218(1995 ) ; + a. berera , m. gleiser , r. o. ramos , phys .
* 83 * : 264(1999 ) .
m. bellini , gen .
. grav . * 34 * : 1953(2002 ) .
h. p. de oliveira and r. o. ramos , phys
. rev . * d57 * : 741(1998 ) .
a. albrech , p. j. steinhardt and m. s. turner , phys .
lett . * 48 * : 1437(1982 ) .
l. kofman , a. linde , and a. a. starobinsky , phys .
* 76 * : 1011(1996 ) .
d. boyanovsky , h. j. de vega , r. holman and j. f. j. salgado , phys . rev . *
d54 * : 7570(1996 ) .
b. a. basset , f. tamburini , phys .
lett . * 81 * : 2630(1998 ) .
j. m. romero , m. bellini , nuovo cim .
* b124 * : 861(2009 ) . k. nozari , m. shoukrani , astrophys.space sci .
* 339 * : 111(2012 ) . | we examine the transition from a fresh inflationary scenario to a radiation dominated universe when the inflaton field is coupled to gravity .
we show that this transition is very natural when this coupling is @xmath0 at the end of fresh inflation .
this is a very important result that shows as fresh inflation has a natural exit to the radiation dominated epoch . |
the price dynamics of financial markets has long been a focus of economics and econophysics research @xcite .
studying the volatility time series is not only crucial for revealing the underlined mechanism of financial markets dynamics , but also useful for traders . for example
, it helps traders to estimate the risk and optimize the portfolio @xcite .
the volatility series is known to be long - term power - law correlated @xcite . to better understand these correlations and characterize temporal scaling features in volatilities , recently yamasaki et al . @xcite and wang et al .
@xcite studied the statistics of return intervals @xmath0 between volatilities that are above a given threshold @xmath1 , which is an alternative way to analyze long - term correlated time series ( see ref @xcite and references therein ) .
they find that scaling and memory in the return intervals of daily and intraday financial records are similar to that found in the climate and earthquake data @xcite .
studies of financial records show that the scaling in the return intervals distribution can be well approximated by a scaling function @xcite . however , financial time series are known to show complex behavior and are not of uniscaling nature @xcite and non - linear features @xcite .
recent studies @xcite of stock markets show that the distribution of activity measure such as the intertrade time has multiscaling behavior .
thus , a detailed analysis of the scaling properties of the volatility return intervals is of interest . it might improve our understanding of the return intervals statistics and shed light on the underlined complex mechanism of the volatility
our analysis suggests that for all standard & poor s index constituents , the cumulative distributions of the return intervals depart slightly but systematically from a single scaling law .
we also find that the moments @xmath10 are consistent with the deviations from scaling .
however , using the corresponding surrogate records @xcite which remove the non - linearities , @xmath11 almost does not depend on @xmath4 and no deviation from scaling occur .
therefore , our results suggest that non - linear correlations in the volatility account for the deviations from a scaling law .
the paper is organized as follows : in section ii we introduce the database and define the volatility . in section iii
we discuss the scaling and investigate the deviations from scaling in the cumulative distributions of the return intervals .
we also describe the stretched exponential form suggested for the distribution and the generation of the surrogate records . section iv deals with the moments of the return intervals .
we quantify the deviation from the scaling that exhibits multiscaling behavior .
we simulate the return intervals with different sizes and show the finite size effect for long @xmath4 .
we also study the discreteness effect for short @xmath4 and explore the relation between the moment and its order . in section
v we present a discussion .
we analyze the trades and quotes ( taq ) database from new york stock exchange ( nyse ) , which records every trade for all the securities in united states stock markets .
the duration is from jan 1 , 2001 to dec 31 , 2002 , which has a total of 500 trading days .
we study all 500 companies which consist of the standard & poor s 500 index ( s&p 500 ) @xcite , the benchmark for american stock markets .
the volatility is defined the same as in @xcite .
first we take the absolute value for the logarithmic price change , then remove the intraday u - shape pattern , and finally normalize it with the standard deviation . here
the price is the closest tick to a minute mark .
thus the sampling time is 1 minute and a trading day usually has 391 points after removing the market closing hours .
for each stock , the size of dataset is about 200,000 records .
the probability density function ( pdf ) for the return intervals @xmath0 of the financial volatilities is well - approximated by the following form , @xmath12 as analyzed by yamasaki et al .
@xcite and wang et al .
@xcite . here
@xmath13 stands for the average over the dataset and @xmath14 depends on the threshold @xmath1 .
it was suggested that the scaling function can be approximated by a stretched exponential @xcite , @xmath15 for financial records , which is consistent with other long - term correlated records @xcite . here
@xmath16 and @xmath17 are fitting parameters and @xmath18 is the exponent characterizing the long - term correlation @xcite . from the normalization of pdf
follows @xcite , @xmath19 from the definition of @xmath4 follows , @xmath20 thus , using eqs .
( [ normal.eq ] ) and ( [ mean.eq ] ) , the parameters @xmath16 and @xmath17 can be expressed by @xmath18 , @xmath21 here @xmath22 is the gamma function .
thus , if the stretched exponential distribution is valid for the scaled interval @xmath14 , it is completely determined by @xmath18 . for @xmath23 ,
the record has no long - term correlations and the return interval distribution indeed follows an exponential distribution , represented by a poissonian statistics , as expected .
though the scaling in the return intervals distribution is a good approximation , we find slight deviations that as shown below are attributed to non - linear features . to explicitly explore the quality of the scaling in return interval distributions
, we study all s&p 500 constituents and show the results of four representative stocks , citigroup ( c ) , general electric ( ge ) , coca cola ( ko ) and exxon mobil ( xom ) .
all other stocks studied here usually show similar features .
first , we examine the cumulative distribution of the scaled intervals .
@xmath24 if the scaling function @xmath25 is valid , the cumulative distributions should also collapse to a single curve .
otherwise , the cumulative distributions , which integrate deviations , may show clearer deviations from scaling .
indeed , in fig . [ fig1 ]
we show cumulative distributions for three thresholds @xmath26 , @xmath27 and @xmath28 .
note that the volatility is normalized by its standard deviation , the threshold @xmath1 is in units of standard deviations and therefore @xmath29 is a quite large volatility .
it is clearly seen that those distributions are close to each other but do not collapse to a single curve .
more important , they show apparent deviations from the scaling , which are systematic with the threshold . for small scaled intervals ( @xmath30 ) , the cumulative distribution decreases with @xmath1 , while for large scaled intervals ( @xmath31 ) , it increases with @xmath1 @xcite .
in other words , the scaled interval prefers to be larger for higher threshold .
this systematic trend suggests multiscaling in the return intervals , which might be related to the non - linear correlations in the volatility . to better understand the systematic trends and test
if it is not due to finite size effect or discreteness of minutes , we also measure the cumulative distribution of return intervals for surrogate records of volatilities using the schreiber method @xcite where non - linearities are removed .
for a given time series , we store the power spectrum and randomly shuffle the sequence , then we apply the following iterations .
each iteration consists of two consecutive steps : \(i ) we perform the fourier transform of the shuffled series , replace its power spectrum with the original one , then take the inverse fourier transform to achieve a series .
this step enforces the desired power spectrum to the series , while the distribution of volatilities usually is modified .
\(ii ) by ranking , we exchange the values of the resulting series from step ( i ) with that of the original record .
the largest value in the resulting series is replaced by the largest one in the original series , the second largest value is replaced by the second largest one , and so on .
this step restores the original distribution but now the power spectrum is changed . to achieve the convergence to the desired power spectrum and distribution , we repeat these two steps 30 times . by this way , a `` surrogate '' series is generated .
because of the wiener - kinchine theorem @xcite , the surrogate record has the same linear correlations as the original , as well as the distribution .
the only difference is that the original record has the non - linear correlations ( if they exist ) but the surrogate does not have any non - linear features . in fig .
[ fig1 ] we also plot the cumulative distribution for the surrogate with the same three thresholds as the original .
since the surrogate records lost the non - linear correlations , they are similar to each other , we only show results for ge s surrogate .
it is seen that the collapse of the surrogate for different @xmath1 values is significantly better than that of the original and the deviation tendency with the threshold in the original records disappears .
this indicates that the scaling deviations in the original are due to the non - linear correlations in the volatility . to further test this hypothesis
, we analyze the moments @xmath10 in sec .
( iv ) and show similar and consistent deviations from scaling .
we also compare our results to the stretched exponential distribution ( dashed lines ) .
this curve is very close to the empirical results , in particular for the surrogate records which contain only the linear correlations .
this suggests that pdf of return intervals is well approximated by a stretched exponential .
the cumulative distribution shows clear systematic trend with @xmath1 , which is difficult to see from the pdf directly @xcite . to further analyze the systematic tendency in the distribution
, we calculate the moments @xmath11 averaged over a stock dataset as a function of @xmath4 , where a mean interval @xmath4 corresponds to a threshold @xmath1 and therefore characterizes a return interval series .
we study moments for a wide range of @xmath4 , from @xmath32 minutes ( to avoid the artificial effects due to discreteness close to @xmath33 ) to thousands minutes ( few trading days or even a week ) . assuming a single scaling function for the pdf @xmath34 , eq .
( [ pdf.eq ] ) , it follows @xmath35 which only depends on @xmath2 and on the form of the scaling function @xmath36 but independent of @xmath4 .
thus , if @xmath11 depends on @xmath4 , it suggests deviation from the assumption of scaling .
first we examine the relation between the moments @xmath11 and the mean interval @xmath4 .
[ fig2 ] shows four representative moments @xmath37 , @xmath38 , @xmath39 and @xmath27 for stock c , ge , ko and xom . ignoring small fluctuations , which is usually due to
limited size data , all moments @xmath11 for the original records deviate significantly from a horizontal line , which is expected for a perfect scaling of the pdf .
they depend on @xmath4 and show some systematic tendency . for @xmath40 ,
moments have similar convex structure , first @xmath11 increases with @xmath4 and then decreases , where the crossover starts earlier for larger @xmath2 . for @xmath41 , moments also show similar tendency but in the opposite direction compared to @xmath40 .
these deviations from scaling in @xmath11 are consistent with the deviations seen in the cumulative distributions shown in fig .
moments of large @xmath2 ( @xmath40 ) represent large @xmath14 in the pdf and they initially ( for @xmath42 ) increase with @xmath4 , while moments of small @xmath2 ( @xmath41 ) represent small @xmath14 and they initially ( for @xmath42 ) decrease with @xmath4 . to further test
if the systematic deviations are not due to finite size effects and discreteness , we also examine moments for the surrogate records which are more flat for most range , as shown in fig .
[ fig2 ] . for the same order @xmath2 , the moment of the surrogate obviously differs from that of the original , especially in the medium range of @xmath4 ( @xmath5 ) .
this discrepancy suggests that the non - linear correlations exist in the original volatility and accounts for the scaling deviations .
nevertheless , all moments of surrogate show small curvature from a perfect straight line at both short and long @xmath4 , which are much weaker compared to the original records .
the weak curvature suggests that some additional effects , not related to the non - linear correlations , affect the moments . for small @xmath4
, the resolution discrete limit seems to have some influence on the moments .
we will discuss this effect in section c. for large @xmath4 , the moments are gradually approaching the horizontal line and are more fluctuating , the effect seems to be related to limited size of the record .
this effect will be discussed in section b. for the original volatility records , the systematic tendency in the distribution of @xmath0 and the moments implies that the return intervals may have multiscaling features . to avoid as much as possible the effect of discreteness and finite size
, we calculate the moments only for some medium range of @xmath4 where the effects are small .
since there is no non - linear correlations in the surrogate records , the curvature in their moments is only due to the additional effects , we use the surrogate curve as our reference . for small @xmath4 ,
the increasing ( decreasing ) range for @xmath40 ( @xmath41 ) almost ends at @xmath43 minutes . for large @xmath4
, the curves start to decrease ( increase ) from different positions , but at @xmath44 , all curves do not or just start to decrease ( increase ) .
thus we choose to study @xmath11 in the region , @xmath5 , represented by the shadow areas in fig .
[ fig2 ] . in this range ,
we find a clear trend for the original records while the surrogate is almost horizontal . to quantify the tendency
, we fit the moments with a power - law , @xmath45 if the distribution of @xmath14 follows a scaling law , the exponent @xmath6 should be some value very close to @xmath46 .
if @xmath6 is significantly different from @xmath46 , it suggests multiscaling .
to examine the multiscaling behavior for the whole market , we calculate @xmath6 for all 500 stocks of s&p 500 constituents and plot the histogram for @xmath37 to @xmath39 .
[ fig3 ] shows that each histogram has a narrow distribution , which suggests that @xmath6 are similar for the 500 stocks .
for the original records , almost all @xmath6 significantly differ from @xmath46 , thus the moments clearly depend on the mean interval . moreover , the mean value of @xmath6 shifts with order @xmath2 from @xmath47 for @xmath37 to @xmath48 for @xmath49 which means the dependence varies with the order @xmath2 .
this behavior suggests multiscaling in the return intervals distribution .
indeed , histograms for the surrogate records are more centered around values close to @xmath9 .
the uniscaling behavior for the surrogate suggests that the non - linear correlations in the volatility are responsible for the multiscaling behavior in the original . to remove fluctuations and show the tendency clearer
, we plot the dependence of @xmath50 on @xmath2 , where @xmath50 is the average @xmath6 over all 500 stocks . in fig .
[ fig4 ] we show this relation for a wide range of @xmath2 , @xmath51 , and the plot shows two different behaviors . for small @xmath2 ( roughly @xmath52 ) ,
@xmath50 for the original records clearly deviates from 0 and demonstrates the multiscaling behavior , while @xmath50 for the surrogate is closer to @xmath46 . for large @xmath2 ( @xmath53 ) , the two curves have similar decreasing trend .
since large @xmath0 dominates high order moments , this similarity may be due to finite size effects . to test the finite size effects we simulate surrogate return intervals by assuming a stretched exponential distribution i.i.d .
process with 3 sizes ( number of all @xmath0 in the series ) , @xmath54 , @xmath55 ( the size of the empirical dataset ) and @xmath56 .
without loss of generality , we choose @xmath57 , which is the correlation exponent for ge of @xmath26 . to be consistent with the 500 stocks , we perform 500 realizations and plot their average exponent @xmath50 . as shown in the inset of fig .
[ fig4 ] , all @xmath50 show a similar decreasing trend as that of the empirical curves .
however , it is seen that the trend starts earlier for smaller size , and thus the size limit has a strong influence on high order moments .
[ fig4 ] also shows the error bars for the two records , which is the standard deviation of 500 @xmath6 values .
note that the error bars for the volatility records do not overlap those of their corresponding surrogate , indicating the significance of our results . for small @xmath4 ( @xmath58 ) ,
the behavior of @xmath11 as a function of @xmath4 was attributed to the discreteness . here
we examine this effect . due to the limits in recording
, we can not have a continuous but discrete record . in our study
the volatility is recorded in 1 minute .
the relative errors in moments will be considerable large for small @xmath4 close to @xmath59 minute . by starting from @xmath60
, we only partially avoid the discreteness in the moments . to test the discreteness effects ,
since we can not increase the resolution , we reduce it and compare the moments @xmath11 with 3 resolutions , 1 minute , 5 minutes and 10 minutes . fig . [
fig5](a ) shows this comparison for ge for @xmath61 and @xmath49 .
the three resolutions have a similar trend , showing that the curves become flatter for the higher resolution . for other stocks
, we find similar behavior .
this systematic tendency suggests that the recording limit ( 1 minute ) strongly affects the moments at short @xmath4 . to reduce it
, we should raise the recording precision or study the moments of larger @xmath4 . to further test this result , we simulate artificial return intervals with an i.i.d .
process from stretched exponential distribution with @xmath57 , same as the empirical @xmath18 for ge of @xmath26 .
we examine moments of @xmath61 and @xmath49 with the same three resolutions ( 1 , 5 and 10 time units ) as in the empirical test done above .
the simulated size is 200 thousands points for each trial and we use the average over 100 trials for each resolution .
[ fig5](b ) shows curves similar to that of empirical ( fig .
[ fig5](a ) ) . for the higher resolution
, the curve is closer to the horizontal line and finally may reach the line when we raise the resolution high enough . to show this , we also simulate continuous return intervals and find constant moments , as expected ( fig .
[ fig5](b ) ) .
therefore , the discreteness effects can be overcome if the resolution is improved enough .
note that for the empirical data , we expect the moment not to be constant for small @xmath4 even if we have a much better resolution , since the return intervals has the multiscaling behavior , as shown for larger @xmath4 in the range @xmath5 which is not affected by discreteness . the moments @xmath11 have systematic dependence on @xmath2 , as seen in figs .
[ fig2 ] and [ fig3 ] where the moments are plotted as the function of @xmath4 .
it is of interest to explore the relation between the moments and @xmath2 directly .
for a fixed @xmath4 , representing a given threshold @xmath1 one can study , the return intervals and their moments of various orders which exhibit information on different scales of @xmath0 . moments of large @xmath2 represent large @xmath0 and vice versa . if @xmath14 follows a single distribution without corrections due to effects such as discreteness and finite size , curves of @xmath11 vs. @xmath2 for different @xmath4 should collapse to a single one , which only depends on the scaling function @xmath36 from eq .
( [ moment.eq ] ) . in fig .
[ fig6 ] we plot @xmath11 vs. @xmath2 for both the original and surrogate records .
we plot @xmath11 for @xmath2 between @xmath62 and @xmath63 for three @xmath4 values : 10 , 80 and 400 minutes . for the original ( fig .
[ fig6](a ) ) , there is substantial deviations from a single curve .
this supports our suggestion that the return intervals has multiscaling behavior .
moments for the surrogate ( fig . [ fig6](b ) ) converge to a single curve for @xmath64 but become diverse for high orders , which agrees with the strong influence of the finite size effects . as a reference , we also plot the analytical moments ( fig .
[ fig6](c ) ) from the stretched exponential distribution .
substituting eq .
( [ scaling.eq ] ) into eq .
( [ moment.eq ] ) , we obtain @xmath65 fig .
[ fig6](c ) shows analytical curves for various correlation exponent @xmath18 .
we study the scaling properties of the distribution of the volatility return intervals for all s&p 500 constituents . we find small but systematic deviations from scaling assumption with the threshold @xmath1 in the cumulative distribution . compared to the good collapse for the surrogate records where non - linearities are removed
, this suggests that the origin of this trend is due to non - linear correlations in the original volatility .
moreover , we find similar systematic deviations for the moments @xmath11 , which are also attributed to the non - linear correlations in the volatility .
we distinguish these deviations from the deviations due to the discreteness for small @xmath4 and finite size effect for large @xmath4 .
further , we explore the dependence of the moment @xmath11 on its order @xmath2 . when compare to surrogate records and to analytical curves , the results support the multiscaling hypothesis of the return intervals .
thus , the scaling assumption in the return interval distributions although it is a good approximation can not be exact .
also , the stretched exponential form of the scaling function can only be an approximation .
recently eisler et al .
@xcite exhibits that the distribution of intertrade times has similar multiscaling behavior and the market activity depends on the company capitalization .
it would be interesting to connect the intertrade times with the return intervals and test size dependence in the return intervals .
we thank y. ashkenazy for his kind help in the simulations , r. mantegna , j. kertsz and z. eisler for fruitful discussions , and the nsf and merck foundation for financial support .
h. takayasu , h. miura , t. hirabayashi , and k. hamada , physica a * 184 * , 127 ( 1992 ) ; h. takayasu , a. h. sato , and m. takayasu , phys .
lett . * 79 * , 966 ( 1997 ) ; h. takayasu and k. okuyama , fractals * 6 * , 67 ( 1998 ) .
y. liu , p. gopikrishnan , p. cizeau , m. meyer , c .- k .
peng , and h. e. stanley , phys .
e * 60 * , 1390 ( 1999 ) ; v. plerou , p. gopikrishnan , x. gabaix , l. a. nunes amaral , and h. e. stanley , quant .
finance * 1 * , 262 ( 2001 ) ; v. plerou , p. gopikrishnan , and h. e. stanley , phys .
e * 71 * , 046131 ( 2005 ) .
g. w. schwert , j. finance * 44 * , 1115 ( 1989 ) ; k. chan , k. c. chan , and g. a. karolyi , rev .
financ . stud . * 4 * , 657 ( 1991 ) ; t. bollerslev , r. y. chou , and k. f. kroner , j. econometr .
* 52 * , 5 ( 1992 ) ; a. r. gallant , p. e. rossi , and g. tauchen , rev .
financ . stud . * 5 * , 199 ( 1992 ) ; b. le baron , j. business * 65 * , 199 ( 1992 ) .
here we assume that the scaled interval @xmath14 is continuous .
this assumption is not precisely accurate since the return intervals are discrete .
however , the minimum scaled interval usually is not very large .
for example ,
@xmath66 for ge return interval of @xmath26 , the corresponding minimum is @xmath67 | the distribution of the return intervals @xmath0 between price volatilities above a threshold height @xmath1 for financial records has been approximated by a scaling behavior . to explore
how accurate is the scaling and therefore understand the underlined non - linear mechanism , we investigate intraday datasets of 500 stocks which consist of the standard & poor s 500 index .
we show that the cumulative distribution of return intervals has systematic deviations from scaling .
we support this finding by studying the @xmath2-th moment @xmath3 , which show a certain trend with the mean interval @xmath4 .
we generate surrogate records using the schreiber method , and find that their cumulative distributions almost collapse to a single curve and moments are almost constant for most range of @xmath4 .
those substantial differences suggest that non - linear correlations in the original volatility sequence account for the deviations from a single scaling law .
we also find that the original and surrogate records exhibit slight tendencies for short and long @xmath4 , due to the discreteness and finite size effects of the records respectively .
to avoid as possible those effects for testing the multiscaling behavior , we investigate the moments in the range @xmath5 , and find the exponent @xmath6 from the power law fitting @xmath7 has a narrow distribution around @xmath8 which depend on @xmath2 for the 500 stocks .
the distribution of @xmath6 for the surrogate records are very narrow and centered around @xmath9 .
this suggests that the return interval distribution exhibit multiscaling behavior due to the non - linear correlations in the original volatility . |
since the discovery of bose - einstein condensation ( bec ) of atoms in optical traps , the field of ultracold atoms has experienced tremendous developments in the recent years.@xcite a first important step has been the use of fano - feshbach resonances@xcite to tune the strength of atom - atom interaction.@xcite fano - feshbach resonances take place when the energy difference between the molecular state in the closed channel and the threshold of the two - atom continuum in the open channel , known as the detuning @xmath0 , is zero@xcite . near a fano - feshbach resonance ,
the atom - atom scattering length possesses a singularity . for @xmath1 , atoms are stable , but the existence of the virtual molecular state results in an effective attraction . for @xmath2 ,
the molecules are formed and possess a weakly repulsive interaction . since the value of @xmath0 can be controlled by an applied magnetic field , this allows to tune the sign and strength of the atomic and molecular interactions.@xcite in particular , the use of fano - feshbach resonances has allowed the observation of pairs of fermionic@xcite or bosonic@xcite atoms binding together to form bosonic molecules . at sufficiently low temperature , for @xmath2 , these molecules can form a bose - einstein condensate . in the case of a fermionic system , for @xmath1 , due to attractive interactions
a bcs superfluid is expected .
since the bec and the bcs state break the same u(1 ) symmetry , a smooth crossover between the two states is expected as @xmath0 is tuned through the resonance .
indeed , the bec of molecules@xcite and the crossover to a strongly degenerate fermi gas@xcite have been observed as a gas of cold fermionic atoms is swept through the fano - feshbach resonance .
measurement of the radio - frequency excitation spectra@xcite and of the specific heat@xcite as well as observation of vortices in a rotating system@xcite on the @xmath1 side revealed the presence of a superfluid bcs gap , thus proving the existence of a bec - bcs crossover .
such a crossover is naturally described by the boson - fermion model , @xcitefirst introduced in the 1950s in the context of the theory of superconductivity@xcite and later reinvestigated in the 1980s in the context of polaronic@xcite and high - tc superconductivity theory.@xcite a second important parallel development has been the possibility to form quasi-1d condensates using anisotropic traps@xcite , two - dimensional optical lattices@xcite or atoms on chips.@xcite in one dimensional systems interactions are known to lead to a rich physics.@xcite in particular , strongly correlated states of fermions , where individual particles are replaced by collective spin or density excitations , are theoretically expected.@xcite when the interactions between the fermions are repulsive , both the spin and density fluctuations are gapless with linear dispersion and this state is known as the luttinger liquid@xcite . for attractive interactions between the fermions , the spin degrees of freedom develop a gap , yielding a state known as the luther - emery liquid.@xcite similarly , bosons are expected to be found in a luttinger liquid state , with individual particles being replaced by collective density excitations@xcite . moreover
, strong repulsion can lead to the fermionization of interacting bosons i.e. the density matrix becomes identical to that of a non - interacting spinless fermion system , the so - called tonks - girardeau ( tg ) regime.@xcite experiments in elongated traps have provided evidence for one - dimensional fluctuations@xcite . however , in these systems , the bosons remain weakly interacting . with two - dimensional optical lattices ,
it is possible to explore a regime with stronger repulsion . in particular , it was possible to observe the tg regime with @xmath3rb atoms@xcite by increasing the transverse confinement .
the tg regime can also be reached by applying a 1d periodic potential along the tubes to increase the effective mass of the bosons@xcite . using a stronger 1d potential , it is possible to drive a one - dimensional mott transition between the superfluid state and an insulating state@xcite .
another characteristic of atoms in a one - dimensional trap is that transverse confinement can give rise to a type of fano - feshbach resonance as a function of the trapping frequency called the confinement induced resonance ( cir).@xcite recently , experiments have been performed on @xmath4k fermionic atoms in a one dimensional trap forming bound states either as a result of fano - feshbach resonances or of cir.@xcite both types of bound states have been observed and the results can be described using the boson - fermion model.@xcite this prompts the question of whether a one dimensional analogue of the bec - bcs crossover could be observed in such a system .
it is well known that in one dimension , no long range bec or bcs order can exist.@xcite however , quasi - long range superfluid order is still possible . for fermions with attractive interactions , it was shown using the exactly solved gaudin - yang model@xcite that for weakly attractive interactions , a luther - emery state with gapless density excitations and gapful spin excitations was formed , whereas for strongly attractive interactions the system would crossover to a luttinger liquid of bosons.@xcite the boson - fermion model was also considered in the case of a broad fano - feshbach resonance.@xcite in that case only bosons or fermions are present ( depending on which side of the resonance the system is ) and the results are analogous to those obtained with the gaudin - yang model .
in fact , in the three dimensional case , it is possible to derive a mapping of the boson - fermion model with a broad resonance to a model with only fermions and a two - body interaction.@xcite in the narrow resonance case , such a mapping is valid only very close to the resonance .
it was therefore interesting to investigate what happens in one dimension in the case of a narrow resonance .
indeed , in the latter case , it has been shown previously@xcite that a richer phase diagram could emerge with a phase coherence between a fluid of atoms and a fluid of molecules at weak repulsion and a decoupling transition for stronger repulsion .
analogous effects have been discussed in the context of bosonic atoms with a fano - feshbach resonance in@xcite .
due to the concrete possibility of forming 1d fermi and bose gas with optical lattices @xcite some of the theoretical predictions in the narrow resonance case may become testable experimentally in the future .
experimental signature of the phase coherence between the two fluids include density response and momentum distribution function . in the present paper ,
we investigate in more details the phase in which the atomic and the molecular fluid coexist .
in particular , we study the equilibrium between the atomic and the molecular fluid as the detuning is varied . also , we investigate the effect of placing the system in a periodic potential and show that the phase coherence between the atomic and molecular fluid hinders the formation of the mott state in systems at commensurate filling .
such conclusion is in agreement with a study in higher dimension@xcite .
the plan of the paper is the following . in sec.[sec :
hamiltonian ] we introduce the boson - fermion hamiltonian both in the lattice representation and in the continuum .
we discuss its thermodynamics in the limit of an infinitesimal boson - fermion conversion term and show under which conditions atoms and molecules can coexist . in sec.[sec : boson - appr ] we derive the bosonized expression for the boson - fermion hamiltonian valid in the region where atoms and molecules coexist .
this hamiltonian is valid for a system in an optical lattice provided it is at an incommensurate filling ( i.e. with a number of atoms per site which is not integer ) .
we show that for not too strong repulsion in the system , a phase where the atomic and the molecular superfluid become coherent can be obtained .
this phase possesses a spin gap .
we show that in this phase the order parameter for the bec and the bcs superfluidity order parameter are identical , while charge density wave correlations present an exponential decay .
we discuss the phase transitions induced by the detuning , the magnetic field and the repulsion .
we also exhibit a solvable point where some correlation functions can be obtained exactly . in sec .
[ sec : mott - insul - state ] , we consider the case where the number of atoms per site in the optical lattice is integer . we show that a phase transition to a mott insulating state can be obtained in that case . however , there is no density wave order in this mott state . finally , in sec .
[ sec : param - boson - hamilt ] , we discuss the applicability of our results to experiments .
we consider a system of 1d fermionic atoms with a fano - feshbach resonance.@xcite this 1d system can be obtained by trapping the fermions in a two dimensional or a three dimensional optical lattice . in the first case ,
the fermions are trapped into 1d tubes , in the second case , a periodic potential is superimposed along the direction of the tubes . in the case in which the fermions are injected in a uniform potential ,
the hamiltonian of the system reads : @xmath5,\end{aligned}\ ] ] where @xmath6 annihilates a molecule , @xmath7 a fermion of spin @xmath8 , @xmath9 is the mass of the isolated fermionic atom , @xmath10 the mass of the molecule , @xmath11 are ( respectively ) the molecule - molecule , atom - molecule and atom - atom interactions .
since these interactions are short ranged , it is convenient to assume that they are of the form @xmath12 .
the term @xmath0 is the detuning .
finally , the term @xmath13 allows the transformation of a pair of fermions into a fano - feshbach molecule and the reverse process .
this term can be viewed as a josephson coupling@xcite between the order parameter of the bec of the molecules , and the order parameter for the superfluidity of the fermions . as a result of the presence of this term , pairs of atoms are converted into molecules and vice - versa , as in a chemical reaction@xcite . as a result of this ,
only the total number of atoms ( paired and unpaired ) , @xmath14 ( where @xmath15 is the number of molecules and @xmath16 is the number of unpaired atoms ) is a conserved quantity . in the case where atoms are injected in a periodic potential , @xmath17 it is convenient to introduce the wannier orbitals@xcite of this potential . in the single band approximation
the hamiltonian reads : @xcite @xmath18 where @xmath19 annihilates a fermion of spin @xmath8 on site @xmath20 , @xmath21 , @xmath22 creates a fano - feshbach molecule ( boson ) on the site @xmath20 , and @xmath23 .
the hopping integrals of the fermions and bosons are respectively @xmath24 and @xmath25 .
the quantity @xmath0 is the detuning .
the parameters @xmath26 , @xmath27 and @xmath28 measure ( respectively ) the fermion - fermion , boson - boson , and fermion - boson repulsion .
the case of hard core bosons corresponds to @xmath29 .
the conversion of atoms into molecules is measured by the term @xmath30 .
again , only the sum @xmath31 is conserved .
we note that within the single band approximation , there should exist a hard core repulsion between the bosons . in this section
, we wish to study the behavior of the density of unpaired atoms @xmath33 and of the density of atoms paired in molecules @xmath34 as a function of the total density of atoms ( pair and unpaired ) @xmath35 in the limit of @xmath36 .
in such a limit , the fermion - boson conversion does not affect the spectrum of the system compared to the case without fermion - boson conversion .
however , it is imposing that only the total total number of atoms @xmath37 is conserved .
therefore , it this limit there is a single chemical potential @xmath38 and the partition function reads : @xmath39=\mathrm{tr}[e^{-\beta[h_\lambda-\mu(n_f+2n_b ) ] } ] , \end{aligned}\ ] ] and : @xmath40 in the absence of fermion - boson conversion , @xmath15 and @xmath16 would be separately conserved , and one would have a chemical potential @xmath41 for the molecules and @xmath42 for the atoms .
the partition function of this hypothetical system would read : @xmath43=\mathrm{tr}[e^{-\beta[h_0-\mu_f n_f- \mu_b n_b ] } ] , \end{aligned}\ ] ] and thus : @xmath44=z_0[\mu,2\mu ] . \end{aligned}\ ] ] if we further assume that @xmath45 , we have @xmath46 , where @xmath47 is the hamiltonian of the fermion subsystem and @xmath48 is the hamiltonian of the bosonic subsystem , and the partition function ( [ eq : partition - no - conversion ] ) factorizes as @xmath49=z_f[\mu_f ] z_b[\mu_b]$ ] , where @xmath50=\mathrm{tr}[e^{-\beta[h_\nu-\mu_\nu n_\nu}]$ ] for @xmath51 .
thus , in the limit @xmath52 , we obtain the following expression of the number of unpaired atoms @xmath16 and the number of atoms paired in molecules @xmath15 .
@xmath53 we now use these equations ( [ eq : atom - number ] ) and ( [ eq : molecule - number ] ) to study the coexistence of bosons and fermions as the detuning @xmath0 is varied .
two simple cases can be considered to illustrate this problem of coexistence .
first , one can consider bosonic molecules with hard core repulsion and noninteracting fermionic atoms .
in such a case , the thermodynamics of the gas of molecules is reduced to that of a system of spinless fermions by the jordan - wigner transformation@xcite , and the expression of the densities of unpaired atoms and molecules can be obtained in closed form . in this simple case , it is straightforward to show that for sufficiently negative detuning all atoms are paired into molecules , and for sufficiently positive detuning all the atoms remain unpaired .
the case of intermediate detuning is more interesting as coexistence of unpaired atoms with atoms paired into molecules becomes possible .
the physical origin of this coexistence is of course the molecule - molecule repulsion that makes the chemical potential of the gas of molecules increase with the density so that in a sufficiently dense gas of molecules , it becomes energetically favorable to create unpaired atoms . to show that the above result is not an artifact of having a hard core repulsion ,
we have also considered a slightly more realistic case of molecules with contact repulsion and non - interacting atoms .
although in that case we can not anymore obtain closed form expressions of the density of molecules , we can still calculate numerically the density of molecules using the lieb - liniger solution@xcite .
we will see that having a finite repulsion between the molecules indeed does not eliminate the regime of of coexistence .
in that case we assume that the boson - boson repulsion @xmath54 in the lattice case and @xmath55 in the continuum case is going to infinity . using the jordan - wigner transformation@xcite
, one shows that the partition function of these hard core bosons is equal to that of free spinless fermions thanks to the jordan - wigner transformation .
for positive temperature , the number of unpaired atoms and the number of atoms paired into molecules read : @xmath56 for @xmath57 , these equations reduce to : @xmath58 where @xmath59 is the fermi momentum of the atoms and @xmath60 is the fermi momentum of the spinless fermions ( i.e. the pseudo - fermi momentum of the molecules ) . up to now , we have not specified the dispersion of the atoms and of the molecules . in the lattice case , these dispersion are obtained from eq .
( [ eq : lattice ] ) as @xmath61 and @xmath62 . a graphical solution of ( [ eq : hard - core - conditions ] ) is shown on fig .
[ fig : chemical ] for three different values of the chemical potential @xmath38 and @xmath1 .
three different regimes are obtained . in the first one , for @xmath63 , only unpaired atoms are present . in the second one for @xmath64 ,
unpaired atoms and molecules coexist . in the last one , for @xmath65 , all the available levels of unpaired atoms are filled , and the available levels for molecules are partially filled . as a result , the system behaves as if only molecules were present .
this last phase is in fact a degenerate tonks - girardeau gas of molecules@xcite . in the intermediate regime ,
the fermions form a two - component luttinger liquid@xcite and the bosons form a single component luttinger liquid.@xcite similar calculations can be performed in the case of fermions and bosons in the continuum described by eq .
( [ eq : nolattice ] ) . with free fermions and hard core bosons in the continuum eqs .
( [ eq : hard - core - conditions ] ) become : @xmath66 with @xmath67 the total density of atoms . eliminating @xmath59 in eq .
( [ eq : conditions - hardcore - continuum ] ) , the problem is reduced to solving a second degree equation : @xmath68 the solutions of eq .
( [ eq : kb - hardcore ] ) are : @xmath69 and these solutions are physical when they yield both @xmath59 and @xmath60 positive . for @xmath1 , eq . ( [ eq : densities - hardcore ] ) yields @xmath70 provided @xmath71
. when @xmath72 , the density of molecules is vanishing and @xmath73 . above the critical density , atoms and molecules coexist , with densities given by eq .
( [ eq : conditions - hardcore - continuum ] ) . at the critical density ,
the slope of @xmath60 versus @xmath74 is discontinuous , being @xmath75 below the critical density and @xmath76 above the critical density .
the fermi wavevector @xmath59 also possesses a slope discontinuity at the critical density , the slope being zero above the critical density .
the behavior of @xmath59 and @xmath60 as a function of the density is represented on fig .
[ fig : kf - kb1 ] .
for @xmath2 , eq . ( [ eq : densities - hardcore ] ) yields @xmath77 provided @xmath78 .
when , @xmath79 the density of unpaired atoms vanishes , and @xmath80 . above the critical density ,
atoms and molecules coexist with densities given by eq .
( [ eq : conditions - hardcore - continuum ] ) . as before ,
the slope of the curve @xmath59 versus @xmath74 is discontinuous at the critical density , being zero below and @xmath81 above .
the behavior of @xmath59 and @xmath60 as a function of the density for @xmath2 is represented on fig .
[ fig : kf - kb2 ] .
the slope discontinuities in @xmath60 and @xmath59 have important consequences for the compressibility .
indeed , using eq .
( [ eq : conditions - hardcore - continuum ] ) , it is easy to see that above the critical density , the chemical potential varies as @xmath82 .
since the compressibility @xmath83 is defined as @xmath84 , this implies that the compressibility of the system becomes infinite as the critical density is approached from above , signalling a first - order phase transition .
such first order transitions associated with the emptying of a band have been analyzed in the context of luttinger liquid theory in refs .
@xcite . .
for @xmath85 , the fermion band is partially filled , and the hardcore boson band is empty . for @xmath86 , the hardcore boson band is partially filled , and the fermion band is totally filled . in the case
@xmath87 , both band are partially filled . in the rest of the paper we will only consider the latter case.,width=340 ] and
@xmath60 for positive detuning @xmath1 as a function of the total density @xmath74 . for low densities ,
only atoms are present ( @xmath88 ) . at higher densities
such that @xmath89 , a nonzero density of molecules appear .
at the critical density , the slopes of @xmath59 and @xmath60 versus @xmath74 are discontinuous . on the figure we have taken @xmath90 .
] and @xmath60 for negative detuning @xmath2 . for low densities ,
only the molecules are present ( @xmath91 ) . for @xmath92
, molecules coexist with atoms . at the critical density , the slopes of @xmath59 and @xmath60 versus @xmath74 are discontinuous . ]
we have seen in the previous section that in the case of hard core repulsion between the molecules , both in the lattice case and in the continuum case , that having @xmath2 did not prevent the formation of unpaired atoms provided the total density of atoms was large enough .
this was related with the increase of the chemical potential of bosons as a result of repulsion when the density was increased . in this section ,
we want to analyze a slightly more realistic case where the repulsion between bosons is finite and check that coexistence remains possible . in the lattice case ,
the problem is untractable by analytic methods and one needs to rely on numerical approaches.@xcite in the continuum case , however , it is well known that bosons with contact repulsions are exactly solvable by bethe ansatz techniques.@xcite the density of molecules can therefore be obtained by solving a set of integral equations.@xcite they read : @xmath93 where : @xmath94 @xmath55 being the boson - boson interaction defined in eq .
( [ eq : nolattice ] ) .
the parameter @xmath95 plays the role of a pseudo fermi momentum . for @xmath96
, we have @xmath97 .
we also have @xmath98.@xcite it is convenient to introduce dimensionless variables@xcite : @xmath99 and rewrite @xmath100 , @xmath101 , @xmath102 , @xmath103 .
the dimensionless integral equations read : @xmath104 using @xmath105 one has the following integral equation for @xmath106 : @xmath107\end{aligned}\ ] ] once this equation has been solved , the chemical potential of the bosons is obtained by : @xmath108 .
\end{aligned}\ ] ] knowing @xmath109 gives immediately @xmath110 . from @xmath111 one finds @xmath112 and @xmath113 .
finally , using the definition of the total density @xmath114 one can map the molecule density and the free atom density as a function of the total density of atoms .
the resulting equation of state can be written in terms of dimensionless parameters as : @xmath115 the behavior of the boson density @xmath116 and fermion density @xmath117 as a function of total density @xmath74 can be understood in qualitative terms .
let us first discuss the case of negative detuning . for sufficiently low densities ,
only bosons are present . however , in that regime , the boson - boson repulsion is strong , and the boson chemical potential is increasing with the boson density . as a result ,
when the density exceeds a critical density @xmath118 , the fermion chemical potential becomes positive , and the density of fermions becomes non - zero .
the appearance of fermions is causing a cusp in the boson density plotted versus the total density .
when the density of particles becomes higher , the boson - boson interaction becomes weaker , and the boson chemical potential barely increases with the density . as a result ,
the fermion density becomes almost independent of the total density . in the case of positive detuning , for low density ,
only fermions are present .
again , the increase of fermion density results in an increase of chemical potential and above a certain threshold in fermion density , bosons start to appear , creating a cusp in the dependence of the fermion density upon the total density . at large density ,
the detuning becomes irrelevant , and the fermion density barely increases with the total density . to illustrate this behavior ,
we have solved numerically the integral equations ( [ eq : lieb_dimensionless ] ) , and calculated the resulting fermion and boson densities . a plot of the density of bosons as well as the density of fermions is shown on fig .
[ fig : density - positive ] for @xmath1 and on fig .
[ fig : density - negative ] for @xmath2 .
the slope discontinuities at the critical density remain visible .
obviously , this implies that the divergence of the compressibility is still present when the repulsion between the molecule is not infinite . and unpaired atoms @xmath117 as a function of the total density @xmath74 in the case of a repulsion @xmath119 between the bosons and for positive detuning @xmath120 . at large density
, the fermion density is increasing more slowly than the boson density .
inset : the behavior of the boson and fermion densities near the origin .
note the cusp in the fermion density as the boson density becomes nonzero as in the @xmath121 case .
] and unpaired atoms @xmath117 as a function of the total density @xmath74 in the case of a repulsion @xmath119 between the bosons and for negative detuning @xmath122 . at large density
, the fermion density is increasing more slowly than the boson density .
inset : the behavior of the boson and fermion densities near the origin .
note the cusp in the boson density as the fermion density becomes nonzero as in the @xmath121 case . ]
we have thus seen that generally we should expect a coexistence of fermionic atoms and bosonic molecules as soon as repulsion between the molecules is sufficiently strong . moreover , the repulsion between the molecules results in a finite velocity for sound excitations in the molecule bose gas . as a result
, we can expect that the gas of molecules will behave as a luttinger liquid .
till now however , we have assumed that the term converting atoms into molecules was sufficiently small not to affect significantly the spectrum of the system . in the following
, we will treat the effect of a small but not infinitesimal conversion term in eqs .
( [ eq : lattice ] ) and ( [ eq : nolattice ] ) using bosonization techniques .
we will show that this term can lead to phase coherence between the atoms and the molecules , and we will discuss the properties of the phase in which such coherence is observed .
in this section , we consider the case discussed in sec .
[ sec : hamiltonian ] where neither the density of molecules nor the density of atoms vanishes . as discussed in sec .
[ sec : hamiltonian ] , this requires a sufficiently large initial density of atoms . as there is both a non - zero density of atoms and of molecules , they both form luttinger liquids@xcite .
these luttinger liquids are coupled by the repulsion between atoms and molecules @xmath123 and via the conversion term or josephson coupling @xmath13 . to describe these coupled luttinger liquids
, we apply bosonization@xcite to the hamiltonians ( [ eq : lattice]) ( [ eq : nolattice ] ) . for the sake of definiteness
, we discuss the bosonization procedure in details only in the case of the continuum hamiltonian ( [ eq : nolattice ] ) . for the lattice hamiltonian ( [ eq : lattice ] ) , the steps to follow are identical provided the system is not at a commensurate filling . at commensurate
filling , umklapp terms must be added to the bosonized hamiltonian and can result in mott phases@xcite .
this case is treated in sec .
[ sec : mott - insul - state ] . to derive the bosonized hamiltonian describing the low - energy spectrum of the hamiltonian ( [ eq : nolattice ] ) ,
we need first to consider the bosonized description of the system when all atom - molecule interactions are turned off . for @xmath124 , both @xmath16 and @xmath15
are conserved and the bosonized hamiltonian equivalent to ( [ eq : lattice ] ) or ( [ eq : nolattice ] ) is given by : @xmath125 \nonumber \\
\frac{dx}{2\pi } \left [ u_\rho k_\rho ( \pi \pi_\rho)^2 + \frac { u_\rho } { k_\rho}(\partial_x\phi_\rho)^2\right ] \nonumber \\
h_\sigma&=&\int \frac{dx}{2\pi } \left [ u_\sigma k_\sigma ( \pi \pi_\sigma)^2 + \frac { u_\sigma } { k_\sigma}(\partial_x\phi_\sigma)^2\right ] -\frac{% 2g_{1\perp}}{(2\pi\alpha)^2 } \int dx \cos \sqrt{8}\phi_\sigma\end{aligned}\ ] ] where @xmath126=i% \delta(x - x^{\prime})\delta_{\nu,\nu^{\prime}}$ ] , ( @xmath127 ) . in the context of cold atoms , the hamiltonian ( [ eq : bosonized - spin ] )
have been discussed in @xcite .
the parameters @xmath128 , the luttinger exponent , and @xmath129 , the charge and spin velocities , are known functions of the interactions @xcite , with @xmath130 in the non - interacting case , @xmath131 is a marginally irrelevant interaction , and at the fixed point of the rg flow @xmath132 . for the bosonic system , the parameters @xmath133 can be obtained from numerical calculations@xcite in the lattice case or from the solution of the lieb - liniger model@xcite in the continuum case . in the case of non - interacting bosons @xmath134 and in the case of hard core bosons
@xmath135.@xcite an important property of the parameters @xmath136 and @xmath137 is that they decrease as ( respectively ) the boson - boson and fermion - fermion interaction become more repulsive .
the bosonized hamiltonian ( [ eq : bosonized - spin ] ) is also valid in the lattice case ( [ eq : lattice ] ) provided that both @xmath16 and @xmath15 do not correspond to any commensurate filling .
the fermion operators can be expressed as functions of the bosonic fields appearing in ( [ eq : bosonized - spin ] ) as @xcite : @xmath138(x)}}{\sqrt{2\pi\alpha}},\end{aligned}\ ] ] where the index @xmath139 indicates the right / left movers , @xmath140 is a cutoff equal to the lattice spacing in the case of the model eq .
( [ eq : lattice ] ) .
similarly , the boson operators are expressed as@xcite : @xmath141.\end{aligned}\ ] ] in eqs .
( [ eq : fermion - bosonized])-([eq : boson - bosonized ] ) , we have introduced the dual fields@xcite @xmath142 ( @xmath143 ) , @xmath144 , and @xmath145 where @xmath146 is the length of the system .
the fermion density is given by@xcite : @xmath147 and the boson density by@xcite : @xmath148 the detuning term in ( [ eq : nolattice ] ) is thus expressed as : @xmath149 we now turn on a small @xmath13 and a small @xmath123 .
the effect of a small @xmath123 on a boson - fermion mixture has been investigated previously@xcite .
the forward scattering contribution is : @xmath150 and as discussed in @xcite , it can give rise to a phase separation between bosons and fermions if it is too repulsive . otherwise , it only leads to a renormalization of the luttinger exponents .
the atom molecule repulsion term also gives a backscattering contribution : @xmath151 however in the general case , @xmath152 this contribution is vanishing . in the special case of @xmath153 , the backscattering can result in the formation of a charge density wave .
this effect will be discussed in sec .
[ sec : quantum - ising ] .
the contribution of the @xmath13 term is more interesting.@xcite using eqs .
( [ eq : fermion - bosonized])-([eq : boson - bosonized ] ) , we find that the most relevant contribution reads : @xmath154 in the next section , we will see that this term gives rise to a phase with atom - molecule coherence when the repulsion is not too strong .
the effect of the term ( [ eq : lambda - bosonized ] ) on the phase diagram can be studied by renormalization group techniques@xcite .
a detailed study of the renormalization group equations has been published in @xcite . here , we present a simplified analysis , which is sufficient to predict the phases that can be obtained in our system .
the scaling dimension of the boson - fermion coupling term ( [ eq : lambda - bosonized ] ) is : @xmath155 . for small @xmath13
it is reasonable to replace @xmath156 with its fixed point value @xmath157 .
therefore , the rg equation for the dimensionless coupling @xmath158 ( where @xmath159 is one of the velocities @xmath160 ) reads : @xmath161 where @xmath162 is related to the renormalized cutoff @xmath163 .
we thus see that for @xmath164 , this interaction is relevant .
since for hardcore bosons@xcite @xmath165 and for non - interacting bosons @xmath166 , while for free fermions @xmath167 and in the lattice case for @xmath168 one has @xmath169@xcite , we see that the inequality is satisfied unless there are very strongly repulsive interactions both in the boson system and in the fermion system .
when this inequality is not satisfied , for instance in the case of fermions with nearest - neighbor repulsion@xcite , in which one can have @xmath170 and hardcore bosons with nearest neighbor repulsion@xcite , in which one can have @xmath171 , the atoms and the molecules decouple .
this case is analogous to that of the mixture of bosons and fermions@xciteand charge density waves can be formed if @xmath60 and @xmath59 are commensurate .
the phase transition between this decoupled phase and the coupled phase belongs to the berezinskii - kosterlitz - thouless ( bkt ) universality class.@xcite as pointed out in @xcite , in the decoupled phase , the effective interaction between the fermions can be attractive . in that case ,
a spin gap is formed @xcite and the fermions are in a luther - emery liquid state with gapless density excitations .
let us consider the coupled phase in more details .
the relevance of the interaction ( [ eq : lambda - bosonized ] ) leads to the locking of @xmath172 , i.e. it results in the formation of a spin gap . to understand the effect of the term @xmath173
, it is better to perform a rotation : @xmath174 and the same transformation for the @xmath175 .
this transformation preserves the canonical commutation relations between @xmath176 and @xmath177 . under this transformation
, @xmath178 becomes : @xmath179 \nonumber \\ & & + \int \frac{dx}{2\pi } [ g_1 ( \pi \pi_+)(\pi \pi_- ) + g_2 \partial_x\phi_+ \partial_x\phi_-],\end{aligned}\ ] ] where : @xmath180 while , @xmath181 defined in ( [ eq : lambda - bosonized ] ) becomes : @xmath182 after the rotation , we see that when @xmath13 is relevant , the field @xmath183 is also locked , but @xmath184 remains gapless . since the field @xmath185 is the difference of the superfluid phase of the atoms and the one of the molecules , this means that when @xmath13 becomes relevant , unpaired atoms and molecules share the same superfluid phase i.e. they become coherent .
the gap induced by the term @xmath13 can be estimated from the renormalization group equation ( [ eq : rg - coupling ] ) . under the renormalization group
, the dimensionless parameter @xmath186 grows until it becomes of order one at a scale @xmath187 where the perturbative approach breaks down . beyond the scale @xmath188 , the fields @xmath185 and @xmath172 behave as classical fields
therefore , the associated energy scale @xmath189 is the scale of the gap . from this argument
, we obtain that the gap behaves as : @xmath190 the gapful excitations have a dispersion @xmath191 and are kinks and antikinks of the fields @xmath185 and @xmath172.@xcite more precisely , since a kink must interpolate between degenerate classical ground states of the potential ( [ eq : lambda - change ] ) , we find that when a kink is present @xmath192 and @xmath193 .
this indicates that a kink is carrying a spin @xmath194 , and is making the phase @xmath195 of the bosons jump by @xmath196 and the phase of the superfluid order parameter @xmath197 of the fermions jump by @xmath198 .
since the current of bosons is @xmath199 and the current of fermions is @xmath200 , this indicates that counterpropagating supercurrents of atoms and molecules exist in the vicinity of the kinks .
therefore , we can view the kinks and antikinks are composite objects formed of vortices bound with a spin 1/2 .
we note that the kinks and antikinks may not exhaust all the possible gapful excitations of the system .
in particular , bound states of kinks and antikinks , known as breathers may also be present@xcite . however , these gapful excitations present a larger gap than the single kinks .
let us now turn to the gapless field @xmath184 .
this field has a simple physical interpretation . considering the integral @xmath201 showing that @xmath202 measures the total number of particles in the system @xmath203 .
thus @xmath204 describe the total density excitations of the system .
the resulting low - energy hamiltonian describing the gapless total density modes reads : @xmath205,\end{aligned}\ ] ] where @xmath206 denote renormalized values of @xmath207 .
this renormalization is caused by the residual interactions between gapless modes and the gapped modes measured by @xmath208 in eq .
( [ eq : b - plus - rho ] ) .
since @xmath184 measures the total density , the hamiltonian ( [ eq : gapless - modes ] ) describes the propagation of sound modes in the 1d fluid with dispersion @xmath209 .
we note than in ref.@xcite , dispersion relation similar to ours were derived for the sound modes and the superfluid phase difference modes using different methods .
having understood the nature of the ground state and the low excited states when @xmath13 is relevant we turn to the effect of the detuning term .
( [ eq : detun - bosonized ] ) and ( [ eq : rot ] ) show that the detuning term can be expressed as a function of @xmath210 as : @xmath211 this shows that the detuning does not affect the boson - fermion coupling ( [ eq : lambda - bosonized ] ) since it can be eliminated from the hamiltonian by a canonical transformation @xmath212 , where @xmath213 and @xmath214 .
for a fixed total density , changing the detuning only modifies the wavevectors @xmath60 and @xmath59 . as discussed extensively in sec .
[ sec : continuum - case ] , for a detuning sufficiently large in absolute value , only molecules or only atoms are present , and near the critical value of the detuning , the compressibility of the system is divergent . we therefore conclude that in one - dimension , the crossover from the bose condensation to the superfluid state , as the detuning is varied , is the results of the band - filling transitions at which either the density of the atoms or the molecules goes to zero . at such band
filling transitions,@xmath215 ( respectively @xmath216 ) and bosonization breaks down @xcite .
the two band filling transitions are represented on fig .
[ fig : band - filling ] .
the cases at the extreme left and the extreme right of the phase diagram have been analyzed in @xcite , where it was shown that in the case of a broad fano - feshbach resonance , the zone of coexistence was very narrow . in the narrow fano - feshbach resonance case we are investigating , the zone of coexistence can be quite important .
application of a magnetic field can also induce some phase transitions .
the interaction with the magnetic field reads : @xmath217 the effect of the magnetic field is to lower the gap for the creation of kink excitations ( remember that they carry a spin @xmath194 ) . as a result ,
when it becomes larger than the gap , the magnetic field induces a commensurate incommensurate transition@xcite that destroys the coherence between atoms and molecules and gives back decoupled luttinger liquids@xcite . in that regime ,
the behavior of the system is described by the models of ref . .
commensurate - incommensurate transitions have been already discussed in the context of cold atoms in @xcite . in the problem we are considering , however ,
since two fields are becoming gapless at the same time , @xmath185 and @xmath172 , there are some differences@xcite with the standard case@xcite , in particular the exponents at the transition are non - universal . to conclude this section , we notice that we have found three types of phase transitions in the system we are considering .
we can have kosterlitz - thouless phase transitions as a function of interactions , where we go from a phase with locked superfluid phases between the bosons and the fermions at weak repulsion to a phase with decoupled bosons and fermions at strong repulsion .
we can have band - filling transitions as a function of the detuning between the phase in which atoms and molecule coexist and phases where only atoms or only molecules are present . finally , we can have commensurate - incommensurate transitions as a function of the strength of the magnetic field . in the following section , we discuss the correlation functions of superfluid and charge density wave order parameters in the phase in which molecules and atoms coexist with their relative superfluid phase @xmath185 locked . in the case of @xmath218 ,
the backscattering term ( [ eq : cdw - locking ] ) is non - vanishing .
this term induces a mutual locking of the densities of the bosons and the fermions@xcite and favors charge density wave fluctuations .
this term is competing with the josephson term ( [ eq : lambda - bosonized ] ) which tends to reduce density wave fluctuations .
for @xmath218 the relevant part of the hamiltonian given by the combination of the terms ( [ eq : cdw - locking ] ) and ( [ eq : lambda - bosonized ] ) reads : @xmath219 \cos \sqrt{2}\phi_\sigma\end{aligned}\ ] ] using a transformation @xmath220 , @xmath221 , and introducing the linear combinations @xmath222 and similar combinations for the dual fields , we can rewrite the interaction term ( [ eq : competing ] ) as : @xmath223 \cos \sqrt{2}\phi_\sigma\end{aligned}\ ] ] from this hamiltonian , it is immediate to see that a quantum ising phase transition occurs between the density wave phase @xmath224 and the superfluid phase @xmath225 at a critical point @xmath226.@xcite indeed , the field @xmath172 being locked , we can replace @xmath227 by its expectation value in eq .
( [ eq : quantum - ising ] ) , and rewrite ( [ eq : quantum - ising ] ) as a free massive majorana fermions hamiltonian.@xcite at the point @xmath228 , the mass of one of these majorana fermions vanishes giving a quantum critical point in the ising universality class@xcite .
on one side of the transition , when @xmath229 , the system is in the superfluid state discussed in sec .
[ sec : bf - coupling ] , on the other side @xmath230 , the charge density wave state discussed in @xcite is recovered . in order to better characterize the phase in which @xmath13 is relevant , we need to study the correlation function of the superfluid and the charge density wave operators .
let us begin by characterizing the superfluid order parameters .
first , let us consider the order parameter for bec of the molecules . as a result of the locking of the fields @xmath185 and @xmath172 , the boson operator eq .
( [ eq : boson - bosonized ] ) becomes at low energy : @xmath231 an order of magnitude of @xmath232 can be obtained from a scaling argument similar to the one giving the gap .
since the scaling dimension of the field @xmath233 is @xmath234 , and since the only lengthscale in the problem is @xmath235 , we must have @xmath236 .
similarly , the order parameter for s - wave superconductivity of the atoms @xmath237 becomes : @xmath238 thus indicating that the order parameters of the bec and the bcs superfluidity have become identical in the low energy limit@xcite .
this is the signature of the coherence between the atom and the molecular superfluids .
the boson correlator behaves as : @xmath239 as a result , the molecule momentum distribution becomes @xmath240 .
one can see that the tendency towards superfluidity is strongly enhanced since the divergence of @xmath241 for @xmath242 is increased by the coherence between the molecules and the atoms .
this boson momentum distribution can , in principle be measured in a condensate expansion experiment@xcite .
having seen that superfluidity is enhanced in the system , with bec and bcs order parameters becoming identical , let us turn to the density wave order parameters .
these order parameters are simply the staggered components of the atom and molecule density in eqs .
( [ eq : fermion - density - bosonized])([eq : boson - density - bosonized ] ) . in terms of @xmath243 , the staggered component of the molecule density is reexpressed as : @xmath244,\end{aligned}\ ] ] and the staggered component of the fermion density as : @xmath245,\end{aligned}\ ] ] where we have taken into account the long range ordering of @xmath172 .
we see that the correlations of both @xmath246 and @xmath247 decay exponentially due to the presence of the disorder field @xmath248 dual to @xmath185 . in more physical terms , the exponential decay of the density - wave correlations in the system results from the constant conversion of molecules into atoms and the reciprocal process which prevents the buildup of a well defined atomic fermi - surface or molecule pseudo - fermi surface .
the exponential decay of the density wave correlations in this system must be be contrasted with the power - law decay of these correlations in a system with only bosons or in a system of fermions with attractive interactions.@xcite in fact , if we consider that our new particles are created by the operator @xmath249 , we can derive an expression of the density operators of these new particles by considering the product @xmath250 . using the haldane expansion of the boson creation and annihilation operators@xcite ,
we can write this product as : @xmath251 \times e^{i\sqrt{2 } \theta_\rho}\left[\cos \sqrt{2}\phi_\sigma + \cos ( \sqrt{2 } \phi_\rho - 2 k_f x)\right ] \nonumber \\ & \sim & \langle e^{-i(\theta_b-\sqrt{2}\theta_\rho ) } \cos \sqrt{2 } \phi_\sigma \rangle \cos ( \sqrt{6 } \phi_+ -2 ( k_f+k_b ) x),\end{aligned}\ ] ] where @xmath252 can be interpreted as the pseudo fermi wavevector of composite bosons .
a scaling argument shows that the prefactor in the expression varies as a power of @xmath13 . as a result , when there is coherence between atoms and molecule , power - law correlations appear in the density - density correlator near the wavevector @xmath253 and the intensity of these correlations is proportional to the @xmath254 .
the resulting behavior of the fourier transform of the density - density correlator is represented on fig .
[ fig : chi - q ] . and @xmath255 .
a new peak at @xmath256 is obtained as a result of boson - fermion coherence .
, width=340 ] another interesting consequence of the existence of atom / molecule coherence is the possibility of having non - vanishing cross - correlations of the atom and the molecule density . in the three - dimensional case
such cross correlations have been studied in @xcite .
if we first consider cross - correlations @xmath257 we notice that due to the presence of different exponentials of @xmath184 in eqs .
( [ eq : boson - staggered - gap]) ( [ eq : fermion - staggered - gap ] ) , this correlator vanishes exactly .
therefore , no cross correlation exists between the staggered densities .
however , if we consider the cross correlations of the uniform densities , we note that since they can all be expressed as functions of @xmath258 , such cross correlations will be non - vanishing . more precisely ,
since : @xmath259 at low energy we have : @xmath260 .
in this section , we will obtain detailed expressions for these correlation functions at a special exactly solvable point of the parameter space . at this point , the kinks of the fields @xmath172 and @xmath185 become free massive fermions .
this property , and the equivalence of free massive fermions in 1d with the 2d non - critical ising model@xcite allows one to find exactly the correlation functions . as we have seen , after the rotation ( [ eq : rot ] ) , if we neglect the interaction terms of the form @xmath261 or @xmath262 , the hamiltonian of the massive modes @xmath263 can be rewritten as : @xmath264 \nonumber \\ & & + \int \frac{dx}{2\pi } \left [ u_{- } k_{- } ( \pi \pi_{-})^2 + \frac { u_{- } } { k_{-}}(\partial_x\phi_{-})^2\right ] \nonumber \\ & & + \frac{\lambda}{\sqrt{2\pi^3\alpha^3 } } \int dx \cos \sqrt{3}\theta_- \cos \sqrt{2}\phi_\sigma,\end{aligned}\ ] ] where @xmath157 . when the luttinger exponent is @xmath265 , it is convenient to introduce the fields : @xmath266 and rewrite the hamiltonian ( [ eq : massive ] ) as : @xmath267 \nonumber \\ & & + \int \frac{dx}{2\pi } \left [ u_{- } ( \pi \overline{\pi})^2 + { u_{- } } ( \partial_x\overline{\phi})^2\right ] \nonumber \\ & & + \frac{\lambda}{\sqrt{2\pi^3}\alpha } \int dx \cos \sqrt{2 } \overline{\phi% } \cos \sqrt{2}\phi_\sigma.\end{aligned}\ ] ] if we neglect the velocity difference , i.e. assume that @xmath268 , and introduce the pseudofermion fields : @xmath269}}{\sqrt{2\pi\alpha}},\end{aligned}\ ] ] we see immediately that the hamiltonian ( [ eq : massive - bar ] ) is the bosonized form of the following free fermion hamiltonian : @xmath270.\end{aligned}\ ] ] as a result , for the special value of @xmath265 , the excitations can be described as massive free fermions with dispersion @xmath271 , where @xmath272 .
this is known as luther - emery solution.@xcite one can see that the fermions carry a spin @xmath194 and a jump of the phase @xmath185 equal to @xmath273 .
therefore they can be identified to the kinks obtained in the semiclassical treatment of sec .
[ sec : bf - coupling ] . also , making all velocities equal and @xmath45 in ( [ eq : corresp - uk ] ) , we find the relation @xmath274 and thus the gap given by the rg varies as @xmath275 . for @xmath265
this expression reduces to the one given by the fermion mapping . to calculate the correlation functions it is convenient to introduce the fields : @xmath276 and reexpress operators by : @xmath277 this leads us to the following expression of the fermion density @xmath247 : @xmath278 } ( \psi^\dagger_{-,\uparrow } \psi^\dagger_{+,\downarrow } + \psi^\dagger_{+,\uparrow } \psi^\dagger_{-,\downarrow } ) + \text{h .
c. } , \end{aligned}\ ] ] which enables us to find exactly its correlations .
we introduce @xmath279 to simplify the notations .
the green s functions of the fermions read : @xmath280 using ( [ eq : rho ] ) and wick s theorem we find that in real space the density density correlations read : @xmath281.\end{aligned}\ ] ] these correlation functions decay exponentially , with a correlation length @xmath282 .
note that expression ( [ eq : fermion - le - realspace ] ) is exact . on the other side
the boson density is given by : @xmath283 } e^{i(\theta_\uparrow + \theta_\downarrow ) } + h. c.\end{aligned}\ ] ] to calculate the correlation functions in this case , we can use the equivalence of the dirac fermions in ( 1 + 1)d with the 2d non critical ising model@xcite to express the boson fields in terms of the order and disorder parameters of two non - critical ising models , @xmath284 respectively , by : @xmath285 we find that : @xmath286 where we have used @xmath287 and similarly for @xmath38 . in the bosonic case
, the mapping on the 2d ising model allows to calculate the correlation functions using the results of ref .
@xcite , where an exact expression of the correlation functions of the ising model in terms of painlev iii transcendants@xcite was derived .
in fact , since we are interested in the low - energy , long - distance properties of the system , it is enough to replace the painlev transcendants with an approximate expression in terms of modified bessel functions .
the resulting approximate expression is : @xmath288 knowing the correlation function in matsubara space allows us to obtain them in the reciprocal space via fourier transforms .
the fourier transform of the density - density response functions ( [ eq : fermion - le - realspace ] ) and ( [ eq : boson - le - realspace ] ) can be obtained in matsubara space from integrals derived in the appendix [ app : integral ] .
we find that the bosonic structure factor is : @xmath289 where @xmath290 is the gamma function and @xmath291 is a generalized hypergeometric function.@xcite of course , since eq .
( [ eq : boson - le - realspace ] ) is a long distance approximation , the expression ( [ eq : boson_structure_factor ] ) is also approximate .
more precisely , the exact expression possesses thresholds at higher frequencies associated with the excitation of more than one pair of kinks in the intermediate state .
however , the expression ( [ eq : boson_structure_factor ] ) is _ exact _ as long as @xmath292 is below the lowest of these thresholds . for the fermions ,
the expression of the structure factor is exact and reads : @xmath293\end{aligned}\ ] ] the response functions are then obtained by the substitution @xmath294 . since the generalized hypergeometric functions @xmath295 are analytic for @xmath296 @xcite , the imaginary part of the response functions is vanishing for @xmath297 . for @xmath298 , the behavior of the imaginary part is obtained from a theorem quoted in @xcite .
according to the theorem , @xmath299 provided that : @xmath300 is not an integer . therefore , if @xmath301 , the generalized hypergeometric function possesses power - law divergence as @xmath302 .
if @xmath303 , it has a cusp singularity . in our case , for the fermion problem , we have @xmath304 and in the boson problem @xmath305 therefore , for @xmath306 small , both the fermion and the boson density - density response functions show power law singularities for @xmath307 . for @xmath308 , the singularity in the boson density - density response
is replaced by a cusp .
this cusp disappears when @xmath309 .
for @xmath310 the singularity in the fermion density density correlator also disappears .
the behavior of the imaginary part of the boson density - density correlation function is shown in fig.[fig : imcorrfun ] .
an exact expression of the imaginary part of the @xmath311 function can be deduced from the calculations in @xcite .
indeed , in @xcite , it was found that : @xmath312 where @xmath313 is defined by a series that converges absolutely for @xmath314 and thus has no cut along @xmath315 $ ] , and @xmath316 is singular along @xmath315 $ ] .
@xmath316 can be expressed in terms of a higher hypergeometric function of two variables , the appell function @xmath317 , as : @xmath318 since only @xmath316 has a cut along @xmath319 $ ] , for @xmath320 , @xmath321 , the imaginary part can be expressed as a function of @xmath317 only .
this is particularly useful in the case of the fermion density correlators , because the expression ( [ eq : fermion_structure_factor ] ) is exact . in the case of the bosons , thresholds associated with the excitation of a larger number of majorana fermions
will appear at energies @xmath322 .
the imaginary parts of correlation functions eq .
( [ eq : boson_structure_factor]) ( [ eq : fermion_structure_factor ] ) can be measured by bragg spectroscopy@xcite . in fig.[fig : imcorrfun ] we plot the imaginary part of density correlation functions for the fermionic system with @xmath323 and @xmath324 as a function of frequency . in the first case , we obtain a divergence of the density - density response near the threshold , whereas in the second case , only a cusp is obtained . .
] we also wish to calculate the spectral functions of the original fermions @xmath325 ( not the pseudofermions @xmath326 ) . to obtain these spectral functions
, we express the operators @xmath327 as a function of the fields @xmath328 and their dual fields .
we find : @xmath329 } e^{-\frac i 6 \phi_\sigma } \\
\psi_{-,\sigma}(x)&=&\frac 1 { \sqrt{2\pi\alpha } } e^{\frac i { \sqrt{6 } } ( \theta_++\phi_+ ) } e^{i \left[\theta_{\sigma } + \frac 5 6 \phi_{\sigma}\right]% } e^{-\frac i 6 \phi_{-\sigma}}\end{aligned}\ ] ] therefore , the green s functions of the original fermions factorize as : @xmath330 where : @xmath331 @xmath332(x,\tau ) } e^{i \left[\theta_{\sigma } -\frac 5 6 \phi_{\sigma}\right](0,0)}\rangle,\end{aligned}\ ] ] and : @xmath333 the correlator in eq .
( [ eq : for - bernard ] ) satisfies differential equations that were derived in @xcite .
however , since the fields @xmath334 are long range ordered , we can simply replace the terms @xmath335 by their expectation value @xmath336 .
this approximation only affects the behavior of the fermion correlator at high energy .
therefore , we are left with ( [ eq : for - lz ] ) to evaluate . to do this , we can use exact results derived in @xcite to obtain a form - factor expansion of eq .
( [ eq : for - lz ] ) . the first term of the form factor expansion yields : @xmath337 writing : @xmath338 we finally obtain that : @xmath339 the fourier transform of @xmath340 is given by a weber - schaefheitlin integral@xcite . in final form ,
the fourier transform of the fermion green s function reads : @xmath341 when this green s function is analytically continued to real frequency , it is seen@xcite that it has a power law singularity for @xmath342 , and is analytic for @xmath292 below this threshold . as a result
, the fermion green s function vanishes below the gap , as it would do in a superconductor@xcite and could be checked experimentally .
note however that the anomalous green s function remains zero@xcite , due to the fluctuations of the phase @xmath343 .
till now , we have only considered the case of the continuum system ( [ eq : nolattice ] ) or incommensurate filling in the lattice system ( [ eq : lattice ] ) .
we now turn to a lattice system at commensurate filling .
we first establish a generalization of the lieb - schultz - mattis theorem@xcite in the case of the boson - fermion mixture described by the hamiltonian ( [ eq : lattice ] ) .
this will give us a condition for the existence of a mott insulating state without spontaneous breakdown of translational invariance .
then , we will discuss using bosonization the properties of the mott state .
we note that mott states have been studied in the boson - fermion model in refs .
@xcite , but not in a one - dimensional case .
finally , we will consider the case when the molecules or the atoms can form a mott insulating case in the absence of boson - fermion conversion , and we will show that this mott state is unstable .
a generalized lieb - schultz - mattis theorem can be proven for the boson fermion mixture described by the lattice hamiltonian ( [ eq : lattice])@xcite .
let us introduce the operator : @xmath344,\end{aligned}\ ] ] such that @xmath345 .
following the arguments in ref.@xcite , one has : @xmath346 where @xmath347 is the ground state of the system , @xmath348 is the translation operator and @xmath349 is the full hamiltonian .
the quantity @xmath0 is defined by : @xmath350 for noninteger @xmath0 , it results from the analysis of @xcite that there is a state @xmath351 of momentum @xmath352 $ ] which is orthogonal to the ground state @xmath347 and is only @xmath353 above the ground state .
this implies either a ground state degeneracy ( associated with a spontaneous breaking of translational symmetry ) or the existence of gapless excitations ( if the spontaneous translational symmetry is unbroken and the ground state is unique ) . for integer @xmath0 , the ground state and the state @xmath351 have the same momentum . in that case
, a gapped state without degeneracy can be obtained .
this state is analogous to the mott insulating state in the half - filled hubbard model in one - dimension@xcite or the mott insulating state in the bose - hubbard model with one boson per site@xcite .
we note that for @xmath354 in the hamiltonian ( [ eq : lattice ] ) fermions and bosons are separately conserved , and the corresponding fermi wavevectors are : @xmath355 the momentum of the state @xmath351 is thus equal to @xmath356 .
the condition to have a mott insulating state in the hubbard model , @xmath357 is thus generalized to @xmath358 i.e. @xmath359 . in this section , we provide a derivation of the umklapp term valid in the case of the lattice system ( [ eq : lattice ] ) .
let us consider the @xmath360 and @xmath361 components of the atom and molecule charge density , given respectively by eq .
( [ eq : fermion - density - bosonized ] ) and eq .
( [ eq : boson - density - bosonized ] ) .
these terms yield an interaction of the form : @xmath362\cos \sqrt{2 } \phi_\sigma \nonumber \\ & & + \frac c 2 \int \cos [ 2\phi_b -\sqrt{2}\phi_\rho -2(k_b - k_f ) x]\cos \sqrt{2 } \phi_\sigma\end{aligned}\ ] ] in eq .
( [ eq : comm - repuls ] ) , the last line is the backscattering term of ( [ eq : cdw - locking ] ) , and the second line is the umklapp term . let us consider a case with @xmath152 , and let us concentrate on the effect of the umklapp term . using the rotation ( [ eq : rot ] ) , we can reexpress it as : @xmath363\cos \sqrt{2 } \phi_\sigma \nonumber \\.\end{aligned}\ ] ] in the following we will consider the cases corresponding to one or two atoms per site .
let us consider first the case of @xmath364 .
then , the term ( [ eq : umk - source ] ) is oscillating . in second order perturbation theory
, it gives rise to the umklapp term : @xmath365 the condition for the appearance of the umklapp term ( [ eq : umklapp ] ) can be seen to correspond to having one fermion atom per site of the atomic lattice .
let us briefly mention two alternative derivations of ( [ eq : umklapp ] ) .
a simple derivation can be obtained by considering the combination of the @xmath366 term in the boson density with the @xmath367 term in the fermion density in haldane s expansion@xcite .
a second derivation can be obtained by considering the effect of a translation by one lattice parameter on the phases @xmath368 and @xmath369 @xcite .
the expressions of the densities ( [ eq : fermion - density - bosonized ] ) and ( [ eq : boson - density - bosonized ] ) imply that upon a translation by a single site @xmath370 and @xmath371 .
therefore , the combination @xmath372 transforms as : @xmath373 . for @xmath374 , the term @xmath375 is invariant upon translation , thus leading again to ( [ eq : umklapp ] ) .
the presence of the umklapp term ( [ eq : umklapp ] ) in the hamiltonian can result in the opening of a charge gap and the formation of a mott insulating state .
since the umklapp term is of dimension @xmath376 this implies that a mott insulating state is possible only for @xmath377 i.e. very strong repulsion . for free fermions ,
the mott transition would occur at @xmath130 i.e. for weakly repulsive interaction .
thus , we see that the josephson coupling is very effective in destabilizing the mott state . in the mott insulating state , the superfluid fluctuations become short ranged .
since cdw fluctuations are also suppressed , the system shows some analogy with the haldane gapped phase of spin-1 chains@xcite in that it is totally quantum disordered .
in fact , this analogy can be strengthened by exhibiting an analog of the vbs ( valence bond solid ) order parameter@xcite . in haldane
gapped chains , this nonlocal order parameter measures a hidden long range order in the system associated with the breakdown of a hidden discrete symmetry in the system . the equivalent nonlocal order parameter for the atom - molecule system is discussed in appendix [ app : non_loc_ord ] .
another commensurate filling , where a mott insulating state is possible is obtained for @xmath378 .
this case corresponds to having one molecule ( or two atoms ) per site of the optical lattice . in that case , the term in ( [ eq : umk - source ] ) is non - oscillating , and it gives rise to an umklapp term of the form : @xmath379 we notice that this umklapp term is compatible with the spin gap induced by the josephson term ( [ eq : lambda - bosonized ] ) .
when the josephson coupling is large , we can make @xmath380 and we see that the term ( [ eq : umklapp - boson ] ) becomes relevant for @xmath381 . for weaker josephson coupling ,
the dimension becomes @xmath382 , and this term is relevant only for @xmath383 .
since @xmath384 corresponds to hard core bosons , this means that for weak josephson coupling , the mott state with a single boson per site becomes trivial .
interestingly , we note that increasing the josephson coupling is _ enhancing _ the tendency of the system to enter a mott insulating state as a result of the formation of a spin gap .
if we compare with a system of bosons at commensurate filling , we note however that the mott transition would obtain for @xmath385@xcite . therefore
, the josephson coupling still appears to weaken the tendency to form a mott insulating state .
such tendency was also observed in @xcite
. when the atomic subsystem is at commensurate filling ( @xmath386 ) , an umklapp term : @xmath387 must be added to the hamiltonian .
such umklapp term can create a gap in the density excitations of the unpaired atoms .
however , we must also take into account the term ( [ eq : lambda - bosonized ] ) .
this term is ordering @xmath185 and thus competes with the umklapp term ( [ eq : umklapp - fermions ] ) . to understand what happens when @xmath185 is locked , it is convenient to rewrite the umklapp term ( [ eq : umklapp - fermions ] ) as @xmath388 .
the terms generated by the renormalization group are of the form @xmath389 , with @xmath390 an integer . when @xmath185 is locked , replacing the terms @xmath391 by their expectation values , we find that all these terms vanish . therefore , no term @xmath392 can appear in the low energy hamiltonian .
a more formal justification of the absence of the @xmath393 term in the low energy hamiltonian can be given by noting that when the hamiltonian is expressed in terms of @xmath176 it has a continuous symmetry @xmath394 and @xmath395 .
as a result , terms of the form @xmath396 are forbidden by such symmetry .
the consequence of the absence of @xmath392 terms in the hamiltonian when @xmath185 is locked is that , even if the unpaired atom density is at a commensurate filling , the umklapp terms do not destabilize the coupled phase .
however , in the opposite case of a strong umklapp term and a weak boson - fermion conversion term , it is the field @xmath368 that will be ordered .
the previous arguments can be reversed and show that the formation of a mott gap for the fermions will prevent the formation of the coupled phase . using the method of ref.@xcite , one can show that the phase transition between the coupled and the decoupled state is identical to the phase transition that occurs in two non - equivalent coupled two - dimensional xy models .
this phase transition was studied by the renormalization group in @xcite .
it was found that in the case of interest to us , this phase transition was in the ising universality class .
thus , one expects a quantum ising phase transition between the state where the fermions are decoupled from the bosons and form a mott insulator and the state where the fermions and bosons are coupled and form a superfluid . of course , the same arguments can also be applied to the bosons at commensurate filling , the role of the fields @xmath369 and @xmath368 being simply reversed .
to connect experiments in quasi - one - dimensional confining waveguides with theoretical models in 1d , it is necessary to obtain estimates of the parameters that enter the hamiltonians ( [ eq : lattice]),([eq : nolattice ] ) and the bosonized hamiltonian ( [ eq : bosonized - spin]),([eq : lambda - bosonized ] ) and ( [ eq : detun - bosonized ] ) . since the parameters in the hamiltonian ( [ eq : lattice ] ) depend on the periodic optical trapping potential , we will mainly focus on the parameters that enter in the continuum hamiltonian eq .
( [ eq : nolattice ] ) and in the bosonized hamiltonian , i.e. the luttinger exponent @xmath137 , the velocity , and the fermion - boson coupling @xmath13 . before giving an estimate of the parameters we need first to remind that , at the two - body level
, there is a connection between the 1d boson - fermion model and the quasi-1d single channel model@xcite , thus we will use one or the other depending on the physical parameter wherein we are interested .
experimentally , molecules have been formed from fermionic atoms @xmath397li@xcite and @xmath4k@xcite . for @xmath397li ,
the mass is roughly 6 times the mass of the proton , @xmath398 , and for @xmath4k it is @xmath399 .
then , we need to determine the effective interaction of atoms under cylindrical confinement . for the interaction , we will assume a contact form , i.e. @xmath400 and the hamiltonian describing the confined atoms reads : @xmath401 where @xmath402 , the second term represents the harmonic confinement potential , and : @xmath403 is the atom - atom repulsion .
the coupling constant @xmath404 is expressed as a function of the atom - atom scattering length @xmath405 as@xcite @xmath406 we introduce the following decomposition of the fermion annihilation operator : @xmath407 where @xmath408 , and the eigenstates of the transverse hamiltonian @xmath409 satisfy the schrdinger equation : @xmath410 and the orthonormal condition @xmath411 .
we will assume that we have a very elongated trap , with @xmath412 so that we can neglect the longitudinal confinement .
the first line of the hamiltonian ( [ eq : atatham ] ) can then be rewritten as : @xmath413\end{aligned}\ ] ] if the transverse zero - point energy is much higher than the interaction energy per atom , the transverse motion is frozen in the ground state .
it is known that virtual transitions to higher states can give rise to a divergence in 1d scattering length known as confinement induced resonance ( cir ) which is a kind of fano - feshbach resonance@xcite .
we will ignore cir for the moment and restrict ourselves to consider the lowest energy level @xmath414 for the transverse motion .
the interaction in ( [ eq : atatham ] ) can be rewritten as : @xmath415 where the effective potential @xmath416 reads : @xmath417 using the expression of the ground state wave function of the transverse motion : @xmath418 substituting it into ( [ eq : ueff ] ) , using the definition of the interaction ( [ eq:3dcontact ] ) , and integrating over the transverse coordinate , we obtain @xmath419 with the effective one - dimensional coupling@xcite : @xmath420 eq .
( [ eq : effcoup ] ) can be generalized to the case in which atoms of different species , with different trapping frequencies @xmath421 and @xmath422 are interacting with each other . in that case , @xmath423 where @xmath424 is the atom - atom scattering length .
knowing @xmath425 , we can obtain the luttinger exponent of fermions in ( [ eq : bosonized - spin ] ) as : @xmath426 in the case of bosons , the luttinger exponent @xmath136 must be extracted from the lieb - liniger equations@xcite .
having obtained the form of the effective interactions , we turn to the determination of the josephson coupling @xmath13 in the boson - fermion model ( [ eq : nolattice ] ) . in the 3d case@xcite , the boson - fermion conversion factor is given by : @xmath427 with : @xmath428 where @xmath429 is the atom - atom scattering length far from resonance , @xmath430 is the width of the resonance and @xmath431 is the difference of magnetic moment between atom and molecule . using the projection on the lowest level ,
we obtain : @xmath432 knowing the interaction @xmath13 in terms of the relevant physical parameters , we finally have to determine the spatial cutoff @xmath140 to use in the bosonization and comment on the validity of bosonization approach . in the case of the optical lattice eq .
( [ eq : lattice ] ) , the obvious spatial cutoff is the lattice spacing .
the cutoff to use in bosonization for the continuum case of eq .
( [ eq : nolattice ] ) is obtained in the following way .
bosonization is applicable as long as the kinetic energy of longitudinal motion of the particles is much smaller than the trapping energy @xmath433 .
thus , we have to impose the condition : @xmath434 , where @xmath435 is the momentum cutoff@xcite .
the real space cutoff in the continuum case is thus @xmath436 .
the condition for perturbation theory to be valid is that the energy associated with the formation of molecules , @xmath437 is small with respect to the energy cutoff @xmath438 .
therefore , perturbation theory is applicable when : @xmath439 i.e. @xmath440 using the values given in refs .
@xcite , we find that this parameter is small for @xmath441 m/s . since @xmath442 can be expected to be of the order of @xmath443 m/s , this is not unreasonable . in fact , using the values of the trapping frequency given by moritz et al.@xcite we find that : @xmath444 therefore , we see that with @xmath397li at the narrow resonance , the ratio is of order @xmath445 and we can expect that our theory is valid qualitatively . concerning @xmath137 , we find that : @xmath446 i.e. interactions between fermions can be neglected .
since the interaction between the molecules @xcite has a scattering length @xmath447 one sees that molecules are only weakly interacting . as we have seen in sec .
[ sec : expt - no - pot ] , in the case of a two - dimensional optical lattice without periodic potential along the tubes , the repulsion between the bosonic molecules is weak , making the decoupling transition or the mott transition impossible to observe . to increase the effect of the repulsion
, one needs to increase the effective mass of the atoms by adding a periodic potential along the tubes .
a periodic potential can be imposed along the tubes by placing the atoms in a three dimensional optical lattice .
the atoms experience a potential : @xmath448 where @xmath449 is the wavelength of the laser radiation , and @xmath450 so that the system remains quasi one - dimensional .
the strength of the potential is measured in unit of the recoil energy @xmath451 as @xmath452 .
typical values for @xmath453 are in the range @xmath454 to @xmath455 . for lithium atoms@xcite ,
the typical value of @xmath456 is @xmath457khz .
if the potential is sufficiently strong , the atoms tend to localize in the lowest trap states near the minima of this potential . in our case , since the periodic potential along the tubes has shallower minima than in the transverse directions , the small overlap between the trap states in the longitudinal direction yields the single band hamiltonian ( [ eq : lattice])@xcite .
an expression of the parameters of the lattice model ( [ eq : lattice ] ) in terms of the microscopic parameters has been derived in@xcite . on the lattice
, the fermi velocity of the atoms and the pseudo - fermi velocity of the molecules can be reduced by increasing the depth of the periodic potential in the longitudinal direction .
this allows in principle to move the system near the decoupling transition@xcite or the mott transition , by reducing @xmath137 .
a second possible setup@xcite is to use a cigar shaped potential : @xmath458 with @xmath459 , so that the atoms and the molecules are strongly confined in the transverse direction , and to apply a periodic potential : @xmath460 in order to form the one dimensional structure described by the model ( [ eq : lattice ] ) .
the main difficulty of experiments in optical lattices is that the reduction of the bandwidth results in a reduction of the fermi velocity @xmath442 .
since the perturbative regime is defined by @xmath461 , this implies that by increasing the depth of the potential in the longitudinal direction one is also pushing the system in the regime where the boson - fermion conversion term must be treated non - perturbatively@xcite .
however , in that regime there is nt anymore coexistence of atoms and molecules and the decoupling transition does not exist .
moreover , in that regime , the mott transition becomes the usual purely fermionic or purely bosonic mott transition.@xcite
we have studied a one - dimensional version of the boson - fermion model using the bosonization technique .
we have found that at low - energy the system is described by two josephson coupled luttinger liquids corresponding to the paired atomic and molecular superfluids . due to the relevance of the josephson coupling for not too strong repulsion ,
the order parameters for the bose condensation and fermion superfluidity become identical , while a spin gap and a gap against the formation of phase slips are formed . as a result of these gaps , we have found that the charge density wave correlations decay exponentially , differently from the phases where only bosons or only fermions are present@xcite .
we have discussed the application of a magnetic field that results in a loss of coherence between the bosons and the fermion and the disappearance of the gap , while changing the detuning has no effect on the existence of the gaps until either the fermion or the boson density is reduced to zero .
we have discussed the effect of a backscattering term which induces mutual locking of the density of bosons and fermions favoring charge density wave fluctuations resulting in a quantum ising phase transition between the density wave phase and the superfluid phase .
we have found a luther - emery point where the phase slips and the spin excitations can be described in terms of pseudofermions . for this special point in the parameter space , we have derived closed form expressions of the density - density correlations and the spectral functions .
the spectral functions of the fermions are gapped , whereas the spectral functions of the bosons remain gapless but with an enhanced divergence for momentum close to zero .
finally , we have discussed the formation of a mott insulating state in a periodic potential at commensurate filling .
we have first established a generalization of the lieb - schulz - mattis theorem , giving the condition for the existence of a mott - insulating state without spontaneous breakdown of translational invariance .
then , we have discussed the properties of the mott - state in the case of one atom or two atoms per site showing that in the first case the josephson coupling is very effective in destabilizing the mott state .
finally , we have considered the case when the atoms or the molecules can form a mott state in absence of boson - fermion conversion and shown that this mott - state is unstable . to connect our results with experiments in quasi - one - dimensional confining waveguides we have derived estimates of the parameters that enter the bosonized hamiltonian , as the luttinger exponents , using the values of the trapping frequency and density used in experiments .
we have seen that bosons are only weakly interacting and the necessary small fermionic luttinger parameter required to realize a strongly interacting system , render the mott insulating and decoupled phases difficult to observe in experiments .
a nontrivial challenge is the experimental realization of the coupled luttinger liquids phase with parameters tunable through the exactly solvable point ( the luther - emery point ) .
we suggest that a fano - feshbach resonantly interacting atomic gas confined in a highly anisotropic ( 1d ) trap and subject to a periodic optical potential is a promising candidate for an experimental measurement of the physical quantities ( correlation functions ) discussed here .
finally we would like to comment on the fact that an interesting edge states physics is expected when open boundary conditions ( or a cut one - dimensional boson - fermion system ) are considered .
the existence of edge states at the end of the system could lead to significant contribution to the density profile that could be tested in experiments .
the physics of the edge states will be similar to the one of haldane gap systems , like the valence bond solid model , and a study along this direction is in progress .
in this appendix we will derive a slightly more general integral than those of eqs .
( [ eq : boson_structure_factor ] ) and ( [ eq : fermion_structure_factor ] ) .
namely , we will consider : @xmath462 to find ( [ eq : integral - general ] ) explicitly , we use the series expansion of the bessel function @xmath463 from @xcite [ eq .
( 9.1.10 ) ] .
we find that @xmath464 the integral that appears in the expansion in powers of @xmath465 is a well known weber - schaefheitlin integral@xcite with two modified bessel functions .
its expression is : @xmath466 the resulting expression of @xmath467 can be rearranged using the duplication formula for the gamma function , eq .
( 6.1.18 ) in @xcite .
the final expression of @xmath468 is : @xmath469 this series expansion is readily identified with the definition of the generalized hypergeometric function @xmath470 given in @xcite .
so we find finally that : @xmath471 for @xmath472 , the function @xmath470 reduces to a simpler @xmath311 function .
this leads to eqs .
( [ eq : boson_structure_factor ] ) and ( [ eq : fermion_structure_factor ] ) .
the mott insulating state can be characterized by the expectation value of a non - local order parameter as the haldane gap state in a spin-1 chain@xcite .
the non - local order parameter is defined as follows : @xmath473 the string operator : @xmath474 is a product of exponentials . as a result of its definition , we see that it is counting the number of fermions located between the sites @xmath475 and @xmath476 , either unbound or forming a molecule . to derive a bosonized expression of this operator ,
we notice that @xmath477 since @xmath478 has only integer eigenvalues and rewrite the string operator as : @xmath479 using ( [ eq : boson - staggered - gap ] ) and ( [ eq : string - bosonized ] ) we obtain the nonlocal order parameter ( [ eq : vbs ] ) as : @xmath481 in the mott insulating state with one fermion per site , we have @xmath482 where @xmath390 is an integer . taking @xmath483 , we see that the expectation value of the order parameter is non - vanishing in the mott state . in bosonized form , we have : @xmath485 and again this order parameter is non - vanishing .
the physical interpretation of the non - zero expectation value of these nonlocal order parameters is that both bosons and fermions possess a hidden charge density wave order in the mott insulator .
this charge density wave is hidden as a result of the fluctuation of the density of fermions and the density of bosons . | we study 1d fermions with photoassociation or with a narrow fano - feshbach resonance described by the boson - fermion resonance model . using the bosonization technique ,
we derive a low - energy hamiltonian of the system .
we show that at low energy , the order parameters for the bose condensation and fermion superfluidity become identical , while a spin gap and a gap against the formation of phase slips are formed . as a result of these gaps , charge density wave correlations decay exponentially in contrast with the phases where only bosons or only fermions are present .
we find a luther - emery point where the phase slips and the spin excitations can be described in terms of pseudofermions .
this allows us to provide closed form expressions of the density - density correlations and the spectral functions .
the spectral functions of the fermions are gapped , whereas the spectral functions of the bosons remain gapless .
the application of a magnetic field results in a loss of coherence between the bosons and the fermion and the disappearance of the gap . changing the detuning has no effect on the gap until either the fermion or the boson density is reduced to zero .
finally , we discuss the formation of a mott insulating state in a periodic potential .
the relevance of our results for experiments with ultracold atomic gases subject to one - dimensional confinement is also discussed . |
neutrino masses are one of the main motivations we have at present for physics beyond the standard model ( sm ) .
although a minimal extension of the sm by adding three right - handed neutrinos with dirac mass terms is still allowed by all neutrino data , this is not the preferred scenario for it does not explain the smallness of neutrino masses or why lepton number should exactly be conserved .
scenarios in which the smallness of neutrino masses is linked to the non - conservation of lepton number are usually considered more natural .
the simplest versions of these scenarios are realized at tree level and are known under the name of the seesaw mechanisms .
seesaws of type i @xcite and iii @xcite extend the sm with new fermions , singlet and triplet respectively , with a majorana mass term , @xmath2 , which breaks explicitly lepton number . on the other hand ,
seesaw of type ii @xcite adds only a scalar triplet with hypercharge 2 . in this case ,
lepton number is broken explicitly in the scalar potential by a trilinear coupling , @xmath3 ( see also @xcite for a variation with lepton number broken spontaneously ) .
seesaws of type i and iii give neutrino masses of order @xmath4 with @xmath5 being the yukawa coupling and @xmath6 the vacuum expectation value ( vev ) of the sm doublet . then
, if @xmath7 , masses can naturally be small .
however , the same parameters that appear in the neutrino masses also appear in the mixing of new fermions with ordinary fermions , which is proportional to @xmath8 and must be very small .
this makes these types of models very difficult to test .
seesaw of type ii , however , generates neutrino masses of order @xmath9 , where now @xmath5 is the yukawa coupling of the scalar triplet to lepton doublets and @xmath2 the triplet mass .
the trilinear coupling @xmath3 is protected by symmetry and can be naturally small .
this allows for small neutrino masses compatible with a rich phenomenology ( the new scalars of the model could be produced at the lhc @xcite and there could be lepton flavor violating processes like @xmath10 or @xmath11 @xcite ) , raising the possibility to test the mechanism of neutrino masses in non - oscillation experiments . in an effort to fixate the scalar spectrum of the model , in this article we revisit the scalar potential of the type ii seesaw model taking into account unitarity and the stability of the potential
moreover , on the phenomenological side , we will also consider the constraints coming from the oblique @xmath1-parameter and higgs decay branching ratios , in particular @xmath12 , which are largely independent on the other sectors of the theory ( yukawa couplings and neutrino masses ) .
in fact , the one - loop @xmath1-parameter only depends , to a good approximation , on the mass splitting between the scalars , while the new contributions to @xmath12 can be expressed in terms of the scalar masses and the mixing angle in the neutral scalar sector .
a very important ingredient of our analysis is the requirement that the triplet scalar vev , @xmath13 , is much smaller than the electroweak scale .
this is , in part , required by the tree - level @xmath14-parameter and in part , by the requirement of small neutrino masses .
it has been noticed , in this case , that the some relations involving the scalar masses can be obtained and that the @xmath1-parameter constrains the splitting to be less that about @xmath15 gev ( see for instance @xcite ) .
we will re - derive these constraints by using perturbative unitarity and the @xmath1-parameter .
after this , the spectrum of scalar masses is practically fixed up to a global scale and a small splitting between masses and can be expressed in terms of three parameters ( apart from the known parameters , the higgs mass and the electroweak scale , and the triplet vev which becomes irrelevant for @xmath16 gev ) .
then , we will derive a lower bound on the global scale by adding present data on @xmath12 .
similar analyses have been performed several times in the literature @xcite .
but we differ from those in the sense that we express our results in terms of the experimentally measurable quantities . in the process
, we have been able to unravel some features which , we believe , have not been emphasized before . additionally we also point out that the current measurement of the diphoton signal strength enables us to put lower limits on the nonstandard scalar masses , which are competitive or , in some cases , better than the direct search limits . in section [ s :
potential ] we study the scalar potential , the scalar mass spectrum and the constraints from unitarity and stability . in section [ s : analysis ] we perform a complete numerical analysis by including also the constraints from the @xmath14-parameter . in section [ s : hgg ]
we study the impact of the data on @xmath12 and obtain the lower bound on the masses .
finally in section [ s : conclusions ] we present our conclusions .
the type ii seesaw model extends the higgs sector of the standard model by adding one scalar @xmath17 triplet ( @xmath18 ) with hypercharge , @xmath19 .
the most general scalar potential involving this triplet and the standard @xmath17 doublet , @xmath20 , is given by@xcite @xmath21 + \lambda_4\left(\phi^\dagger\delta \delta^\dagger\phi \right ) \,,\end{aligned}\ ] ] where ` @xmath22 ' represents the trace over @xmath23 matrices and @xmath24 is the second pauli matrix .
we can take all the parameters in the potential to be real without any loss of generality@xcite . denoting by @xmath6 ( @xmath13 ) the vev of the doublet ( triplet ) scalar field , the minimization conditions read
[ e : min ] @xmath25 after the spontaneous symmetry breaking , we represent the scalar multiplets in the following way : @xmath26 and the electroweak vev is then given by , @xmath27 from the observed value of the electroweak @xmath14-parameter , @xmath13 is expected to be @xmath28 or less@xcite .
using we first trade @xmath29 and @xmath30 for @xmath6 and @xmath13 in the potential of .
the fields , @xmath31 , represent the doubly charged scalar with mass @xmath32 the mass squared matrix in the singly charged sector can be rotated to the physical basis through @xmath33 to obtain the charged goldstone ( @xmath34 ) along with a singly charged scalar ( @xmath35 ) with mass @xmath36 the assumption of real vevs allows us to define electrically neutral mass eigenstates which are also eigenstates of @xmath37 .
the mass squared matrix in the @xmath37-odd sector can be rotated to the physical basis through @xmath38 to obtain the neutral goldstone ( @xmath39 ) along with a pseudoscalar ( @xmath40 ) with mass @xmath41 finally , for the @xmath37-even part we have : @xmath42 we can obtain the physical eigenstates through the following rotation : @xmath43 [ [ sm - like - limit ] ] sm like limit : + + + + + + + + + + + + + + even with @xmath44 , to make the tree level couplings of the lightest @xmath37-even scalar ( @xmath45 ) close to those in the sm , we need to set @xmath46 . from we see that there are two different ways to obtain this limit as discussed below : [ [ casei ] ] case i : the first option is to invoke a fine tuning so that the numerator in the expression for @xmath47 vanishes .
the condition reads : [ e : case1 ] @xmath48 note that , in this case , we can make the tree level couplings of @xmath45 to be close to sm _ without _ demanding @xmath13 to be exactly zero .
this limit is interesting in view of the fact that we can still have small neutrino masses through a small value for @xmath13 and at the same time be very close to the sm .
but later we will show that the charged scalars , in this limit , can contribute substantially in the diphoton decay amplitude and therefore this limit is _ ruled out_. [ [ caseii ] ] case ii : in a second and more conventional approach , the sm - like limit is obtained as a direct consequence of @xmath13 being arbitrarily small . to be more precise , in the limit and @xmath49
, we can approximate as @xmath50 in this case the expression for @xmath47 in can be simplified into @xmath51 as we will show later , this limit corresponds to the decoupling of heavy nonstandard scalars .
it is now instructive to count the number of free parameters in the scalar potential .
note that , contains eight free parameters .
as mentioned before , @xmath29 and @xmath30 can be traded in favor of @xmath6 and @xmath13 .
the remaining six parameters can be exchanged for five physical masses and the mixing angle , @xmath52 , as follows@xcite : [ e : inverted ] @xmath53 \ , , \\
\lambda_3 & = & \frac{1}{v_t^2}\left [ \frac{2v_d^2m_+^2}{v_d^2 + 2v_t^2 } - \frac{v_d^2m_a^2}{v_d^2 + 4v_t^2}- m_{++}^2\right ] \
\lambda_4 & = & \frac{4m_a^2}{v_d^2 + 4v_t^2 } - \frac{4m_+^2}{v_d^2 + 2v_t^2 } \,.\end{aligned}\ ] ] among the eight redefined parameters that appear on the rhs of , not all are unknown .
we already know @xmath54 gev and under the assumption that the lightest cp - even higgs is what has been found at the lhc , @xmath55 gev is also known .
the compatibility of higgs signal strengths into different decay channels with their corresponding sm expectations tells us to focus near the sm - like limit , @xmath56 .
we will see how the smallness of @xmath13 in association with unitarity and stability entail strong correlations among the remaining four nonstandard masses , @xmath57 , making the scalar potential of type ii seesaw model constrained very strongly .
we need to ensure that there is no direction in the field space along which the potential becomes infinitely negative .
the conditions for the potential of to be bounded from below read@xcite [ e : stability ] @xmath58 \,.\end{aligned}\ ] ] the @xmath59-matrix eigenvalues that will be constrained from unitarity of the scattering amplitudes are also listed below@xcite : [ e : unieigen ] @xmath60
as a first level of simplification , we can use into the inequality ( [ e : u10 ] ) and remembering @xmath61 , we may write @xmath62 therefore , for @xmath63 , we can very well approximate @xmath64 this automatically implies @xmath65 . similar considerations for the inequalities ( [ e : u8 ] ) or ( [ e : u9 ] ) enable us to express the doubly charged scalar mass as follows : @xmath66 we will exemplify our numerical results by setting @xmath67 gev .
we also take @xmath68 gev and @xmath69 gev as input parameters
. then we perform random scan over the @xmath70 space by varying the parameters within the following ranges : @xmath71 \ , , ~~ m_h \in [ 125 , 2000]~{\rm gev}\ , , ~~ m_{+}\in
[ 0,2000]~{\rm gev}\,,\end{aligned}\ ] ] and calculate @xmath72 and @xmath73 through the relations ( [ e : ma1 ] ) and ( [ e : mcc1 ] ) and @xmath74 will cause negligible deviations from .
we have taken this effect into account in our numerical analysis . ] . in anticipation that the higgs data will continue to agree with the sm with increasing accuracy in the upcoming runs of the lhc , we vary @xmath75 in a rather narrow range around @xmath46 .
next we compute the @xmath76s using and check whether the unitarity and stability conditions , given in , are satisfied .
note that , since we are scanning in terms of the physical parameters , positivity of the masses are guaranteed and therefore , we are in a local minimum .
moreover , we have explicitly checked that @xmath77 for every point in our scan .
we discuss below the results of our analysis .
-@xmath78 plane from unitarity and stability for @xmath67 gev .
the continuous lines are contours for @xmath73 drawn using for @xmath56 .
the allowed region from @xmath1-parameter has also been shaded assuming @xmath79.__,width=302,height=226 ] -@xmath78 plane from unitarity and stability for @xmath67 gev .
the continuous lines are contours for @xmath73 drawn using for @xmath56 .
the allowed region from @xmath1-parameter has also been shaded assuming @xmath79.__,width=302,height=226 ] from fig .
[ f : samh ] we see that a heavy @xmath80 requires @xmath81 , _
i.e. _ , we are automatically pushed towards a sm - like situation for small @xmath13 and heavy nonstandard scalars .
however a closer inspection of fig .
[ f : samh ] reveals that the peak does not occur precisely at @xmath82 but at a small non negative value of @xmath75 .
the reason for this can be understood from the unitarity condition ( [ e : u7 ] ) which , in terms of the physical masses reads @xmath83 where , we have used to substitute for @xmath72 .
one interesting thing to note from above inequality is that , when @xmath84 , the value of @xmath80 is bounded from above in the sm - like limit , @xmath85 .
but what is more interesting is to find that , for large @xmath80 and small @xmath75 , the expression on the lhs of ( [ e : peak ] ) vanishes for @xmath86 and so the inequality is trivially satisfied .
thus the location of the peak on the horizontal axis of fig .
[ f : samh ] is a direct reflection of a small value of @xmath13 .
smaller the value for @xmath13 , closer is the peak to @xmath46 . strictly speaking , the peak in fig .
[ f : samh ] does not extend up to infinity along the vertical axis .
this is because the unitarity condition ( [ e : u3 ] ) which in terms of physical masses reads @xmath87 always puts an upper bound on @xmath80 for nonzero @xmath75 .
but for @xmath88 the peak of fig .
[ f : samh ] at @xmath46 which corresponds to the decoupling limit as defined in and in this case , the bound from ( [ e : peak ] ) can be alleviated and infinitely heavy nonstandard scalars can be allowed .
[ f : mhmc ] depicts that the splitting between @xmath80 and @xmath78 can be restricted from unitarity and stability . in figs .
[ f : samh ] and [ f : mhmc ] we have also drawn the contours of @xmath72 and @xmath73 to emphasize that due to the correlation between different parameters , experimental bound on any of the nonstandard masses can be translated into indirect bounds on the other masses too . to illustrate , if we can rule out a doubly charged scalar below 500 gev from direct searches at the lhc , then , from fig .
[ f : mhmc ] , we also forbid a singly charged scalar below 350 gev . note that , these correlated bounds do not crucially depend on the numerical value of @xmath13 as long as it is small . in passing we also remark that although the contours of @xmath73 in fig .
[ f : mhmc ] have been drawn for @xmath46 , they are not appreciably modified for @xmath89 .
things become more interesting when we superimpose , in fig .
[ f : mhmc ] , the constraint arising from the electroweak @xmath1-parameter . for @xmath67 gev or less
, the major contribution to the @xmath1-parameter comes from the loops involving the new nonstandard scalars . with @xmath81 and @xmath79 ( these two approximations
can already be justified from fig .
[ f : samh ] ) , the new physics contribution to the electroweak @xmath1-parameter is given by@xcite @xmath90 \,,\end{aligned}\ ] ] where , @xmath91 and @xmath92 are the weinberg angle and the @xmath93-boson mass respectively , and @xmath94 in we further use the relation ( [ e : mcc1 ] ) to substitute for @xmath73 . taking the new physics contribution to the @xmath1-parameter as@xcite @xmath95
we draw the allowed region in fig .
[ f : mhmc ] .
from there we see that the combined constraint implies that all the nonstandard scalars should be nearly degenerate . using the correlation of , and denoting the typical mass difference , @xmath96 by @xmath97 ( @xmath98 )
, we can simplify the expression of the @xmath1-parameter as follows : @xmath99 using the experimental number we can then find @xmath100 gev .
since the quarks couple only with the doublet and the physical scalar , @xmath45 , in general , is a mixed state of doublet and triplet fields , the tree level couplings of @xmath45 will be modified from those in the sm .
we define a generic modification factor for the fermions and vector bosons as follows : @xmath101 then , for @xmath102 , one can easily calculate@xcite @xmath103 where , @xmath104 represents any fermion and @xmath105 . denoting by @xmath106 the percentage of @xmath45 produced via the channel @xmath107 , we can express the modification of the higgs production cross section as follows : @xmath108 where , @xmath109 , @xmath110 , @xmath111 and @xmath112 at a cm energy of 7 and 8 tev@xcite . we know that a 125 gev sm higgs decays into @xmath113 channel with nearly 24% branching ratio and the rest decays almost entirely into two - body fermionic channels and into two gluons .
thus the modification of the total decay width can be expressed as @xmath114 [ cols="^,^ " , ] now we turn our attention to the modification of the partial decay widths of loop induced higgs decays like @xmath115 . to display conveniently the contribution of the charged scalar loops to the decay amplitude
, we define dimensionless parameters @xmath116 and @xmath117 in the following way , [ kappa ] @xmath118 where , the general expressions for @xmath119 and @xmath120 are given by [ g ] @xmath121 \,.\end{aligned}\ ] ] with these , the modification of the diphoton decay width can be written as @xmath122 where , using the notation , @xmath123 , the @xmath124 functions can be written as@xcite @xmath125 ^ 2 \ , , & \text{for } \tau\ge 1 \ , , \\
-\frac{1}{4}\left[\ln\left(\frac{1+\sqrt{1-\tau}}{1-\sqrt{1-\tau}}\right)-i\pi \right]^2 \ , , & \text{for } \tau < 1 \ , .
\end{cases}\end{aligned}\ ] ] now we can write the modified higgs signal strength as follows : @xmath126 in fig .
[ f : mcmu ] we have shown how @xmath127 behaves with the charged scalar masses .
here , one should note that the signal strength is suppressed compared to the sm expectations for heavy charged scalars .
thus observation of an excess in the diphoton channel , in future runs of the lhc , will disfavor the possibility of heavy charged scalars in this model . in passing ,
we comment that although we have assumed @xmath67 gev in fig .
[ f : mcmu ] , the above conclusions do not crucially depend on our numerical choice of @xmath13 as long as it remains small .
-parameter and the observed diphoton signal strength at 95% cl .
the inner green region represents how the constraint will tighten if @xmath127 is measured to be @xmath128 .
contours of @xmath73 are also shown.__,width=302,height=226 ] -parameter and the observed diphoton signal strength at 95% cl .
the inner green region represents how the constraint will tighten if @xmath127 is measured to be @xmath128 .
contours of @xmath73 are also shown.__,width=302,height=226 ] next , we can also define , similar to , the modified signal strength for the @xmath129 decay channel , @xmath130 .
however , for brevity , we do not display the explicit expressions here .
interested readers can find the relevant formulas in some earlier papers@xcite .
as has been noted in these references , since the sign of the @xmath131 vertex is opposite to that of the @xmath132 vertex , the singly charged scalar loop interfere destructively with the doubly charged one in the @xmath129 amplitude .
consequently , in some region of the parameter space depending on which loop dominates , we can have an _ anti - correlation _ between @xmath127 and @xmath130 , _
i.e. _ , one is enhanced compared to the sm while the other is suppressed .
but the new thing that we want to add to this context is the understanding of the behavior of @xmath127 and @xmath130 in the sm - like scenario .
one should note that , as the charged scalars become heavy , the function @xmath133 in saturates to @xmath134 .
hence the decoupling of the charged scalars from the loop induced higgs decays depends on how @xmath116 ( @xmath117 ) behaves with increasing @xmath78 ( @xmath73 ) .
we can easily check that , in the sm - like case defined by , the trilinear couplings in take the following forms : [ glim ] @xmath135 \approx -\frac{2m_{++}^2}{v_d } \,,\end{aligned}\ ] ] where , in the final step we have used .
thus it follows from that @xmath136 when @xmath137 .
consequently the charged scalars contribute substantially to the diphoton or @xmath138-photon decay amplitudes even when their masses lie in the tev regime .
in fact , one can check from that @xmath139 when @xmath46 for finite @xmath13 .
therefore the sm - like limit defined by is already forbidden from the current combined fit value of the diphoton signal strength , @xmath140@xcite .
this feature has been clearly depicted in fig .
[ f : overlap ] where we can see that the region around @xmath46 is ruled out from the higgs to diphoton data .
the reason for this large quantum effect for heavy masses can be understood once we realize that both the vevs in this case are nonzero .
this means the charged scalars obtain their masses entirely from spontaneous symmetry breaking ( ssb ) , which leads to large quantum effects as also has been noted in the multi doublet context@xcite .
but in the sm - like limit defined by one can obtain [ glim2 ] @xmath141 here we can easily see that the charged scalars can be decoupled in the limit , @xmath142 let us investigate the above limit in some greater detail
. since @xmath88 in this case , the triplet remains inert and @xmath30 in serves as a free mass squared parameter not connected to ssb . using to recover @xmath30 from @xmath13 and then plugging it in we can see that for this type of scenario , @xmath143 .
thus the condition ( [ dec ] ) essentially implies that the charged scalars obtain all their masses from a non - ssb origin , which , not surprisingly , lead to decoupling in the same way as in the inert doublet models@xcite . from the combined region in fig .
[ f : overlap ] , we see that the space for @xmath80 ( or , @xmath72 ) is severely constrained ( @xmath144 gev ) when the value of @xmath75 departs from @xmath145 .
as we will discuss below , even in the region @xmath86 we can obtain lower bounds on masses that are competitive with the direct experimental bounds .
much attention has been received by the doubly charged scalars because of the possibility of detecting them in the same sign dilepton channel . in this article
, we have shown that unitarity and @xmath1-parameter impart a degeneracy among the nonstandard scalars for @xmath146 gev .
because of this , the doubly charged scalar can now decay into mainly three channels@xcite
( i ) same sign dileptons , ( ii ) a pair of same sign @xmath93-bosons and ( iii ) a @xmath93-boson and a singly charged scalar . as we will explain shortly , continuation of the agreement of the diphoton signal strength with the corresponding sm expectation will result in tightening the degree of degeneracy between the nonstandard scalar masses . in that case , for the usual type ii scenario , the decay mode @xmath147 is likely to be suppressed and the doubly charged scalar will decay almost exclusively into same sign dileptons and/or same sign @xmath93-bosons .
when @xmath148 decays completely into two same sign dileptons , it is easy to look for it in the experiments and the bound on its mass is quite strong ( @xmath149 gev)@xcite . in the type ii seesaw model , the dominant decay mode of @xmath148 depends on the vev of the triplet@xcite . for @xmath150 gev ,
the doubly charged scalar decays entirely into a pair of leptons and the bound from direct searches applies .
but in the region @xmath151 gev , the @xmath148 decays mainly into a pair of @xmath93-bosons and therefore , it is very difficult to search for . in this case ,
a very weak bound has been placed on the mass of the doubly charged scalar using the lhc data@xcite . in this context
, we note that too low values for the masses of the charged scalar may give substantial contribution to the diphoton decay amplitude and therefore , it might be possible to set a lower bound on the masses from the observed value of the diphoton signal strength .
since all the nonstandard scalar masses are correlated , this bound can be translated into bounds on other scalar masses also .
taking into account the experimental bound from lep-2 , @xmath152 gev@xcite , along with the mass relations of we plot the @xmath153 allowed region from @xmath1-parameter and the observed diphoton signal strength in fig .
[ f : diphoton ] .
although we have displayed our results for @xmath67 gev , the plot remains essentially the same for any value of @xmath146 gev . from fig .
[ f : diphoton ] , we read the following lower bounds on the nonstandard scalar masses : @xmath154 gev , @xmath155 gev . in particular , for @xmath150 gev when @xmath156 gev applies , we can give the following lower bounds on the other nonstandard scalar masses : @xmath157 gev , @xmath158 gev .
from we see that a more precise measurement of @xmath127 has the potential to constrain the mass splittings more strongly than the @xmath1-parameter . to illustrate this point , in fig .
[ f : diphoton ] , we have also showed the futuristic projection of the @xmath153 allowed parameter region ( in green ) assuming that the diphoton signal strength will be measured to be consistent with the sm within 5% accuracy level . in that case , the bounds for @xmath150 gev can be improved to @xmath159 gev , @xmath160 gev .
in this paper we have revisited the constraints on the scalar sector of the type ii seesaw model .
we have worked under the assumption that the lightest cp - even scalar ( @xmath45 ) has been observed the lhc .
although we have exemplified our results for @xmath67 gev , our conclusions are mostly generic and valid for any @xmath13 of @xmath28 or less . to begin with
, we have re - derived the correlations between the nonstandard masses using unitarity .
we have expressed all our results in terms of physical masses and mixing angles .
consequently , we have noticed that @xmath75 becomes restricted within a narrow range around @xmath161 just from unitarity , whenever the neutral nonstandard scalar is heavier than 300 gev .
moreover , for nonzero @xmath13 , we have argued how a thin strip around @xmath46 can be ruled out from @xmath0 , which shrinks further the allowed band for @xmath75 ( around @xmath161 ) . using the mass relations in conjunction with the @xmath1-parameter , the separation between the masses , @xmath162 ,
can be restricted to be less than 50 gev .
we have also found that , for @xmath163 gev , an enhancement in @xmath127 is hardly possible . finally ,
using the experimental range of @xmath127 and @xmath1-parameter we plotted the @xmath153 allowed region in the @xmath80-@xmath78 plane for @xmath164 . using @xmath152 gev from lep-2 , we obtain the following limits on the other masses : @xmath165 moreover , when @xmath166 gev , the direct search bound from lhc , @xmath156 gev applies and then we can improve the bounds on other masses as follows : @xmath167 remembering that @xmath1-parameter restricts the mass splitting , we may conclude that the spectrum can be determined by essentially one mass parameter for @xmath166 gev .
we have also hinted how , in future , @xmath127 can play a very important role in restricting the splittings between nonstandard masses .
we thank ipsita saha for useful discussions . this work has been partially supported by the spanish mineco under grants fpa2011 - 23897 , fpa2014 - 54459-p , by the `` centro de excelencia severo ochoa '' programme under grant sev-2014 - 0398 and by the `` generalitat valenciana '' grant gvprometeoii2014 - 087 .
p. ramond , _ the family group in grand unified theories _ , in _ international symposium on fundamentals of quantum theory and quantum field theory palm coast , florida , february 25-march 2 , 1979 _ , pp . 265280 , 1979 .
http://arxiv.org/abs/hep-ph/9809459[hep-ph/9809459 ] .
e. ma and d. p. roy , _ heavy triplet leptons and new gauge boson _ , http://dx.doi.org/10.1016/s0550-3213(02)00815-5[_nucl .
_ * b644 * ( 2002 ) 290302 ] , [ http://arxiv.org/abs/hep-ph/0206150 [ hep - ph/0206150 ] ] .
f. cuypers and s. davidson , _ bileptons : present limits and future prospects _ , http://dx.doi.org/10.1007/s100520050157[_eur .
phys . j. _ * c2 * ( 1998 ) 503528 ] , [ http://arxiv.org/abs/hep-ph/9609487[hep-ph/9609487 ] ] .
f. del aguila and j. a. aguilar - saavedra , _ distinguishing seesaw models at lhc with multi - lepton signals _ , http://dx.doi.org/10.1016/j.nuclphysb.2008.12.029[_nucl .
phys . _ * b813 * ( 2009 ) 2290 ] , [ http://arxiv.org/abs/0808.2468 [ 0808.2468 ] ] . p. fileviez perez , t. han , g .- y .
huang , t. li and k. wang , _ neutrino masses and the cern lhc : testing type ii seesaw _ , http://dx.doi.org/10.1103/physrevd.78.015018[_phys .
rev . _ * d78 * ( 2008 ) 015018 ] , [ http://arxiv.org/abs/0805.3536 [ 0805.3536 ] ] .
a. melfo , m. nemevsek , f. nesti , g. senjanovic and y. zhang , _ type ii seesaw at lhc : the roadmap _ , http://dx.doi.org/10.1103/physrevd.85.055018[_phys .
_ * d85 * ( 2012 ) 055018 ] , [ http://arxiv.org/abs/1108.4416 [ 1108.4416 ] ] .
f. del aguila and m. chala , _ lhc bounds on lepton number violation mediated by doubly and singly - charged scalars _ , http://dx.doi.org/10.1007/jhep03(2014)027[_jhep_ * 03 * ( 2014 ) 027 ] , [ http://arxiv.org/abs/1311.1510[1311.1510 ] ] .
e. j. chun and p. sharma , _ search for a doubly - charged boson in four lepton final states in type ii seesaw _ , http://dx.doi.org/10.1016/j.physletb.2013.11.056[_phys .
_ * b728 * ( 2014 ) 256261 ] , [ http://arxiv.org/abs/1309.6888 [ 1309.6888 ] ] .
f. del aguila , m. chala , a. santamaria and j. wudka , _ discriminating between lepton number violating scalars using events with four and three charged leptons at the lhc _ , http://dx.doi.org/10.1016/j.physletb.2013.07.014[_phys .
lett . _ * b725 * ( 2013 ) 310315 ] , [ http://arxiv.org/abs/1305.3904 [ 1305.3904 ] ] .
a. g. akeroyd , m. aoki and h. sugiyama , _ lepton flavour violating decays tau > anti - l ll and mu
> e gamma in the higgs triplet model _ , http://dx.doi.org/10.1103/physrevd.79.113010[_phys
. rev . _ * d79 * ( 2009 ) 113010 ] , [ http://arxiv.org/abs/0904.3640 [ 0904.3640 ] ] .
a. arhrib , r. benbrik , m. chabab , g. moultaka , m. c. peyranere , l. rahili et al .
, _ the higgs potential in the type ii seesaw model _ , http://dx.doi.org/10.1103/physrevd.84.095005[_phys .
rev . _ * d84 * ( 2011 ) 095005 ] , [ http://arxiv.org/abs/1105.1925 [ 1105.1925 ] ] .
e. j. chun , h. m. lee and p. sharma , _ vacuum stability , perturbativity , ewpd and higgs - to - diphoton rate in type ii seesaw models _ , http://dx.doi.org/10.1007/jhep11(2012)106[_jhep_ * 11 * ( 2012 )
106 ] , [ http://arxiv.org/abs/1209.1303[1209.1303 ] ] . f. arbabifar , s. bahrami and m. frank , _ neutral higgs bosons in the higgs triplet model with nontrivial mixing _ , http://dx.doi.org/10.1103/physrevd.87.015020[_phys .
_ * d87 * ( 2013 ) 015020 ] , [ http://arxiv.org/abs/1211.6797 [ 1211.6797 ] ] .
p. s. bhupal dev , d. k. ghosh , n. okada and i. saha , _ 125 gev higgs boson and the type - ii seesaw model _
, http://dx.doi.org/10.1007/jhep03(2013)150 , 10.1007/jhep05(2013)049[_jhep _ * 03 * ( 2013 ) 150 ] , [ http://arxiv.org/abs/1301.3453[1301.3453 ] ] .
p. dey , a. kundu and b. mukhopadhyaya , _ some consequences of a higgs triplet _ , http://dx.doi.org/10.1088/0954-3899/36/2/025002[_j .
phys . _ * g36 * ( 2009 ) 025002 ] , [ http://arxiv.org/abs/0802.2510 [ 0802.2510 ] ] .
s. kanemura and k. yagyu , _ radiative corrections to electroweak parameters in the higgs triplet model and implication with the recent higgs boson searches _ , http://dx.doi.org/10.1103/physrevd.85.115009[_phys .
* d85 * ( 2012 ) 115009 ] , [ http://arxiv.org/abs/1201.6287 [ 1201.6287 ] ] .
c. bonilla , r. m. fonseca and j. w. f. valle , _ consistency of the triplet seesaw model revisited _ , http://dx.doi.org/10.1103/physrevd.92.075028[_phys .
rev . _ * d92 * ( 2015 ) 075028 ] , [ http://arxiv.org/abs/1508.02323 [ 1508.02323 ] ] .
l. lavoura and l .- f .
li , _ making the small oblique parameters large _ , http://dx.doi.org/10.1103/physrevd.49.1409[_phys .
_ * d49 * ( 1994 ) 14091416 ] , [ http://arxiv.org/abs/hep-ph/9309262 [ hep - ph/9309262 ] ] .
collaboration , _ measurements of the higgs boson production and decay rates and constraints on its couplings from a combined atlas and cms analysis of the lhc pp collision data at @xmath172 = 7 and 8 tev _ , _ atlas - conf-2015 - 044 _ ( 2015 ) .
chen , c .- q .
geng , d. huang and l .- h .
@xmath173 in type - ii seesaw neutrino model _ , http://dx.doi.org/10.1016/j.physletb.2013.05.007[_phys .
_ * b723 * ( 2013 ) 156160 ] , [ http://arxiv.org/abs/1302.0502 [ 1302.0502 ] ] .
n. haba , h. ishida , n. okada and y. yamaguchi , _ vacuum stability and naturalness in type - ii seesaw _ , http://dx.doi.org/10.1140/epjc/s10052-016-4180-z[_eur .
* c76 * ( 2016 ) 333 ] , [ http://arxiv.org/abs/1601.05217 [ 1601.05217 ] ] .
g. bhattacharyya and d. das , _ nondecoupling of charged scalars in higgs decay to two photons and symmetries of the scalar potential _ , http://dx.doi.org/10.1103/physrevd.91.015005[_phys .
_ * d91 * ( 2015 ) 015005 ] , [ http://arxiv.org/abs/1408.6133 [ 1408.6133 ] ] .
t. han , b. mukhopadhyaya , z. si and k. wang , _ pair production of doubly - charged scalars : neutrino mass constraints and signals at the lhc _ , http://dx.doi.org/10.1103/physrevd.76.075013[_phys
_ * d76 * ( 2007 ) 075013 ] , [ http://arxiv.org/abs/0706.0441 [ 0706.0441 ] ] .
j. garayoa and t. schwetz , _ neutrino mass hierarchy and majorana cp phases within the higgs triplet model at the lhc _ , http://dx.doi.org/10.1088/1126-6708/2008/03/009[_jhep_ * 03 * ( 2008 ) 009 ] , [ http://arxiv.org/abs/0712.1453[0712.1453 ] ] .
m. kadastik , m. raidal and l. rebane , _ direct determination of neutrino mass parameters at future colliders _ , http://dx.doi.org/10.1103/physrevd.77.115023[_phys .
* d77 * ( 2008 ) 115023 ] , [ http://arxiv.org/abs/0712.3912 [ 0712.3912 ] ] .
a. g. akeroyd , m. aoki and h. sugiyama , _ probing majorana phases and neutrino mass spectrum in the higgs triplet model at the cern lhc _ , http://dx.doi.org/10.1103/physrevd.77.075010[_phys .
rev . _ * d77 * ( 2008 ) 075010 ] , [ http://arxiv.org/abs/0712.4019 [ 0712.4019 ] ] .
s. kanemura , k. yagyu and h. yokoya , _ first constraint on the mass of doubly - charged higgs bosons in the same - sign diboson decay scenario at the lhc _ , http://dx.doi.org/10.1016/j.physletb.2013.08.054[_phys .
_ * b726 * ( 2013 ) 316319 ] , [ http://arxiv.org/abs/1305.2383[1305.2383 ] ] .
s. kanemura , m. kikuchi , h. yokoya and k. yagyu , _ lhc run - i constraint on the mass of doubly charged higgs bosons in the same - sign diboson decay scenario _ , http://dx.doi.org/10.1093/ptep/ptv071[_ptep_ * 2015 * ( 2015 ) 051b02 ] , [ http://arxiv.org/abs/1412.7603 [ 1412.7603 ] ] .
z. kang , j. li , t. li , y. liu and g .- z .
ning , _ light doubly charged higgs boson via the @xmath176 channel at lhc _ ,
* c75 * ( 2015 ) 574 ] , [ http://arxiv.org/abs/1404.5207 [ 1404.5207 ] ] .
collaboration , j. abdallah et al . , _
search for doubly charged higgs bosons at lep-2 _ , http://dx.doi.org/10.1016/s0370-2693(02)03125-8[_phys .
lett . _ * b552 * ( 2003 ) 127137 ] , [ http://arxiv.org/abs/hep-ex/0303026 [ hep - ex/0303026 ] ] . | we show that in the higgs triplet model , after the higgs discovery , the mixing angle in the cp - even sector can be strongly constrained from unitarity .
we also discuss how large quantum effects in @xmath0 may arise in a sm - like scenario and a certain part of the parameter space can be ruled out from the diphoton signal strength . using @xmath1-parameter and diphoton signal strength measurements , we update the bounds on the nonstandard scalar masses . + * updated scalar sector constraints in higgs triplet model * + + |
flow around a circular cylinder is a classical topic in hydrodynamics that is of fundamental importance to many scientific fields with numerous applications @xcite .
of particular interest is the formation , at moderate reynolds numbers , of vortex eddies behind a circular cylinder , which then go unstable at higher reynolds numbers and evolve into a karman vortex street @xcite . since an analytic treatment of the problem in terms of the navier - stokes equation is difficult and the computational cost of direct numerical simulation very high , a particularly useful approach to study the basic features of vortex shedding from bluff bodies is to consider the dynamics of point vortices in an inviscid fluid . a point - vortex model for the formation of two recirculating , symmetric eddies in the wake of a circular cylinder
was first introduced by fppl @xcite .
he obtained stationary solutions for a pair of vortices behind the cylinder in a uniform stream and found that the centers of the eddies observed in the experiments lie on the locus of such equilibria now called the fppl curve .
in addition , fppl found that these equlibria , although stable against perturbations that are symmetric with respect to the centerline , were unstable against nonsymmetric perturbations .
this instability is believed to constitute the origin of the vortex shedding process that leads to the formation of the karman vortex street @xcite .
it was later found out independently by several authors @xcite that fppl s stability analysis for symmetric perturbations was in error in that the stationary solution behind the cylinder is not exponentially but only marginally stable .
physically , marginal stability implies , for instance , that if a splitter plate is placed behind the cylinder in the center plane of the wake to suppress vortex shedding @xcite , oscillating forces on the cylinder may still arise owing to the cyclic motion of the vortices around their equilibrium position @xcite .
despite many contributions to the problem , it is fair to say that the nonlinear dynamics of the fppl system is not yet fully understood . in particular , a more complete picture of vortex - pair dynamics in the presence of symmetric perturbations is lacking , and several aspects of the nonlinear dynamics for nonsymmetric perturbations remain unclear . to address these two issues
is the main motivation of the present paper .
it should be emphasized at the outset that a better understanding of the dynamical structure underlying the fppl model is of interest not only because of its practical relevance for vortex shedding , but also in its own theoretical right from the viewpoint of nonlinear dynamics .
the fppl model has inspired a number of studies on several related problems , such as the modeling of vortex wake behind slender bodies in terms of multiple pairs of point vortices @xcite , the hamiltonian structure of a circular cylinder interacting dynamically with point vortices @xcite , the control of vortex shedding @xcite , and the stability of symmetric and asymmetric vortex pairs over three - dimensional slender conical bodies @xcite .
the related problem of desingularization of the fppl pair in terms of vortex patches of finite area was also studied @xcite . a recent review on vortex motion past solid bodies with additional references to the fppl model and related problems
can be found in ref .
[ 28 ] . after formulating the problem of a pair of counter - rotating point vortices placed in a uniform stream around a circular cylinder in sec .
[ sec:2 ] , we begin our analysis of the fppl system in sec . [ sec:3 ] by studying its hamiltonian dynamics restricted to the invariant subspace where the vortices move symmetrically with respect to the centerline .
a phase portrait of the system is presented that fully characterizes the dynamics within this symmetric subspace . in particular , we point out that in addition to the two previously known sets of equilibira , namely , the fppl equilibrium and the equilibrium on the axis bisecting the cylinder perpendicularly to the uniform flow , the system possesses a hitherto unnoticed nilpotent saddle at infinity .
we show furthermore that the homoclinic orbits associated with this nilpotent saddle delimit the region of closed orbits around the fppl equilibrium .
we proceed in sec .
[ sec:4 ] to study the linear and nonlinear dynamics resulting from antisymmetric perturbations of the fppl equilibrium . in the linear regime ,
a mistake that went undetected in fppl s expressions @xcite for the corresponding eigenvalues is now corrected . as for the nonlinear dynamics , the unstable manifold associated with the fppl equilibrium
is computed numerically and its close relation to the vortex shedding instability is pointed out .
the linear stability analysis of the equilibria on the normal line with respect to symmetric and antisymmetric perturbations is also presented for the first time , it seems and the respective nonlinear dynamics is investigated numerically .
a discussion of the physical relevance of our findings and our main conclusions are presented in sec .
[ sec : discuss ] .
we consider the motion of a pair of point vortices of same strength and opposite polarities around a circular cylinder of radius @xmath0 and in the presence of a uniform stream of velocity @xmath1 , as illustrated in fig .
[ fig:1 ] .
it is convenient to work in the complex @xmath2-plane , where @xmath3 , and place the center of the cylinder at the origin .
the upper and lower vortices are located at positions @xmath4 and @xmath5 , respectively .
the complex potential @xmath6 , with @xmath7 being the velocity potential and @xmath8 the stream function , is given by @xcite @xmath9 where @xmath10 is the circulation of the vortex at @xmath11 and bar denotes complex conjugation .
( [ eq:1 ] ) , the first two terms represent the incoming flow and its image ( a doublet at the origin ) with respect to the cylinder , the third term gives the contributions to the complex potential from the upper vortex and its image , and similarly the last term contains the contributions from the lower vortex and its image . as can be inferred from fig .
[ fig:1 ] , a necessary condition for a steady configuration to exist is that the upper ( lower ) vortex be of negative ( positive ) circulation , hence only the case @xmath12 is of interest to us here .
in dimensionless variables @xmath13 the complex potential ( [ eq:1 ] ) becomes @xmath14 where the prime notation has been dropped . according to standard theory of point vortices in an inviscid fluid ,
any given vortex moves with the velocity of the flow computed at the position of that vortex , excluding its own contribution to the flow .
it then follows from eq .
( [ eq:12 ] ) that the velocity @xmath15 of the vortex located at @xmath16 is given by @xmath17 or more explicitly [ eq:24 ] @xmath18 where @xmath19 , @xmath20 .
the velocity @xmath21 of the second vortex is obtained by simply interchanging the indexes @xmath22 in eq .
( [ eq:24 ] ) and letting @xmath23 .
it is not difficult to see from eq .
( [ eq:24 ] ) that if the vortices are initially placed at positions symmetrically located with respect to the centerline , i.e. , @xmath24 , then this symmetry is preserved for all later times , i.e. , @xmath25 for @xmath26 . in this section ,
we study the dynamics within this invariant symmetric subspace , where the motion of the lower vortex is simply the mirror image of that of the upper vortex with respect to the centerline .
symmetry can be enforced experimentally by placing a splitter plate behind the cylinder in the center plane of the wake @xcite . with @xmath27 and @xmath28 , eq .
( [ eq:24 ] ) reduces to [ eq:4 ] @xmath29 , \label{eq:4a}\ ] ] @xmath30 .
\label{eq:4b}\ ] ] here , the subscripts have been dropped with the understanding that in the remainder of the section we restrict our attention to the upper vortex .
as is well known , the equations of motion for point vortices in a two - dimensional inviscid flow , first derived by kirchhoff , can be formulated as a hamiltonian system @xcite .
the dynamics of point vortices in the presence of closed , rigid boundaries was shown by lin @xcite to be also hamiltonian with the same canonical sympletic structure as in the absence of boundaries . for a vortex pair placed in a uniform stream around a circular cylinder ,
the phase space is four - dimensional and has a two - dimensional ( 2d ) invariant subspace corresponding to symmetric orbits .
the hamiltonian restricted to the 2d symmetric subspace is given by @xcite @xmath31 the corresponding dynamical equations @xmath32 where dot denotes time derivative , yield eq .
( [ eq:4 ] ) upon identifying @xmath33 with @xmath34 . a phase portrait of this hamiltonian system for @xmath35 is presented in fig . [ fig : phase ] , where the curves shown are ( unevenly spaced ) level sets of the hamiltonian ( [ eq : h ] ) .
[ for convenience , these curves were obtained from a direct numerical integration of eq .
( [ eq:4 ] ) . ] a detailed description of the main features of this phase portrait will be given below , starting with an analysis of the various equilibrium points and their stability .
the related problem of the symmetric `` moving fppl system , '' where the cylinder advances through the fluid followed by the vortex pair , was recently considered by shashikanth _
@xcite , but there the phase portrait @xcite is quite different from the one shown in fig .
[ fig : phase ] , because of the additional degrees of freedom related to the velocity of the moving cylinder . .
the isolated black dots are the fppl equilibria .
the dashed curves are the stable and unstable branches of the separatrix associated with the equilibrium point on the normal line , and the thick solid lines are the homoclinic loops of the equilibrium point at infinity ; see text.,scaledwidth=60.0% ] the equilibrium positions for the vortex are obtained by setting @xmath36 in eq .
( [ eq:4 ] ) .
three types of equilibrium points can be identified .
the locus of possible equilibrium positions @xmath37 for the upper vortex found by fppl @xcite is the curve @xmath38 with corresponding strength @xmath39 along the fppl curve ( [ eq : pair ] ) , the vortex strength increases with distance from the center of the cylinder and diverges linearly for @xmath40 . for the equilibrium point on the edge of the cylinder ( @xmath41 ) ,
the strength vanishes .
notice that eq .
( [ eq : pair ] ) yields two branches of solution : one in which the vortex pair is behind the cylinder ( @xmath42 ) and the other where the vortex pair is in front of the cylinder ( @xmath43 ) .
the former case models the formation of vortex eddies behind a cylinder in a uniform stream and was the primary motivation of fppl s original study @xcite .
the latter case has attracted far less attention because it is not usually observed in experiments .
we note , however , that recirculating eddies are observed in front of a circular cylinder near a plane boundary when the gap between the cylinder and the plane is sufficiently small @xcite . in this context
, the fppl equlibrium upstream of the cylinder may eventually be relevant for flows around a half - cylinder placed on a plane wall ( or for the closely related situation where a splitter plate is attached to the front of the cylinder ) , although we are unaware of specific experiments in this setting .
this corresponds to the upper vortex being located on the line bisecting the cylinder perpendicularly to the incoming flow @xcite , that is , @xmath44 with strength @xmath45 as in the fppl solution , the strength tends to zero when the edge of the cylinder is reached ( @xmath46 ) and diverges linearly with distance from the center of the cylinder . at large distances ,
the vortex strength for this equilibrium is about twice that of a fppl pair located at the same distance from the origin .
equation ( [ eq:4 ] ) also yields equilibrium points at the positions @xmath47 to the best of our knowledge , the existence of this additional equilibrium point at infinity was not noted before .
its physical origin , however , can be easily understood , as it corresponds to the equilibrium configuration for a vortex pair placed in a uniform stream ( without the cylinder ) . at points infinitely far from the cylinder , the flow induced by the image system ( inside the cylinder ) becomes negligible and hence a stationary configuration is possible if the vortices with given circulation @xmath48 are placed at the appropriate distance ( @xmath49 ) from each other .
the linear stability analysis of the equilibria described above is presented next , together with a discussion of the _ nonlinear _ stability of the fppl equilibrium .
consider a perturbation of the fppl equilibrium ( [ eq : pair ] ) parameterized as : @xmath50 , where @xmath51 , with @xmath52 and @xmath53 being infinitesimal ( real ) quantities .
linearization of eq .
( [ eq:4 ] ) then yields the following dynamical system @xmath54 where the matrix @xmath55 reads @xmath56 @xmath57 @xmath58 its eigenvalues @xmath59 are given by @xmath60 for @xmath61 .
the eigenvalues are thus purely imaginary , and not a complex pair with negative real part as found by fppl @xcite . in other words ,
the fppl equilibrium is a center and not a stable focus .
our equation ( [ eq : ls ] ) agrees with the expression for the eigenvalues of the symmetric modes obtained in ref . [ 7 ] from the linearization of the full 4d dynamical system . as can be seen from fig .
[ fig : phase ] , the fppl solution is in fact a nonlinearly stable center , meaning that when the vortex is displaced from its equilibrium position by a small ( but finite ) amount , it executes a periodic motion around that point , corresponding to the closed orbits in the figure .
this periodic motion around the fppl equilibrium has been observed in numerical simulations of the model carried out by de laat and coene @xcite .
note that since the eigenvalues given in eq .
( [ eq : ls ] ) do not depend explicitly on the coordinate @xmath62 , it follows that the two fppl equilibria , downstream and upstream of the cylinder , have identical stability properties , as is evident from fig .
[ fig : phase ] .
this means , in particular , that if vorticity can be generated upstream of the cylinder then stationary recirculating eddies could form in front of the cylinder a situation observed , for instance , in flows around a cylinder placed above a plane wall @xcite .
linearization of eq .
( [ eq:4 ] ) around the equilibrium point @xmath63 yields for the matrix @xmath55 : @xmath64 @xmath65 @xmath66 the eigenvalues @xmath59 of this matrix are determined by @xmath67 which yields a pair of real eigenvalues , @xmath68 .
the equilibrium point on the normal line is therefore a saddle , having a stable and unstable direction , as is also evident from the phase portrait shown in fig .
[ fig : phase ] .
the eigenvectors @xmath69 associated with the eigenvalues @xmath70 , respectively , read @xmath71 although it was known from numerical simulations @xcite that the equilibrium point on the normal line is unstable ( against generic symmetric perturbations ) , it seems that an explicit linear stability analysis for this case was not carried out before , perhaps because these equilibria were not considered physically relevant since they are not observed in experiment @xcite .
however , when the full nonlinear dynamics is considered , the stable and unstable eigendirections @xmath69 give origin to the respective stable and unstable separatrices , indicated by the dashed curves in fig .
[ fig : phase ] . in this sense , the existence of an equilibrium point on the normal line is dynamically felt by a vortex even if it is placed far from this `` unphysical '' equilibrium .
the matrix @xmath55 of the linearized system around the equilibrium point at infinity is given by @xmath72 which is nilpotent and has two zero eigenvalues . to study the stability of this equilibrium point
, one needs to examine the nonlinear contributions . to this end
, we note that for @xmath73 and @xmath74 , eq . ( [ eq:4b ] ) assumes the form @xmath75 it then follows from a theorem in ordinary differential equations @xcite that , in view of the cubic term in eq .
( [ eq : doty ] ) , the equilibrium point is a _ degenerate _ or _ nilpotent _ saddle @xcite , for which the two eigenvectors are the same .
the behavior of trajectories in the neighborhood of a generic nilpotent saddle is illustrated in fig .
[ fig : saddle ] .
the behavior near the nilpotent saddle at @xmath76 and @xmath77 can be described as follows .
a vortex placed very far downstream and below ( above ) the line @xmath77 will move away from ( towards ) the equilibrium point at @xmath78 .
similarly , a vortex placed very far upstream will move away from ( towards ) the equilibrium point at @xmath79 if @xmath80 ( @xmath81 ) .
the stable and unstable separatrices associated with the nilpotent saddle at infinity form two homoclinic loops @xcite , called nilpotent saddle loops , which are indicated in fig .
[ fig : phase ] by thick solid lines and correspond to the level curves passing through this equilibrium point : @xmath82 the nilpotent saddle loops encircle the fppl equilibria and define their region of nonlinear stability , in the sense that vortex trajectories are closed for initial positions inside the loops and unbounded otherwise . in this way , the nilpotent saddle at infinity , which went unnoticed until now , allows us to fully characterize the nonlinear stability of the fppl equilibrium . for unbounded orbits ,
the long - time asymptotic behavior depends on the location of the vortex initial position with respect to the separatrices associated with the equilibrium point on the normal line .
a vortex placed downstream of the cylinder between the nilpotent saddle loop and the separatrices of the equilibrium point on normal line will eventually be convected away by the free stream ; see fig .
[ fig : phase ] .
in particular , if the vortex starts very far behind the cylinder at a position that is below the nilpotent saddle loop and above the stable separatrix , it first moves towards the cylinder , turns around the fppl equilibrium , and is then `` reflected '' back to infinity .
even more surprising trajectories arise if the vortex is placed downstream below the stable separatrix , for it will be close enough to its image below the centerline to be able to overcome the cylinder and move off to infinity upstream .
( a related phenomenon occurs in the inviscid coupled motion of a cylinder initially at rest and a vortex pair starting at infinity with no imposed background flow @xcite .
when the cylinder is less dense than the fluid , it is found that if the vortices are released sufficiently above the centerline they reverse relative to the moving cylinder ; otherwise , they move over and past the cylinder . ) unbounded trajectories for the fppl system also result for initial positions upstream of the cylinder : i ) if placed above the stable separatrix , the vortex moves downstream to infinity ; and ii )
if placed between the stable separatrix and the nilpotent saddle loop , the vortex goes around the fppl equlibrium in front of the cylinder and returns to infinity upstream ; see fig .
[ fig : phase ] .
it is again the hitherto unnoticed nilpotent saddle at infinity , together with the precise nature of the equilibrium point on the normal line , that allows us to go beyond linear stability analysis and capture the full phase portrait in the symmetric subspace .
we stress that closed orbits exist only when the flow is symmetric .
nonsymmetric perturbations inevitably cause the vortex pair to move off to infinity , as we demonstrate next .
in this section , the effect of antisymmetric perturbations on the equilibria of the fppl system is studied .
we begin by observing that the dynamics of two counter - rotating point vortices possesses a _
conjugation symmetry_. to describe this symmetry , let @xmath83 and @xmath84 denote the trajectories of the upper and lower vortices , respectively , with initial positions @xmath85 and @xmath86 . for the dynamical system defined by eq .
( [ eq:24 ] ) and the corresponding equation for the second vortex , one can verify that the following relations hold [ eq : cc ] @xmath87 @xmath88 in other words , for any given pair of initial positions , @xmath85 and @xmath86 , there exists a `` conjugate pair '' of initial positions , @xmath89 and @xmath90 , such that the vortex trajectories of the first pair are the complex conjugate of those of the second pair .
any perturbation of a vortex - pair equilibrium can be written as the superposition of a symmetric perturbation and an antisymmetric one . to be precise ,
antisymmetric perturbations are of the form @xmath91 where @xmath92 denotes a generic equilibrium point and @xmath93 .
since the antisymmetric subspace of the full 4d phase space is invariant under linear dynamics , we can focus on the upper vortex in carrying out our linear stability analysis .
linearization of eq .
( [ eq:24 ] ) around the fppl equilibrium ( [ eq : pair ] ) with respect to antisymmetric perturbations ( [ eq : daz ] ) yields @xmath94 where the matrix @xmath95 is given by @xmath96 @xmath97 @xmath98 this matrix has a pair of real eigenvalues , @xmath99 , where @xmath100 the fppl equilibrium is therefore a saddle with respect to antisymmetric perturbations , while it is a center with respect to symmetric perturbations , as seen earlier .
that is , the fppl equilibrium is a _ saddle - center _ of the full 4d dynamical system @xcite .
we note in passing that , although fppl obtained a pair of real eigenvalues for the case of antisymmetric perturbations , his original formulae for the eigenvalues are in error @xcite . our expression ( [ eq : las ] )
is in agreement with the eigenvalues of the skew - symmetric modes obtained by smith @xcite from the linearization of the full dynamical system .
the eigenvectors @xmath101 associated with the eigenvalues @xmath70 are readily computed , with the result @xmath102 , in which case @xmath103 and @xmath104 ( black dots ) .
the solid and dashed curves are the trajectories starting along the unstable directions @xmath105 and @xmath106 , respectively , while the short straight lines indicate the axes defined by the stable direction @xmath107 .
the dotted lines represent the loci of the fppl equilibria.,scaledwidth=60.0% ] in fig . [
fig : anti ] , we show in solid curves the pair of vortex trajectories obtained by slightly displacing the vortices from their equilibrium positions in the directions defined by the unstable eigenvector @xmath108 , while the trajectories obtained by slightly displacing the vortices in the opposite directions are shown in dashed curves .
the latter pair of trajectories is the complex conjugate of the former by conjugation symmetry .
note that for the first pair of trajectories , the lower vortex initially moves towards the centerline and upstream , while the upper vortex moves away from the centerline and downstream . at later times , the vortex pair moves off to infinity with the lower vortex trailing behind the upper vortex . for the second pair of trajectories , the upper and lower vortices switch role ; see fig .
[ fig : anti ] . in the flow of a real fluid past a cylinder ,
the two basic instabilities associated with displacements along the unstable directions @xmath109 happen alternately and constitute the origin of vortex shedding that leads to the formation of the karman vortex street @xcite . in like manner , the suppression of vortex shedding by placing a splitter plate behind the cylinder @xcite is consistent with the fact that the fppl equlibrium is nonlinearly stable with respect to symmetric perturbations ; see sec .
[ sec : discuss ] for further discussions on vortex shedding and its suppression by a splitter plate . for small , generic antisymmetric perturbations ,
the vortices move along trajectories that follow closely the ones depicted in fig .
[ fig : anti ] . whether a vortex pair eventually moves up or down is determined by the initial position of the upper vortex relative to the stable direction @xmath107 , which is indicated in fig . [
fig : anti ] by the short straight line passing through the fppl equilibrium .
if the initial position of the upper vortex is to the right ( left ) of the stable direction , then the vortex pair asymptotically moves upwards ( downwards ) .
this explains the behavior seen in the numerical simulations reported in ref .
[ 25 ] , where nearby initial positions around the fppl equilibrium were found to lead to close - by trajectories .
since any degree of antisymmetry in the initial perturbation causes the vortex pair to move off to infinity , the fppl equilibrium is unstable under generic perturbations . as an example , fig .
[ fig : loop ] shows vortex trajectories obtained by displacing the fppl pair ( at @xmath110 ) by the amounts @xmath111 . during the linear stage ,
the trajectories are a superposition of a symmetric orbit and a growing mode associated with the antisymmetric component of the perturbation , which ultimately leads to asymptotic trajectories with the vortices moving parallel to each other . of the fppl pair at @xmath110 ( black dots).,scaledwidth=60.0% ] for antisymmetric perturbations of the equilibrium ( [ eq : b ] ) on the normal line ,
the matrix @xmath95 assumes the form @xmath112
@xmath113 @xmath114 with eigenvalues @xmath59 given by @xmath115 this yields a pair of real eigenvalues , @xmath99 , with respective eigenvectors : @xmath116 of the equilibrium at @xmath117 ( black dots ) .
the dashed curves are trajectories resulting from the antisymmetric perturbation @xmath118.,scaledwidth=60.0% ] in fig .
[ fig : normal ] , we show the vortex trajectories ( solid curves ) obtained by slightly displacing the vortices from their equilibrium position along the unstable direction @xmath105 for @xmath119 .
the initial motion here is somewhat similar to what is seen for a fppl pair , in the sense that one vortex moves upstream towards the centerline and the other moves downstream away from the centerline .
the main difference is that for later times , the vortices now end up moving upstream .
the long - time dynamics in this case is also more sensitive on the initial conditions : for somewhat larger perturbations , the vortices are eventually carried away by the free stream .
an example where this happens is indicated by the dashed curves in fig .
[ fig : normal ] , which represent the vortex trajectories for the antisymmetric perturbation @xmath120 . as already argued in sec .
[ sec : c2 ] , although the equilibrium point on the normal line is not directly observed in experiments , it is important to know its instability properties under both symmetric and antisymmetric perturbations .
this knowledge contributes to a better understanding not only of the full nonlinear dynamics of the fppl system but also of more general flows , such as the case of stationary vortex patches above and below the cylinder in a uniform stream , where similar unstable modes are observed @xcite .
in this paper , we have investigated a two - dimensional vortex model for the formation of recirculating eddies behind a fixed cylinder placed on a uniform stream .
the model , which was first introduced by fppl @xcite almost a century ago , has two main simplifying assumptions : i ) the fluid is treated as inviscid and hence the flow is potential , and ii ) the size of the vortex core is neglected and so the vortices are considered to be point - like . in spite of these simplifications , the model is known to be in qualitative agreement with real flows past a cylinder , as was already pointed out by fppl in his original paper .
several novel features of the fppl model have been obtained in the present work , which help one to better understand the basic dynamics of vortex shedding behind a cylinder . in real flows , governed by the navier - stokes equations , stationary vortices behind a cylinder
are formed at moderate reynolds number ( @xmath121 ) . as the reynolds number increases
past @xmath122 , the configuration loses its symmetry and becomes unstable .
new vortices then start to form alternately on both sides of the cylinder , while the vortices further downstream break away and develop into a karman vortex street , as described by fppl @xcite .
it has been argued by roshko @xcite that `` possibly the breaking away should be regarded as primary , resulting in asymmetry . ''
the analysis presented in sec .
[ sec : unst ] makes it clear that the reverse scenario is more plausible : the asymmetrical disturbances induce the instability of the vortex pair which then breaks away from the cylinder . as vorticity
is continuously generated from the separated boundary layer on both sides of the cylinder , new vortices are formed and alternately shed into the far wake of the cylinder according to the unstable modes shown in fig . [
fig : anti ] .
direct numerical simulations ( dns ) of two - dimensional flows past a cylinder performed by tang and aubry @xcite have confirmed that the mechanism for the instability of the symmetric eddies in real flows is qualitatively described by the instability of the point - vortex model .
it is experimentally observed @xcite that vortex shedding is suppressed if a splitter plate is installed behind the cylinder in the center plane of the wake .
the presence of the splitter plate tends to enforce symmetry of the flow with respect to the centerline , thus effectively reducing the appearance of antisymmetric disturbances behind the cylinder .
the suppression of vortex shedding in this case is thus entirely consistent with the fact that the fppl equilibria of the vortex - point model is nonlinearly stable against symmetric perturbations and that vortex shedding is induced by unstable antisymmetric modes , as discussed above .
this scenario has been confirmed by dns of flows past a cylinder with symmetry imposed along the centerline recently performed by kumar _
the problem of stationary configurations for vortex flows past a cylinder with patches of constant vorticity has also been studied numerically by elcrat _
these authors found two families of solutions , representing desingularized versions of the fppl and the normal equilibria , respectively , which have the same stability properties as the corresponding point - vortex equilibria . in conclusion
, we have seen that the fppl model , where a pair of counter - rotating point vortices move around a circular cylinder in the presence of a uniform stream , is a rich nonlinear dynamical system whose features notably its stability properties bear a direct relevance to our understanding of the vortex shedding mechanism in real flows .
the results obtained here should , in principle , carry over to more general geometries , such as vortex motion around a plate or around a cylinder with noncircular cross section .
b. n. shashikanth , j. e. marsden , j. w. burdick , and s. d. kelly , `` the hamiltonian structure of a two - dimensional rigid circular cylinder interacting dynamically with @xmath123 point vortices , '' phys .
fluids * 14 * , 1214 ( 2002 ) .
( erratum : phys .
fluids * 14 * , 4099 ( 2002 ) . ) | the motion of a pair of counter - rotating point vortices placed in a uniform flow around a circular cylinder forms a rich nonlinear system that is often used to model vortex shedding . the phase portrait of the hamiltonian governing the dynamics of a vortex pair that moves symmetrically with respect to the centerline a case that can be realized experimentally by placing a splitter plate in the center plane is presented .
the analysis provides new insights and reveals novel dynamical features of the system , such as a nilpotent saddle point at infinity whose homoclinic orbits define the region of nonlinear stability of the so - called fppl equilibrium .
it is pointed out that a vortex pair properly placed downstream can overcome the cylinder and move off to infinity upstream .
in addition , the nonlinear dynamics resulting from antisymmetric perturbations of the fppl equilibrium is studied and its relevance to vortex shedding discussed . |
off - diagonal ( or skewed ) parton distributions provide important information about the nonperturbartive structure of the nucleon @xcite .
they can in principle be measured in such processes as deeply virtual compton scattering , diffractive vector meson production or diffractive high-@xmath0 jet production .
just as for the ordinary ( diagonal ) parton distributions , the qcd evolution equations @xcite play an important role in the determination of skewed parton distributions . as usual input distributions are required .
however , this is more complicated than in the diagonal case since the skewed parton distributions depend on additional variable the asymmetry parameter @xmath1 .
we use the symmetric formulation of ji @xcite in which the skewed parton distributions are given by functions @xmath2 with support @xmath3 , see @xcite for more details . here
we discuss various ways to specify the initial conditions and show how the differences in the input distributions disappear on evolution .
in particular we illustrate the conclusion of @xcite that the skewed distributions @xmath2 , at small @xmath4 and @xmath5 , are fixed by the conventional diagonal partons .
the distributions @xmath6 have to fulfil several conditions with respect to the variable @xmath5 .
first , the time - reversal invariance and hermiticity impose the condition @xcite @xmath7 thus @xmath6 is an even function of @xmath5 .
the second condition states that in the limit @xmath8 we recover the ordinary diagonal parton distributions @xmath9 the third condition is more complicated but has a simple origin , @xcite .
the @xmath10 moment of @xmath6 is a polynomial in @xmath5 of the order @xmath11 at most @xmath12}\;a_{n , i}\;\xi^{2i}\ , , \label{eq:4}\ ] ] which also embodies condition ( [ eq:2 ] ) .
finally we impose continuity of @xmath6 at the border @xmath13 between two different physical regions , see @xcite .
this ensures that the amplitude of a physical process , described by skewed distributions , is finite .
all these conditions have to be fulfilled in the construction of the input for evolution .
they are , of course , conserved during the evolution .
there are two equivalent ways to evolve @xmath14 up in the scale @xmath15 . the first method , which is most appropriate at small @xmath16 , uses the evolution of the gegenbauer moments of @xmath6 and the shuvaev transform @xcite to find the final answer in the @xmath17space @xcite . in the second method the solutions
are found after imposing initial conditions and numerically solving the evolution equations , directly in the @xmath17space .
here we supplement the studies of @xcite by presenting results from the second approach .
condition ( [ eq:4 ] ) is difficult to fulfil in order to specify initial distributions at a certain scale @xmath18 .
it may be facilitated by the use of the double distribution @xmath19 @xcite @xmath20 where @xmath21 is the square @xmath22 this prescription introduces a nontrivial mixing between @xmath4 and @xmath5 . condition ( [ eq:2 ] )
is guaranteed if @xmath23 .
we still have freedom to add to ( [ eq:5 ] ) a function @xmath24 , antisymmetric in @xmath25 , contained entirely in the erbl - like region @xmath26 @xcite .
the only problem left is to build in the ordinary diagonal distributions in the prescription ( [ eq:5 ] ) . to do this
we take @xmath27 where different choices of @xmath28 give , via ( [ eq:5 ] ) , different initial conditions for @xmath29 .
three choices , @xmath30 are shown in fig . 1 , with @xmath31 given by grv @xcite at @xmath32 .
the first choice gives simply the diagonal input @xmath33 , independent of @xmath5 .
the second , with @xmath34 for quarks ( gluons ) , generates an input form similar to that obtained from the model of @xcite in which the gegenbauer moments of @xmath6 are @xmath5-independent .
this property is conserved by the evolution .
the exact form of the double distribution in this case can be found in @xcite .
the last choice was selected so as to give an oscillatory input behaviour in the erbl - like region .
fig . 1 shows the quark non - singlet , quark singlet and gluon skewed distributions for the three input models evolved up in @xmath15 for @xmath35 .
the results for larger values of @xmath5 are qualitatively the same @xcite .
we see that the form of @xmath36 has the most influence in the erbl region .
however even in this region evolution soon washes out the differences . already by @xmath37
gev@xmath38 the three curves are close to each other . in the dglap region , @xmath39 ,
the curves are almost identical while in the erbl region , @xmath26 , they approach the asymptotic form for each particular skewed parton distribution .
such behaviour is especially important in view of the fact that one set of the curves is obtained from the pure diagonal input parton distributions . in this way
we illustrate the result presented in @xcite that to a good accuracy at small @xmath5 , the skewed distributions @xmath40 are completely known in terms of conventional partons .
thus , to summarize , the nonperturbative information contained in the diagonal input parton distributions and particular features of the evolution equations for skewed parton distributions are sufficient for their determination .
we thank max klein and johannes blmlein for their efficient organization of dis99 , and the royal society and the eu fourth framework programme ` training and mobility of researchers ' , network ` qcd and the deep structure of elementary particles ' , contract fmrx - ct98 - 0194 ( dg 12-miht ) for support .
9 x. ji , j.phys .
* g24 * ( 1998 ) 1181 .
a. belitsky , these proceedings .
d. martin et al . , preceding paper .
a. radyushkin , phys.lett .
* b449 * ( 1999 ) 81 .
a. shuvaev , hep - ph/9902318 .
a. shuvaev , k. golec - biernat , a.d .
martin and m.g .
ryskin , hep - ph/9902410 .
polyakov , c. weiss , hep - ph/9902451 .
m. glck , e. reya and a. vogt , eur .
j. * c5 * ( 1998 ) 461 .
musatov and a.v .
radyushkin , hep - ph/9905376 .
k. golec - biernat , a.d .
martin and m.g .
ryskin , hep - ph/9903327 . | we briefly discuss the problem of specifying initial conditions for evolution of off - diagonal ( skewed ) parton distributions .
we present numerical results to show that evolution rapidly washes out differences of input . |
the @xmath12 resonances with masses in the region around 4000 mev have stirred the hadron community with a series of states that challenge the common wisdom of mesons as made from @xmath6 .
there has been intense experimental work done at the babar , belle , cleo , bes and other collaborations , and many hopes are pinned in the role that the future fair facility with the panda collaboration and j - parc will play in this field .
there are early experimental reviews on the topic @xcite and more recent ones @xcite . from the theoretical point of view
there has also been an intensive activity trying to understand these intriguing states .
there are quark model pictures @xcite and explicit tetraquark structures @xcite . molecular interpretations
are given in refs .
much progress also has been done using the heavy quark spin symmetry ( hqss ) @xcite .
predictions using qcd sum rules have also brought some light into the issue @xcite .
strong decays of these resonances have been studied to learn about the nature of these states @xcite , while very often radiative decays are invoked as a tool to provide insight into this problem @xcite , although there might be exceptions as discussed in ref .
it has even been speculated that some states found near thresholds of two mesons could just be cusps , or threshold effects @xcite .
however , this speculation was challenged in ref .
@xcite which showed that the near threshold narrow structures can not be simply explained by kinematical threshold cusps in the corresponding elastic channels but require the presence of @xmath13-matrix poles . along
this latter point one should also mention a recent work calling the attention to possible effects of singularities on the opposite side of the unitary cut that enhance the cusp structure for states with mass above a threshold @xcite .
some theoretical reports on these issues can be found in refs .
@xcite .
on the other hand , and somewhat unexpected , recent experiments on the weak decays of @xmath14 meson are proving to be a powerful source of information on hadron dynamics and the nature of hadronic states @xcite .
one of the recent surprises was to see from these experiments a pronounced peak for the @xmath15 in @xmath16 decay into @xmath17 and @xmath18 @xcite while the signal for the @xmath19 was found very small or non - existent .
simultaneously , in the analogous decay of @xmath20 into @xmath17 and @xmath18 @xcite a signal was seen for the @xmath19 production and only a very small fraction was observed for the @xmath15 production .
this is surprising since the @xmath19 couples more strongly to @xmath18 than to the @xmath15 .
some attempt to explain this behaviour was offered in ref .
@xcite in terms of tetraquark structures for the scalar mesons .
furthermore , the @xmath19 and @xmath15 resonances are naturally explained as states dynamically generated from the meson - meson interaction in chiral unitary theory @xcite . from this perspective the features observed in these experiments and different ratios were well described in ref .
@xcite . the basic approach in ref .
@xcite was to identify the dominant mechanism of the decays at the quark level , implementing the hadronization of the final @xmath6 pair into two mesons and allowing them to interact to generate the @xmath19 and @xmath15 resonances .
the picture of ref .
@xcite has been extended to describe many other @xmath14 and @xmath21 decays , with equal success in the description of observed features of the reactions . in ref .
@xcite ratios for the production of vector mesons in the final states were evaluated and predictions for the @xmath22 decay were made . in ref .
@xcite the @xmath23 decays into @xmath24 and @xmath19 , @xmath15 , @xmath25 were studied . a different sort of states ,
dynamically generated from the vector - vector interaction were investigated in ref .
@xcite in the @xmath26 and @xmath27 decays into @xmath17 and @xmath28 . in ref .
@xcite the @xmath20 decay into @xmath23 and @xmath29 or @xmath19 , @xmath15 , @xmath25 and @xmath30 decay into @xmath23 and @xmath31 or @xmath32 were studied .
similarly , the @xmath16 decay into @xmath33 was studied with the aim of learning about @xmath34 scattering and the @xmath35 resonance @xcite .
related work to this latter one was done studying the semileptonic @xmath36 and @xmath14 decays in ref .
@xcite .
related problems of these weak decays have been addressed in refs .
@xcite from a very different point of view , evaluating microscopically the weak matrix elements and parametrizing parts of this interaction plus properties of the resonances produced , then carrying fits to data .
the aims are also usually different , having in mind the construction of full amplitudes that can be used for issues like @xmath37 violation . on the contrary , the work of ref . @xcite and the related ones make predictions for shapes and ratios that are tied to the final state interaction of mesons and , consequently , parameter free predictions
can be made for these observables .
so far , in the study of these @xmath14 decays the production of @xmath12 states has not yet been addressed and the aim of the present paper is to study reactions where these states can be produced , evaluating ratios for different decay modes and estimating the absolute rates .
this should stimulate experimental work that can shed light on the nature of some of these controversial states .
following refs .
@xcite , we plot in fig .
[ fig:1 ] the basic mechanism at the quark level for @xmath38 decay into a final @xmath39 and another @xmath40 pair .
( a ) and @xmath20 ( b ) decays into @xmath41 and a @xmath40 pair.,title="fig : " ] ( a ) and @xmath20 ( b ) decays into @xmath41 and a @xmath40 pair.,title="fig : " ] in ref .
@xcite the @xmath41 went into the production of a @xmath17 and the @xmath42 or @xmath43 were hadronized to produce two mesons which were allowed to interact to produce some resonant states . here
, we shall follow a different strategy and allow the @xmath41 to hadronize into two vector mesons , while the @xmath42 and @xmath43 will make the @xmath44 and @xmath45 mesons respectively .
let us observe that , apart for the @xmath46 transition , most favored for the decay , we have selected an @xmath47 in the final state which makes the @xmath48 transition cabibbo allowed .
this choice magnifies the decay rate , which should then be of the same order of magnitude as the @xmath49 , which also had the same diagram of the quark level prior to the hadronization of the @xmath42 to produce two mesons , in this case @xmath50 that couples later to the @xmath15 .
the next step consists in introducing a new @xmath40 state with the quantum numbers of the vacuum , @xmath51 , and see which combinations of mesons appear when added to @xmath41 .
this is depicted in fig .
[ fig:2 ] .
an easy way to see which vector mesons are produced in the hadronization of @xmath41 is to introduce the @xmath40 matrix @xmath52 note that this matrix corresponds to the su(4 ) vector matrix @xmath53 now we see that @xcite @xmath54 hence , we can write @xmath55 and @xmath56 note that we have produced an @xmath57 combination , as it should be coming from @xmath41 and the strong interaction hadronization , since we have the isospin doublet ( @xmath58 ) , ( @xmath59 ) . the @xmath60 component is energetically forbidden and hence we can write @xmath61 we can reach the result of eq .
( [ eq : vv44 - 1 ] ) in a simpler way neglecting the @xmath62 component in eq .
( [ eq : mm44 ] ) . since @xmath39 is an su(3 )
singlet and so is the @xmath63 combination , then the @xmath10 combination of eq .
( [ eq : vv44 - 2 ] ) is also an su(3 ) singlet .
using the full formalism has the advantage that it already tells one the phase convention of the states , which should be taken consistently with the chiral study of the interaction based on the matrix of eq .
( [ eq : vmatrix ] ) .
following the philosophy of ref .
@xcite we shall now let these vector mesons undergo interaction and we can connect with the work of ref .
@xcite , where using an extension of the local hidden gauge approach @xcite some @xmath12 states were dynamically generated . concretely , in ref .
@xcite four resonances were found , that we summarize in table [ tab : xyzstate ] together with the channel to which the resonance couples most strongly , and the experimental state to which they were associated .
.states found in ref .
@xcite , the channel to which they couple most strongly , and the experimental states to which they are associated ( see also refs .
@xmath64 is a predicted resonance . [
cols="^,^,^,^,^,^ " , ] in ref .
@xcite , another state with @xmath65 was found , but this one can not be produced with the hadronization of @xmath41 .
some of these resonances have also been claimed to be of @xmath10 or @xmath66 molecular nature in refs .
@xcite using for it the weinberg compositeness condition @xcite and in refs .
@xcite using qcd sum rules . also using hqss the same conclusions
are reached in ref . @xcite and with phenomenological potentials in ref .
. the final state interaction of the @xmath10 and @xmath66 proceeds diagrammatically as depicted in fig .
[ fig:3 ] . starting from eq .
( [ eq : vv44 - 2 ] ) the analytical expression for the formation of the resonance @xmath2 is given by @xmath67 where @xmath68 is the loop function of the two intermediate meson propagators and @xmath69 is the coupling of the resonance to the @xmath70 meson pair .
the formalism for @xmath0 runs parallel since the hadronization procedure is identical , coming from the @xmath41 , only the final state of @xmath40 is the @xmath45 rather than the @xmath44 .
hence , the matrix element is identical to the one of @xmath1 , only the kinematics to different masses will change .
there is one last point to consider which is the angular momentum conservation . for @xmath71 , we have the transition @xmath72 .
parity is not conserved but the angular momentum is . by choosing the lowest orbital momentum @xmath73 , we see that @xmath74 for @xmath75 and @xmath76 for @xmath77
. however , the dynamics will be different for @xmath78 .
this means that we can relate @xmath79 with @xmath80 , @xmath81 with @xmath82 , @xmath83 with @xmath84 and @xmath85 with @xmath86 , but in addition we can relate @xmath81 with @xmath83 , and the same for @xmath82 with @xmath84 .
hence in this latter case we also have a @xmath87 state for both resonances and the only difference between them is the different coupling to @xmath10 and @xmath66 , where the @xmath5 couples mostly to @xmath10 , while the @xmath3 couples mostly to @xmath66 . the partial decay width of these transitions is given by @xmath88 which allows us to obtain the following ratios , where the different unknown constants @xmath89 , that summarize the production amplitude at tree level , cancel in the ratios : @xmath90 and @xmath91 where @xmath92 , @xmath93 , @xmath94 and @xmath95 are the @xmath4 , @xmath64 , @xmath5 and @xmath3 , respectively .
the couplings @xmath69 and the loop functions @xmath68 in eq .
( [ eq : amplitut ] ) are taken from ref .
@xcite , where the dimensional regularization was used to deal with the divergence of @xmath68 , fixing the regularization scale @xmath96 mev and the subtraction constant @xmath97 . however , we have taken advantage to make corrections to the work of ref .
@xcite due to the findings of ref .
@xcite concerning heavy quark spin symmetry .
it was found there that a factor @xmath98 has to be implemented in the hidden gauge coupling @xmath99 in order to account for the @xmath100 decay .
however , this factor should not be implemented in the weinberg - tomozawa terms ( coming from exchange of vector mesons ) because these terms automatically implement this factor in the vertices of vector exchange .
however in the work of ref .
@xcite the coupling needed for @xmath100 had been used also in the weinberg - tomozawa term .
the bindings were obtained by fitting subtraction constants in the @xmath101 function .
we also do the same now , but with the present reduced interaction the @xmath101 function becomes more negative .
we use now @xmath102 mev and @xmath103 . in eqs .
( [ eq : ratio-1])-([eq : ratio-2 ] ) we summarize the results that we obtain , @xmath104 and @xmath105 as we can see , all the ratios are of the order of unity , however , the simplicity of these results should not be confused with a triviality .
the ratios close to unity for the @xmath44 or @xmath31 production are linked to the fact that the resonances are dynamically generated from @xmath10 and @xmath66 , which are produced by the hadronization of the @xmath41 pair .
the ratio for the @xmath106 is even more subtle since it is linked to the particular couplings of these resonances to @xmath10 and @xmath66 , which are a consequence of the dynamics that generates these states .
actually , the ratios @xmath107 are based only on phase space and result from the elementary mechanisms of fig [ fig:1 ] .
we would get the same ratios as far as the resonances are @xmath39 based .
hence , even of these ratios do not prove the molecular nature of the resonances , they already provide valuable information telling us that they are @xmath39 based .
the ratio @xmath108 provides more information since it involves two independent resonances and it is not just a phase space ratio .
if we take into account only phase space then @xmath109 instead of the value @xmath110 that we obtain . as for the absolute rates
, we can establish an analogy to the @xmath111 decay , since the amplitudes of the quark level prior to hadronization are identical and both processes require hadronization of a @xmath40 pair into resonances .
hence , we estimate the branching ratio for the production of these states of the order of @xmath7 @xcite . another estimate can be done starting from @xmath112 , which has a branching ratio of @xmath113 and reducing this rate by one order of magnitude which is the reduction factor that is found for hadronization in ref .
@xcite . in both cases
we find branching fractions of the order of @xmath7 , which are an order of magnitude bigger than many rates of the order of @xmath114 already catalogued in the pdg @xcite .
given the fact that the ratios @xmath107 obtained are not determining the molecular nature of the resonances , but only on the fact that they are @xmath115 based , we propose a complementary test in the next section .
in this section we propose a test that is linked to the molecular nature of the resonances .
we study the decay @xmath8 or @xmath9 close to the @xmath10 and @xmath11 thresholds .
let us now look to the process @xmath8 depicted in fig .
. the production matrix for this process will be given by @xmath116 where @xmath117 and @xmath118 stands for the @xmath119 and @xmath120 channels respectively .
the differential cross section for production will be given by @xcite @xmath121 where @xmath122 is the @xmath44 momentum in the @xmath30 rest frame and @xmath123 the @xmath124 momentum in the @xmath10 rest frame . by comparing this equation with eq .
( [ eq : width ] ) for the coalescence production of the resonance in @xmath125 , we find @xmath126 where we have divided the ratio of widths by the phase space factor @xmath127 and multiplied by @xmath128 to get a constant value at threshold and a dimensionless magnitude
. we would apply this method for the three resonances that couple strongly to @xmath119 ( see table [ tab : xyzstate ] ) . in the case of the resonance @xmath129 with @xmath130 that couples mostly to the @xmath120 channel ( see table [ tab : xyzstate ] ) we look instead for the production of @xmath120 . for the @xmath11 production we have @xmath131 and we use eq .
( [ eq : rg1 ] ) but with @xmath120 instead of @xmath119 in the final state .
we have evaluated eq .
( [ eq : rg1 ] ) using the scattering matrices obtained in ref .
@xcite modified as discussed above , and then eq .
( [ eq : tmat ] ) and eq .
( [ eq : tmat2 ] ) .
the results are shown in fig .
[ fig : rgam ] .
we can see that the ratios are different for each case and have some structure .
we observe that there is a fall down of the differential cross sections as a function of energy , as it would correspond to the tail of a resonance below threshold .
we should also note that in the case of @xmath119 , we have produced the @xmath57 combination .
if instead , one component like @xmath132 is observed , the rate should be multiplied by @xmath133 . in the case of @xmath120
there is a single component and the rate predicted is fine . in order to estimate uncertainties , we modify the model of ref .
@xcite by changing a bit the value of @xmath134 in the @xmath101 function expressions such that the values of the masses of the states change @xmath135 mev up or down .
the results are shown in fig .
[ fig : rgamj0 ] .
we find differences of the order of @xmath136% in the @xmath119 , @xmath120 mass distributions , which we accept as systematic uncertainties .
concerning the uncertainties of the ratios , since @xmath107 come from phase space , there are no changes for these ratios when using the second model .
the ratio @xmath108 is modified and we find now @xmath137 .
we have investigated the decays of @xmath0 and @xmath1 with r being the @xmath3 , @xmath4 , @xmath5 and a predicted @xmath138 resonances .
these decays have not been yet investigated .
we estimate them to have branching ratios of the order of @xmath7 .
we used a model in which these states are dynamically generated from the vector - vector interaction , most notably the @xmath10 and @xmath66 states . within this model
we could predict five ratios for the production of these states , where unknown dynamics of the production processes cancels in the ratios .
the procedure used a mechanism in which we have quark production at the elementary level , followed by hadronization of one final @xmath6 pair into two vectors and posterior final state interaction of this pair of vector mesons .
this mechanism has been tested successfully in many other decays and given us confidence on the fairness of the predictions made .
some ratios only tell us that the resonance is build from @xmath39 with or without extra hadronization . in order to offer some extra test for the molecular composition of these states
, we have evaluated the invariant mass distributions for the @xmath8 and @xmath9 close to threshold and have made predictions for these magnitudes relative to the width for the production of the resonances .
the experimental investigation of these decay modes , and comparison with the predictions made , would shed light on the nature of these resonances and we can only encourage the implementation of such experiments .
one of us , e. o. , wishes to acknowledge support from the chinese academy of science ( cas ) in the program of visiting professorship for senior international scientists .
this work is partly supported by the spanish ministerio de economia y competitividad and european feder funds under the contract number fis2011 - 28853-c02 - 01 and fis2011 - 28853-c02 - 02 , and the generalitat valenciana in the program prometeo ii-2014/068 .
we acknowledge the support of the european community - research infrastructure integrating activity study of strongly interacting matter ( acronym hadronphysics3 , grant agreement n. 283286 ) under the seventh framework programme of eu .
this work is also partly supported by the national natural science foundation of china under grant nos .
11165005 and 11475227 .
the project is sponsored by the scientific research foundation for the returned overseas chinese scholars , state education ministry and the american national science foundation , pif grant no .
phy 1415459 .
t. barnes , int .
j. mod .
a * 21 * , 5583 ( 2006 ) .
d. m. asner , t. barnes , j. m. bian , i. i. bigi , n. brambilla , i. r. boyko , v. bytev and k. t. chao _ et al .
_ , int .
j. mod .
a * 24 * , s1 ( 2009 ) .
s. godfrey and s. l. olsen , ann .
nucl . part .
sci . * 58 * , 51 ( 2008 ) .
s. cheng - ping [ belle collaboration ] , chin .
c * 34 * , 615 ( 2010 ) .
s. l. olsen , prog .
* 193 * , 38 ( 2012 ) .
s. lange _ et al . _ [ panda collaboration ] , arxiv:1311.7597 [ hep - ex ] .
x. liu , chin .
bull . * 59 * , 3815 ( 2014 ) .
y. p. guo , epj web conf . * 81 * , 01008 ( 2014 ) .
s. l. olsen , front .
* 10 * , 101401 ( 2015 ) .
g. ortega , d. r. entem and f. fernandez , j. phys .
g * 40 * , 065107 ( 2013 ) .
j. vijande and a. valcarce , phys .
b * 736 * , 325 ( 2014 ) .
a. esposito , a. l. guerrieri , f. piccinini , a. pilloni and a. d. polosa , int .
j. mod .
a * 30 * , no .
04n05 , 1530002 ( 2014 ) t. branz , t. gutsche and v. e. lyubovitskij , phys .
d * 80 * , 054019 ( 2009 ) .
t. branz , t. gutsche and v. e. lyubovitskij , phys .
d * 82 * , 054025 ( 2010 ) .
y. c. yang , z. xia and j. ping , phys .
d * 81 * , 094003 ( 2010 ) .
y. dong , a. faessler , t. gutsche and v. e. lyubovitskij , j. phys .
g * 40 * , 015002 ( 2013 ) .
q. wang , c. hanhart and q. zhao , phys .
* 111 * , no . 13 , 132003 ( 2013 ) .
m. cleven , q. wang , f. k. guo , c. hanhart , u .- g .
meiner and q. zhao , phys .
d * 87 * , no . 7 , 074006 ( 2013 ) .
q. wang , m. cleven , f. k. guo , c. hanhart , u. g. mei ner , x. g. wu and q. zhao , phys .
d * 89 * , no . 3 , 034001 ( 2014 ) . f. aceti , m. bayar , j. m. dias and e. oset , eur .
phys . j. a * 50 * , 103 ( 2014 ) .
f. aceti , m. bayar , e. oset , a. martinez torres , k. p. khemchandani , j. m. dias , f. s. navarra and m. nielsen , phys . rev .
d * 90 * , no . 1 ,
016003 ( 2014 ) .
j. nieves and m. p. valderrama , phys .
d * 86 * , 056004 ( 2012 ) .
m. nielsen , f. s. navarra and s. h. lee , phys .
rept . * 497 * , 41 ( 2010 ) .
k. p. khemchandani , a. martinez torres , m. nielsen and f. s. navarra , phys .
d * 89 * , no . 1 , 014029 ( 2014 ) . m. nielsen and f. s. navarra , mod .
a * 29 * , no .
5 , 1430005 ( 2014 ) .
y. dong , a. faessler , t. gutsche and v. e. lyubovitskij , phys .
d * 88 * , no . 1 , 014030 ( 2013 ) .
t. branz , t. gutsche and v. e. lyubovitskij , phys .
d * 82 * , 054010 ( 2010 ) .
f. aceti , r. molina and e. oset , phys .
d * 86 * , 113007 ( 2012 ) .
l. ma , z. f. sun , x. h. liu , w. z. deng , x. liu and s. l. zhu , phys .
d * 90 * , no . 3 , 034020 ( 2014 ) .
y. dong , a. faessler , t. gutsche and v. e. lyubovitskij , phys .
d * 90 * , no . 7 , 074032 ( 2014 ) . f. k. guo , c. hanhart , y. s. kalashnikova , u .- g .
meiner and a. v. nefediev , phys .
b * 742 * , 394 ( 2015 ) e. s. swanson , phys . rev .
d * 91 * , no . 3 , 034009 ( 2015 ) s. l. zhu , int .
j. mod .
e * 17 * , 283 ( 2008 ) .
n. brambilla , s. eidelman , b. k. heltsley , r. vogt , g. t. bodwin , e. eichten , a. d. frawley and a. b. meyer _ et al .
. phys .
j. c * 71 * , 1534 ( 2011 ) .
t. aaltonen _ et al . _
[ cdf collaboration ] , phys .
d * 84 * , 052012 ( 2011 ) . v. m. abazov _ et al .
_ [ d0 collaboration ] , phys .
d * 85 * , 011103 ( 2012 ) .
r. aaij _ et al . _
[ lhcb collaboration ] , phys .
d * 87 * , 052001 ( 2013 ) .
r. aaij _ et al . _
[ lhcb collaboration ] , phys .
d * 90 * , 012003 ( 2014 ) .
s. stone and l. zhang , phys .
* 111 * , 062001 ( 2013 ) .
j. a. oller and e. oset , nucl .
a * 620 * , 438 ( 1997 ) [ erratum - ibid . a * 652 * , 407 ( 1999 ) ] .
j. a. oller , e. oset and j. r. pelaez , phys .
d * 59 * , 074001 ( 1999 ) [ erratum - ibid .
d * 60 * , 099906 ( 1999 ) ] [ erratum - ibid .
d * 75 * , 099903 ( 2007 ) ] .
n. kaiser , eur .
j. a * 3 * , 307 ( 1998 ) . m. p. locher , v. e. markushin and h. q. zheng , eur .
j. c * 4 * , 317 ( 1998 ) .
j. nieves and e. ruiz arriola , nucl .
a * 679 * , 57 ( 2000 ) .
j. r. pelaez and g. rios , phys .
lett . * 97 * , 242002 ( 2006 ) .
w. h. liang and e. oset , phys .
b * 737 * , 70 ( 2014 ) .
m. bayar , w. h. liang and e. oset , phys .
d * 90 * , no .
11 , 114004 ( 2014 ) .
j. j. xie , l. r. dai and e. oset , phys .
b * 742 * , 363 ( 2015 ) j. j. xie and e. oset , phys .
d * 90 * , no .
9 , 094006 ( 2014 ) . w. h. liang , j. j. xie and e. oset , arxiv:1501.00088 [ hep - ph ] .
m. albaladejo , m. nielsen and e. oset , arxiv:1501.03455 [ hep - ph ] , phys .
b in print .
f. s. navarra , m. nielsen , e. oset and t. sekihara , arxiv:1501.03422 [ hep - ph ] .
a. furman , r. kaminski , l. lesniak and b. loiseau , phys .
b * 622 * , 207 ( 2005 ) .
b. el - bennich , a. furman , r. kaminski , l. lesniak and b. loiseau , phys .
d * 74 * , 114009 ( 2006 ) .
h. -y .
cheng , c. -k .
chua and k. -c .
yang , phys .
d * 73 * , 014017 ( 2006 ) .
d. delepine , j. l. lucio m. and c. a. ramirez , eur .
j. c * 45 * , 693 ( 2006 ) .
b. el - bennich , a. furman , r. kaminski , l. lesniak , b. loiseau and b. moussallam , phys .
d * 79 * , 094005 ( 2009 ) ; [ erratum - ibid .
d * 83 * , 039903 ( 2011 ) ] .
o. leitner , j .-
dedonder , b. loiseau and r. kaminski , phys .
d * 81 * , 094033 ( 2010 ) ; [ erratum - ibid .
d * 82 * , 119906 ( 2010 ) ] .
a. martinez torres , l. s. geng , l. r. dai , b. x. sun , e. oset and b. s. zou , phys .
b * 680 * , 310 ( 2009 ) [ arxiv:0906.2963 [ nucl - th ] ] .
r. molina and e. oset , phys .
d * 80 * , 114013 ( 2009 ) [ arxiv:0907.3043 [ hep - ph ] ] .
bando , t. kugo , s. uehara , k. yamawaki and t. yanagida , phys .
lett . * 54 * , 1215 ( 1985 ) .
m. bando , t. kugo and k. yamawaki , phys .
rept . * 164 * , 217 ( 1988 ) .
m. harada and k. yamawaki , phys .
rept . * 381 * , 1 ( 2003 ) . u .-
meiner , phys .
rept . * 161 * , 213 ( 1988 ) .
s. godfrey and s. l. olsen , ann .
nucl . part .
58 * , 51 ( 2008 ) .
k. a. olive _ et al .
_ [ particle data group collaboration ] , chin .
c * 38 * , 090001 ( 2014 ) .
k. abe _ et al .
_ [ belle collaboration ] , phys .
* 94 * , 182002 ( 2005 ) .
s. uehara _ et al .
_ [ belle collaboration ] , phys .
lett . * 96 * , 082003 ( 2006 ) .
p. pakhlov _ et al .
_ [ belle collaboration ] , phys .
lett . * 100 * , 202001 ( 2008 ) .
t. branz , t. gutsche and v. e. lyubovitskij , phys .
d * 82 * , 054010 ( 2010 ) .
t. gutsche , t. branz , a. faessler , i. w. lee and v. e. lyubovitskij , chin .
c * 34 * , no .
9 , 1185 ( 2010 ) .
t. branz , t. gutsche and v. e. lyubovitskij , phys .
d * 80 * , 054019 ( 2009 ) .
s. weinberg , phys .
* 137 * , b672 ( 1965 ) .
v. baru , j. haidenbauer , c. hanhart , y. kalashnikova and a. e. kudryavtsev , phys .
b * 586 * , 53 ( 2004 ) .
t. sekihara , t. hyodo and d. jido , arxiv:1411.2308 [ hep - ph ] .
k. p. khemchandani , a. martinez torres , m. nielsen and f. s. navarra , phys .
d * 89 * , no . 1 , 014029 ( 2014 ) . z. g. wang , eur .
j. c * 74 * , no . 7 , 2963 ( 2014 ) .
m. nielsen , f. s. navarra and s. h. lee , phys .
rept . * 497 * , 41 ( 2010 ) . c. hidalgo - duque ,
j. nieves and m. p. valderrama , phys .
d * 87 * , no .
7 , 076006 ( 2013 ) . | we investigate the decay of @xmath0 and @xmath1 with @xmath2 being the @xmath3 , @xmath4 , @xmath5 resonances . under the assumption that these states are dynamically generated from the vector - vector interaction ,
as has been concluded from several theoretical studies , we use a reaction mechanism of quark production at the elementary level , followed by hadronization of one final @xmath6 pair into two vectors and posterior final state interaction of this pair of vector mesons to produce the resonances . with this procedure
we are able to predict five ratios for these decays , which are closely linked to the dynamical nature of these states , and also predict the order of magnitude of the branching ratios which we find of the order of @xmath7 , well within the present measurable range . in order to further test the dynamical nature of these resonances we study
the @xmath8 and @xmath9 decays close to the @xmath10 and @xmath11 thresholds and make predictions for the ratio of the mass distributions in these decays and the @xmath1 decay widths .
the measurement of these decays rates can help unravel the nature of these resonances . |
convective stars of various types tend to have small - scale , time - varying magnetic fields , which is interpreted as dynamo action driven by differential rotation and convection / buoyancy instabilities .
in contrast , intermediate - mass main - sequence stars ( @xmath0 ) , which have a small convective core and a radiative envelope , have either no detectable field or a large - scale , steady magnetic fields the ap stars ( see e.g. aurire et al .
many of these stars have roughly dipolar fields but many have rather more complex fields , such as @xmath1 sco ( donati et al .
2006 ) . given the lack of necessary dynamo ingredients , we infer that the magnetic fields are ` fossil ' remnants in some stable equilibrium , formed from the magnetic field left over from the convective / accretive protostellar phase . on the theoretical side ,
the emphasis has been on finding stable magnetic equilibria .
analytically one can produce an equilibrium configuration simply by making sure all forces are balanced , and then test its stability using an energy method ( bernstein et al .
unfortunately , it has been easier to demonstrate the _ instability _ of various equilibria than to find _ stable _ configurations .
for instance , both axisymmetric poloidal fields and axisymmetric toroidal fields have been shown to be unstable ( wright 1973 ; tayler 1973 ; braithwaite 2006 ) .
more recently , it has become possible to find stable equilibria using numerical methods .
braithwaite & nordlund ( 2006 , see also braithwaite & spruit 2004 ) found such a configuration by evolving an arbitrary initial magnetic field in time and watching it find its way into a stable equilibrium .
this equilibrium is roughly axisymmetric , consisting of both toroidal and poloidal components ( logical , since both are unstable on their own ) in a twisted - torus configuration . from the outside , the field looks approximately dipolar , as do the fields of many ap stars .
here , i demonstrate the existence of more complex , non - dipolar configurations .
i first describe simulations of magnetic equilibria evolving from arbitrary initial conditions , then use simple analytic methods to explore the nature and stability of these equilibria .
the setup of the simulations is described in detail in braithwaite & nordlund ( 2006 ) ; i give a brief summary here .
the code used is the stagger code ( gudiksen & nordlund 2005 ) , a high - order finite - difference cartesian mhd code . the star is modelled as a ball of self - gravitating ideal gas of radius @xmath2 arranged in a polytrope of index @xmath3 ( approximating an a star ) , surrounded by an atmosphere of very low electrical conductivity .
the initial magnetic field resembles that expected at the end of a period of convection , i.e. chaotic and small - scale , containing wavenumbers with a flat power spectrum up to @xmath4 ( set to @xmath5 here ) , tapered so that @xmath6 where @xmath7 is the gas density and @xmath8 is a free parameter .
note that if the star forms out of a uniformly magnetised cloud and the same fraction of flux is lost from all fluid elements , then we expect @xmath9 .
the evolution in time of the magnetic field is followed in simulations with various values of the tapering parameter @xmath8 : @xmath10 , @xmath11 , @xmath12 and @xmath13 . in all cases ,
a stable equilibrium is reached after a few alfvn crossing times , but both the geometry and energy of the equilibrium is found to vary .
two types of equilibrium are found : simple axisymmetric ( as in braithwaite & nordlund 2006 ) in the @xmath14 and @xmath9 cases , and more complex non - axisymmetric in the @xmath15 and @xmath16 cases . )
initial field .
bottom left and right ( viewed from either side of the star ) is the equilibrium reached in the @xmath16 case.,title="fig : " ] ) initial field .
bottom left and right ( viewed from either side of the star ) is the equilibrium reached in the @xmath16 case.,title="fig : " ] ) initial field .
bottom left and right ( viewed from either side of the star ) is the equilibrium reached in the @xmath16 case.,title="fig : " ] ) initial field .
bottom left and right ( viewed from either side of the star ) is the equilibrium reached in the @xmath16 case.,title="fig : " ] fig .
[ fig : vapor ] shows the initial conditions ( top - left ) , as well as both the @xmath14 ( top - right ) and @xmath16 ( bottom - left & -right ) runs after around @xmath17 alfvn crossing times . in the @xmath14 case ,
the configuration is roughly axisymmetric and contains both toroidal and poloidal components . when @xmath16 , a non - axisymmetric equilibrium is reached .
the configuration consists of twisted flux tubes below the surface of the star .
the flux tubes lie horizontally and meander around the star at some distance below its surface .
consider a twisted flux tube of length @xmath18 and width ( at the surface of the star ) @xmath19 whose axis lies at some constant radius in a star of radius @xmath2 .
the cross - section of such a flux tube is illustrated in figure [ fig : x - sec ] , as well as its axisymmetric relative .
it is assumed that @xmath20 .
the three components of the magnetic field are toroidal ( parallel to the length of the flux tube ) , latitudinal and radial ; average values are denoted respectively by @xmath21 , @xmath22 and @xmath23 .
the latitudinal and radial components are collectively referred to as the poloidal component . since the flux tube can not have any ends , it must be joined in a loop , which can be wrapped around the star in any fashion . and @xmath23 are indicated by arrows towards the left ; @xmath21 is directed into the page .
poloidal field lines are marked with arrows ; toroidal field is represented by shading . ]
it is shown in fig .
[ fig : x - sec ] that the toroidal component of the field is confined to the volume enclosed by the largest poloidal line closed within the star , and is zero outside this volume .
this has to be the case , because a poloidal line which crosses the stellar surface would ` unwind ' if it contained any toroidal component . in this process ,
toroidal field is redistributed along the whole of the poloidal line .
toroidal field in the atmosphere is then destroyed by reconnection on a short time - scale , and the toroidal field inside the star is redistributed along the poloidal line to where it has been lost .
this occurs on the alfvn time - scale . in this way ,
all toroidal field on the poloidal line is quickly transferred into and destroyed in the atmosphere .
the magnetic energy of the flux tube is @xmath24 , \label{eq : energy}\ ] ] where @xmath18 is the length of the tube ( on the surface of the star ) , @xmath25 is the angle subtended at the centre of the star .
the three terms on the right hand side represent the field energy in the three components .
the field energy outside the star can be ignored ( braithwaite 2008 , in prep . ) . allowing the length and width of the flux tube to change adiabatically , the tube will stretch or contract until @xmath26 , i.e. an equilibrium
is reached at this point , i.e. that the equilibrium is stable to stretching / contracting perturbations . ] .
since we are considering a magnetic field in a stably - stratified star where the magnetic energy density is very much less than the thermal , i.e. @xmath27 , we have the following restrictions on the displacements @xmath28 which take place during this adjustment : @xmath29 the consequence of these restrictions is that the volume of the flux tube stays constant since no matter can leave or enter it , which can be expressed by stating that the area of the flux tube on the stellar surface @xmath30 .
in addition , we have : @xmath31 which follow simply from flux - freezing . differentiating ( [ eq : energy ] ) with respect to @xmath25 and setting @xmath26 , and noting from the geometry that @xmath32
, we have @xmath33 therefore the adjustment to equilibrium consists in stretching or contracting until the toroidal and latitudinal components are roughly equal . the toroidal and poloidal fluxes in the tube are @xmath34 and @xmath35 respectively .
they are conserved during adiabatic changes , and using ( [ eq : btbl ] ) we have @xmath36 so that if we know the area @xmath37 and the ratio of toroidal to poloidal fluxes of a tube , we can calculate its equilibrium width @xmath19 and length @xmath18 .
the above arguments can be applied not only to the flux tube as a whole , but also to any small length @xmath38 of the tube with area @xmath39 . along
the length of the flux tube @xmath40 is constant , but @xmath41 can vary .
a single flux tube need not have uniform width at every point along its length , there is a local equilibrium giving @xmath25 in terms of the toroidal flux and the local value of @xmath23 ( remember from ( [ eq : ddalpha ] ) that @xmath23 is constant during adiabatic changes in @xmath25 ) . in a star with more than one flux tube
, neighbouring flux tubes must have their poloidal components in opposite senses to avoid boundary discontinuities . however , there is no constraint on the sense of the toroidal component , since the tubes are separated from each other by a region with zero toroidal component .
since magnetic helicity is essentially just the product of the poloidal and toroidal components , different tubes can have helicities of opposite sign , and configurations of zero net helicity are possible in principle .
this contradicts the hypothesis that an arbitrary field should evolve into the lowest energy state for its helicity , rather the field can find local energy minima at higher energies .
once an equilibrium has formed , it continues to evolve as a result of magnetic and thermal diffusivities ( @xmath42 and @xmath43 respectively ) .
magnetic diffusion ( a result of finite conductivity ) causes the field to evolve on a timescale @xmath44 where @xmath45 is the relevant length scale . for an a star ,
this is around @xmath46yr if @xmath47 .
thermal diffusion can also cause field evolution . to be in pressure and density equilibrium with its non - magnetised surroundings , a flux tube must be at a lower temperature than its surroundings .
heat diffuses into the tube , causing it to rise on a timescale of order @xmath48 , where @xmath49 is the ratio of thermal to magnetic pressure .
these effects are discussed in braithwaite 2008 ( in prep . ) ; it is likely that both are too slow to be important in a stars , except perhaps in stars with very complex , small - scale equilibria ( i.e. with smaller @xmath45 ) .
whether there is any observational evidence for field evolution in a stars is controversial . that we observe
both axisymmetric and non - axisymmetric fields in ap stars implies that initial conditions vary , and that probably the central concentration of the initial field varies .
the reason for this is unclear .
however , it is a common phenomenon that ap stars share with white dwarfs and neutron stars ( as is the enormous range in field strengths ) , while the apparent cutoff in field strengths below @xmath50 g ( aurire et al .
2008 ) is unique to a stars , so the two problems are probably unconnected .
in summary , i have performed numerical simulations of the formation of stable magnetic equilibria in a non - convective star , starting from turbulent initial conditions .
a sufficiently centrally - concentrated initial field evolves into an approximately axisymmetric equilibrium ; a more spread - out initial field evolves into a more complex , non - axisymmetric equilibrium consisting of twisted flux tubes meandering under the surface of the star . in this case , the adjustment to equilibrium consists in the stretching or contracting of these flux tubes until an energy minimum is reached .
still needed is a more quantitative understanding of the effect of initial conditions on the resulting equilibrium , and more work is also required on the quasi - static diffusive evolution of these equilibria .
the results are applicable to all predominantly non - convective stars : main sequence above @xmath51 , white dwarfs and neutron stars . | an arbitrary initial magnetic field in an a star evolves into a stable equilibrium .
simulations are presented of the formation of non - axisymmetric equilibria consisting of twisted flux tubes meandering under the surface of the star , and analytic arguments are given relating the stability and form of these equilibria .
these results may help to explain observations of ap stars with very non - dipolar fields .
this work is also applicable to other essentially non - convective stars such as white dwarfs and neutron stars . |
the main goal of the fermionic molecular dynamics model proposed by s. drod _ _ et al.__@xcite and b. caurier _ _ et al.__@xcite , and improved thereafter by h. feldmeier@xcite , is to modelise the dynamics of a fermions system where antisymetrisation is exactly taken into account .
the fmd model consists of a single slater determinant parametrised by a set of dynamical variational parameters @xmath0 $ ] : @xmath1 where the single particle states are chosen to be gaussians , in order to localise the particles in phase space and be able to deal with particle density fluctuations and clustering : @xmath2 the real parameters @xmath3 and @xmath4 are the dynamical phases for the evolution of the spin degree of freedom , @xmath5 is the isospin degree of freedom , and the complex parameters @xmath6 and @xmath7 are linked to the classical coordinates of nucleons .
indeed the expectation value of the position and the momentum for a single particle state reads : @xmath8 the dynamical evolution of each variational parameters is obtained imposing a minimisation of the action : @xmath9 where @xmath10 is the hamiltonian of the system .
this equation gives the fmd equation of motion for each variational parameter : @xmath11 where @xmath12 .
an important feature of the fmd model is that the set of dynamical variational parameters includes the widths of the single particle states .
this means that the quantum - mechanical spreading of a free wave packet can be reproduced by the model .
another important point is that the single slater approximation implies that the expectation value of the energy is given by a functional of the one body - density only , namely the hartree - fock energy @xmath13 $ ] .
the hilbert space is restricted respect to tdhf because of the gaussian wave packet constraint , meaning that fmd can be viewed as an approximation of tdhf . on the other side ,
it is interesting to remark that the fmd model gives the exact hamilton equations for classical particles , meaning that all the correlations and fluctuations are taken into account at the classical level .
since the expectation value of the energy for the fmd model can be written as a density functional , the lastest skyrme functionals as the sly parametrisations@xcite can be used , with appropriate restoration of galilean invariance@xcite .
our fmd model based on a skyrme functional can be tested comparing observables with hf calculations and experimental data . in the following
we show calculations of ground state properties and giant monopole resonance ( gmr ) frequencies . concerning ground states , the fmd variational principle coincides with the minimization of the energy @xmath14 .
then fmd becomes a static model where @xmath15 . to obtain the ground state properites
, we have used an iterative algorithm based on a gradient method of first order .
table 1 represents the energies of the ground states for some light even - even nuclei .
we can see that the results given by the fmd model with the sly4 parametrisation are comparable to full hf calculations , and that both are close to the experimental data .
.fmd ground state energy with the sly4 interaction for some even - even nuclei in comparison with results from the hartree - fock model , and experimental data .
[ cols="<,^,^,^",options="header " , ] a similar quality of results is obtained for mean square radii .
a discrepancy is however observed for @xmath16c where fmd underestimates the binding energy .
this can be tentatively understood observing that the fmd result has a pronounced @xmath17 structure ; similar states is experimentally observed as excited states .
another interesting observable is the gmr frequency , which is a stringent test of the compressibility properties of nuclear matter . to excite a giant resonances , at a given time
an exciting operator is applied to the system according to : @xmath18 where @xmath19 for the gmr , which characterizes the isotropic compression of the system at the initial time . in calculating the gmr for @xmath16c with the sly5 parametrisation
, we found that the frequency of the monopole is given by @xmath20mev , which is comparable to the experimental value @xmath21mev extracted by d.h .
et al._@xcite .
calculations for heavier systems and different multipolarities are in progress .
we have seen that the fmd model gives a good description of light systems at t=0mev , ie good ground state properties and compressibility .
we will propose now a method to extract the thermal properties at finite temperature .
the extraction of the phase diagram at finite temperature for a finite system needs the calculation of thermal averages of observable @xmath22 , as well as of the temperature @xmath23 .
if the system is ergodic , the thermal average is given by the time average @xmath24 .
to know the temperature of the system , this latter has to be coupled to a thermometer , if we want the microcanonical temperature , or to a thermal bath if we want the canonical temperature . in both cases
the coupling must be such that the total wave function is a tensor product between the nuclear system and the thermometer @xmath25 , where the total hamiltonian is given by : @xmath26 here , @xmath27 is the weak coupling between the two systems where the expectation values should respect the conditions @xmath28 and @xmath29 .
if the statistical ensemble is the microcanonical ensemble ( thermometer case ) , we should always respect the condition @xmath30 , while for the canonical ensemble ( coupling to a thermal bath ) we should have @xmath31 .
if the thermal eresponse of the thermometer ( or the thermal bath ) @xmath32 is known analytically , it can be used to extract the common temperature of the two systems at equilibrium from the calculated time average @xmath33 . in practical calculations ,
the nucleus is trapped in a potential , and the total hamiltonian for the nuclear system can be written as : @xmath34 for the trap we have chosen a @xmath35 potential such as @xmath36 where @xmath37 is the trap constant . the thermometer has been taken as an harmonic oscillator where @xmath38 , with thermal response @xmath39 .
the coupling between the two system has been chosen as a contact interaction @xmath40 .
even if the parameters @xmath37,@xmath41 and @xmath42 are chosen to optimise the thermalisation properties , the equilibration time is very long , of the order of 50000 fm / c .
for this reason we have made calculations only for a very small system , namely @xmath43he . applying the method explained above at different total energies , we have extracted the temperature as a function of the excitation energy per nucleon , the so - called caloric curve .
figures [ fig : calorique ] represent three calculations for a @xmath43he nucleus using an hybrid ensemble intermediate between the microcanonical and the canonical ensemble , with @xmath44 .
the caloric curves obtained with the sly4d and the sgii parametrisation , with and without coulomb interaction are presented .
it is important to remark that the excitation energy axis contains the constraining potential and as such it is not a direct measure of the deposited energy . at low excitation energies
the trend is not clear , and doubts may be shed on the ergodicity of the dynamics . at higher energy
the curves show the expected trend for a transition from an excited fermion system to a system of non - interacting classical particles .
this can be further confirmed looking at the energy distribution , which is bimodal in the transition region . comparing the three caloric curves
, we can see that that the choice of the nuclear interaction has small effects , and the same is true for the influence of coulomb .
calculations with heavier systems are in progress .
this work is supported by the nexen anr project . | a new fermionic molecular dynamics ( fmd ) model based on a skyrme functional is proposed in this paper . after introducing the basic formalism ,
some first applications to nuclear structure and nuclear thermodynamics are presented . |
although we are constantly learning new things about our galaxy , some major questions about its structure still remain unanswered . among them
is the composition of the galaxy halo . solving this mystery
has been the main goal of the large photometric surveys launched in the 1990s .
the proposal of @xcite has been realised by the surveys like macho @xcite , ogle @xcite , eros @xcite and moa @xcite .
these groups continuously monitored millions of stars in both magellanic clouds in order to detect the transient amplification of brightness caused by objects crossing the observer source line of sight in a phenomenon called gravitational microlensing @xcite .
the rate of such events depends on the number , mass and kinematics of objects populating the line of sight , in the disk and the halo of our galaxy and in the clouds .
the main hypothesis all the surveys have been testing is whether the halo of our galaxy contains dark matter in the form of the massive compact halo objects , called machos .
if the galactic halo were entirely made of machos , with masses between @xmath4 and @xmath5 , in a project like ogle ii we should detect about five microlensing events per year towards the large magellanic cloud ( lmc ) , and the corresponding optical depth should be @xmath6 ( @xcite ) .
however , the actual optical depth measurements reported by the microlensing experiments are significantly smaller . the macho collaboration presented a sample of 10 microlensing event candidates towards the lmc and concluded that the fraction of dark matter in machos @xmath7/@xmath8 20 per cent , suggesting the existence of a new , previously unknown population of objects with masses @xmath9 ( @xcite , @xcite ) .
their estimate of the optical depth towards the lmc is @xmath10 .
however , the eros group reported a null result and derived an upper limit on the optical depth due to machos of @xmath11 , which implies @xmath12 per cent ( @xcite ) .
macho s sample of events is likely to have some contamination with variable stars , novae , and supernovae @xcite , and additional data indeed shows the presence of small contamination @xcite .
however , the optical depth we quote from macho has been corrected for this contamination @xcite .
the large discrepancy between the results may be due to the different samples of stars used by these groups or the fact that they probe different regions of the lmc .
however , it is unclear whether these fully explain the discrepancy .
independent searches for machos have also been carried out towards the andromeda galaxy ( m31 ) using pixel - lensing method .
for example , @xcite found 6 events and estimated the machos halo mass fraction to be higher than 20 per cent . on the other hand
, @xcite presented an automatic procedure that selects unambiguous microlensing events in a pixel light curve data set with a total of only 3 `` gold - plated '' candidates , thus emphasizing the lack of large numbers of clear - cut microlensing events towards the m31 .
@xcite showed 14 events towards m31 and noticed that the observed event rate is consistent with the self - lensing predictions , however their results were still inconclusive , mainly due to low statistics .
in this paper we present the results of the search for microlensing events towards the large magellanic cloud in the independent data set gathered by the optical gravitational lensing experiment ( ogle ) in its second phase during the years from 1996 to 2000 .
the paper is organized as follows .
first , the observational data are described along with the error bar correction procedure .
next , the event search algorithm and its results are presented .
it is followed by the description of the analysis of blending and its impact on the optical depth measurement and the event s detection efficiency .
finally , the optical depth is calculated and the results are discussed .
the data used in this paper were collected during the second phase of the ogle survey , in the years 1996 2000 towards the large magellanic cloud ( lmc ) .
the ogle project used its own dedicated warsaw telescope in the las campanas observatory , chile .
details of the instrumentation for ogle ii can be found in @xcite . in brief , a 2k@xmath132k ccd with a pixel size of 0.417 arcsec was operated in the drift - scan mode , giving an actual size for each observed field of 2k@xmath138k .
there were 21 fields observed towards the lmc covering a total of 4.72 square degrees .
their locations are shown in fig .
[ fig : fields ] .
fields are listed in table [ tab : fields ] , which contains the coordinates of centers of the fields , the number of good objects in the @xmath14band , the blending - corrected number of stars ( see section [ sec : blending ] ) and the blending density group . by good
, we mean all objects having at least 80 epochs during the entire time span of the ogle ii ( from about 15 to 25 per cent of all collected frames of a field ) and mean magnitude brighter than @xmath15 mag .
the limiting magnitude was chosen to be at the peak of the observed luminosity function .
the images in the @xmath16 and @xmath17 bands were de - biased and flat - field corrected `` on - the - fly '' .
the photometric pipeline was based on the difference image analysis method ( dia , @xcite ) .
photometric databases were created for both pass - bands as described in @xcite and are available on - line . the ogle
ii observations of the lmc began in december 1996 ( hjd=2450446 ) , but during the first year only the very central fields ( i.e. those that are , in general , denser ) were observed ( fields 110 and 12 ) . in november 1997 ( hjd=2450726 ) , fields 11 and 1320 were added to the observing queue , and the addition of the field lmc_sc21 in january 1998 ( hjd=2450831 ) eventually formed the entire set of ogle ii fields .
all 21 fields were monitored continuously until november 2000 ( hjd=2451874 ) , yielding about 500 and 300 frames per field in the @xmath16 band for dense and sparse fields , respectively . in the @xmath17 band
, there were no more than 50 frames per field collected in total during the duration of the project . on average , each field
was observed every third night in the @xmath16 band , and every 11-th night in the @xmath17 band .
mean seeing was 1.35 arcsec in the @xmath16 and 1.37 arcsec in @xmath17 .
template images were created by stacking the best seeing images , resulting in images with seeing of about 1.1 arcsec .
the template images contained about 5.5 million objects suitable for our study in the @xmath16 and @xmath17 band .
the detection of microlensing events and the efficiency determination have been performed using @xmath14 band data only , as these were far more numerous and were sampled more frequently as compared to @xmath18band photometry . in this study
we also occasionally made use of ogle iii lmc data from years 20012008 , which covered a much wider area and included the ogle ii fields .
details of the instrumentation and photometric pipeline can be found in @xcite .
analysis of ogle iii lmc data will be presented separately in a forthcoming paper ( wyrzykowski et al .
, in preparation ) . additionally , for the two detected events
, we also use the data gathered by the macho group and provided on their www interface .
the reductions of the images and extraction of the photometry were performed with difference image analysis following @xcite .
cccrrcc field & @xmath19 & @xmath20 & & stellar density & density level + & & & template & estimated real & [ stars / sq.arcmin ] & + lmc_sc1 & 5:33:48.84 & -70:06:07.9 & 288029 & 668799 & 367 & dense + lmc_sc2 & 5:31:17.17 & -69:51:57.0 & 332267 & 771301 & 424 & dense + lmc_sc3 & 5:28:47.95 & -69:48:05.6 & 355075 & 822913 & 453 & dense + lmc_sc4 & 5:26:17.82 & -69:48:06.1 & 390484 & 904432 & 498 & dense + lmc_sc5 & 5:23:47.79 & -69:41:03.3 & 379709 & 878472 & 484 & dense + lmc_sc6 & 5:21:18.34 & -69:37:10.3 & 392781 & 909775 & 501 & dense + lmc_sc7 & 5:18:48.01 & -69:24:09.9 & 388918 & 900339 & 496 & dense + lmc_sc8 & 5:16:17.91 & -69:19:13.7 & 349202 & 807890 & 445 & dense + lmc_sc9 & 5:13:47.92 & -69:14:05.8 & 289391 & 671459 & 369 & dense + lmc_sc10 & 5:11:16.10 & -69:09:17.3 & 258706 & 599375 & 330 & dense + lmc_sc11 & 5:08:41.06 & -69:10:03.1 & 253395 & 588432 & 323 & dense + lmc_sc12 & 5:06:16.15 & -69:38:18.3 & 186951 & 318893 & 238 & sparse + lmc_sc13 & 5:06:14.18 & -68:43:30.3 & 214229 & 365917 & 273 & sparse + lmc_sc14 & 5:03:48.62 & -69:04:44.3 & 222569 & 380394 & 284 & sparse + lmc_sc15 & 5:01:17.27 & -69:04:42.9 & 170657 & 291859 & 218 & sparse + lmc_sc16 & 5:36:17.62 & -70:09:41.9 & 247981 & 575953 & 316 & dense
+ lmc_sc17 & 5:38:47.47 & -70:16:44.1 & 194661 & 332187 & 248 & sparse + lmc_sc18 & 5:41:17.77 & -70:24:48.0 & 166998 & 285126 & 213 & sparse + lmc_sc19 & 5:43:47.91 & -70:34:42.7 & 156353 & 267126 & 199 & sparse + lmc_sc20 & 5:46:17.85 & -70:44:50.0 & 152186 & 260515 & 194 & sparse + lmc_sc21 & 5:21:13.73 & -70:33:18.7 & 143979 & 244667 & 184 & sparse + total & & 5534521 & 11845824 & + coordinates point to the centre of the field , each being @xmath21 . the number of good objects in the template is provided ( number of data points @xmath22 and mean @xmath16-band magnitude @xmath23 mag ) together with the estimated number of real monitored stars ( see section [ sec : blending ] ) .
stellar density in number of stars per square arc minute was used to classify fields into _ dense _ or _ sparse _ classes with the threshold of 300 stars / sq.arcmin .
photometric error bars calculated using the difference image analysis method are known to be generally underestimated .
this effect is most pronounced for bright stars , as noted already by the authors of the dia method @xcite . in microlensing studies
this affects the brightest parts of events , increasing the @xmath24 values in microlensing model fitting . in order to increase the detectability of events
, we derived and applied an error bar correction factor . typically , error bars in microlensing surveys
are corrected by calculating one single correction factor for an entire light curve , derived at the event s baseline ( assuming the baseline is constant ) .
this approach , however , does not account for the need for varying correction factors for different apparent magnitudes , which in case of highly magnified events leaves the error bars at the peak of the event significantly underestimated . also , as shown in @xcite , some events exhibit intrinsic variability in their baselines due to variability of the microlensed source or one of the blended stars , in which case the derived photometric error bars are overestimated . in this study , we processed the entire ogle
ii data set and determined the correction coefficients in an empirical way . for each ogle ii field , we extracted the @xmath16-band light curves of stars and compared their intrinsic error - weighted _ rms _ with the mean error returned by the photometry pipeline . in order to produce statistically comparable results ,
in each light curve we selected only the first 250 measurements for further analysis , as different fields had different numbers of frames taken . in the plot of _ rms _ vs magnitude ( see fig .
[ fig : errors ] ) , the lowest boundary of the _ rms _ is occupied by the least variable ( i.e.,constant ) stars . on the other hand , the lowest outline of mean
error bar is defined by the errors assigned to the best measurements at given magnitude . in an ideal situation
these two curves should follow each other , i.e.,every constant star should have its _ rms _ equal to its mean error bar . as can be seen from the upper panel of fig .
[ fig : errors ] , for our data the mean errors were systematically lower than the _
rms_. .error correction coefficients for ogle ii lmc fields . [ cols="^,^,^",options="header " , ] + @xmath25 colour derived using macho @xmath26-band data +
out of @xmath27 light curves we found only two plausible candidates for microlensing events .
these are labeled ogle - lmc-01 and ogle - lmc-02 .
table [ tab : events ] presents their coordinates , ogle ii field , star i d , baseline @xmath14 and @xmath17-band magnitudes obtained in the microlensing fit and magnitude and colour of the source .
both candidate light curves were checked for available additional data outside the time - span of the ogle ii project .
we cross - correlated them with the ogle iii database , covering years 20012008 ( the most - recent data used were obtained in may 2008 ) and the macho database , covering years 19921999 , yielding total data coverage of 16 years . neither candidate exhibits additional variability apart from the actual event . in the ogle - ii database
there were 44 @xmath17-band observations of ogle - lmc-01 , but none of them were taken during the event . therefore for colour determination of the source , we used re - reduced macho data in the @xmath26 band . by comparing the ogle @xmath17 and @xmath16 cmd with macho @xmath26-band photometry of nearby stars , we found the following transformation for the colour : @xmath28 , where the measured constant is an artefact of the instrumental magnitude system of our reductions of the macho data . for ogle - lmc-02
, we used the ogle @xmath17-band data , as there were 59 observations available , with about 14 taken during the peak .
close - ups of the events are shown in fig .
[ fig : events ] along with the best - fitting microlensing models and their residuals .
lrrrrrr + parameter & & & + @xmath29 & 1337.20 & @xmath30 & 1337.20 & @xmath30 & 1337.20 & @xmath31 + & & & & & & + @xmath32 & 57.2 & @xmath33 & 59.2 & @xmath34 & 65.7 & @xmath35 + & & & & & & + @xmath36 & 0.0258 & @xmath37 & 0.0248 & @xmath38 & 0.0231 & @xmath39 + & & & & & & + @xmath40 & 19.91 & @xmath31 & 19.91 & @xmath31 & 19.92 & @xmath31 + & & & & & & + @xmath41 & 1.04 & @xmath42 & 1.0 & & 0.91 & @xmath43 + & & & & & & + @xmath44 & & & & & 18.28 & @xmath31 + & & & & & & + @xmath45 & & & & & 1.11 & @xmath46 + & & & & & & + @xmath24 & 286.7 & & 286.9 & & 3401.0 & + & & & & & & + @xmath47 & 0.88 & & 0.87 & & 2.68 & + + parameter & & & + @xmath29 & 491.6 & @xmath48 & 491.6 & @xmath48 & 491.3 & @xmath48 + & & & & & & + @xmath32 & 23.8 & @xmath49 & 24.2 & @xmath50 & 27.4 & @xmath51 + & & & & & & + @xmath36 & 0.4120 & @xmath52 & 0.4011 & @xmath53 & 0.3478 & @xmath54 + & & & & & & + @xmath40 & 20.42 & @xmath30 & 20.42 & @xmath30 & 20.44 & @xmath30 + & & & & & & + @xmath41 & 1.04 & @xmath55 & 1.0 & & 0.83 & @xmath56 + & & & & & & + @xmath57 & & & & & 20.68 & @xmath30 + & & & & & & + @xmath58 & & & & & 0.67 & @xmath59 + & & & & & & + @xmath24 & 578.2 & & 578.2 & & 636.2 & + & & & & & & + @xmath47 & 1.21 & & 1.20 & & 1.19 & + the top panel of fig .
[ fig : cmd ] shows the colour - magnitude diagram of lmc stars based on ogle and hst @xcite data , along with the determined positions of the sources in the two ogle candidates . for comparison ,
we have also included in this figure the candidates from the macho collaboration : different colours and shapes differentiate between self lensing candidates according to @xcite , the binary event macho - lmc-9 , the thick disk lens candidates macho - lmc-5 @xcite and macho - lmc-20 @xcite and the remaining candidates . in the bottom panel of fig .
[ fig : cmd ] , the positions of ogle ii and macho candidates are shown on the map of stellar density of the red clump stars from the ogle
iii data .
this was the only event detected towards the lmc by the early warning system ( ews , @xcite ) during the entire second phase of the ogle project and was designated ogle-1999-lmc-01ogle / ogle2/ews/1999/lmc-01.html ] .
it was also independently discovered and alerted by the macho collaboration ( macho-99-lmc-02 ) , but it was not included in @xcite as it occurred outside the 5.7 year time - span of analysed data ( macho star i d : 79.5863.4522 , see also @xcite ) . the ogle
ii data were fitted with the standard paczyski microlensing model , i.e.,a point - source point - lens model ( eq . [ eq : i ] ) . when we fixed blending such that the event was unblended ( @xmath60 ) we obtained @xmath61 days , whilst when @xmath62 was free , we found @xmath63 days ( @xmath64 ) . when ogle @xmath16- and macho @xmath26-band data were fitted simultaneously , the derived timescale was somewhat larger ( @xmath65 days ) , as the blending parameter in @xmath16-band changed slightly .
the results of the fits are summarised in table [ tab : models ] . since the event was relatively long we checked if ogle - lmc-01 was subject to the parallax effect ( e.g.,@xcite ) .
the components of the parallax vector @xmath66 in the ogle ii data fit were @xmath67 and @xmath68 , i.e.,consistent at 1-@xmath69 level with no detection of parallax@xcite .
therefore , the time - scale derived in the fit is not subject to any degeneracy with parallax . in principle
the lack of parallax signatures allows us to place limits on lens distance .
however , for this event the parallax constraints are not very tight and so the limits are not particularly insightful : if we assume that the source lies at the distance of the lmc ( @xmath70 ) and the lens is @xmath71 , we find that the 2-sigma lower - limit on the lens distance is 300 pc . from this , we can not make any significant statements about the population to which the lens belongs . with a maximum amplification of about 40 ( @xmath72 mag )
, this is one of the most secure and best constrained events towards the lmc found to date .
high amplification events are often prone to exhibit finite - source effects in cases where the source and the lens are close to each other .
we checked for the presence of this effect in the ogle and macho data .
in particular , macho data had two observations very close to the peak , where the effect should be most pronounced .
we have not detected any deviation ( see inset in fig . [
fig : events ] ) .
this only allows us to put an upper limit on the source star size in units of the einstein radii of @xmath73 , which along with an estimation of the source size ( @xmath74 ) lead to an upper limit on @xmath75 .
this at the lmc distance is equivalent to about @xmath76 au .
hence , the limit on the projected transverse velocity of the lensing object @xmath77 km / s , which is satisfied by typical lenses in both the lmc and galaxy halo . a simultaneous microlensing model fit to the ogle @xmath16-band and macho @xmath26-band data gives an apparent source magnitude of @xmath78 and colour @xmath79 . magnitudes and
location of the source on the cmd in fig .
[ fig : cmd ] indicate it belongs to the red giants branch of the lmc .
the second event , ogle - lmc-02 , was found just at the limit of our magnitude cut ( the actual mean brightness was 20.29 mag , which means that it passed cut 0 ) . the maximum amplification of @xmath802.6 produced a rise in brightness of @xmath801 mag .
this event occurred in the very first season of the lmc observations in the ogle
ii , when the sampling was much denser than in subsequent seasons .
therefore , the microlensing light curve was very well covered and prominent ( see fig . [
fig : events ] ) .
again , we used the standard point - source point - lens model to fit the ogle ii data . the unblended fit ( @xmath60 ) gave a time - scale @xmath81 days , and for unconstrained @xmath62 we found @xmath82 days ( @xmath83 ) .
the large error bars in @xmath62 and @xmath84 were also present in the simultaneous fit to @xmath16- and @xmath17-band observations , which yielded a time scale of @xmath85 days .
this object was also observed by the macho collaboration ( macho star i d : 77.8152.2917 ) .
however , the event was not detected by them , probably due to noisy data and sparse sampling .
the peak was clearly revealed in both red and blue filters only after dia re - reductions of the original macho images and its best - fitting microlensing model had a time - scale of around 25 days , consistent with the fit to the ogle data . from the fit to the two - band ogle data , we derived the magnitude and colour of the lensed source .
the position of the source with magnitude @xmath86 and colour @xmath87 , is marked on the fig .
[ fig : cmd ] and indicates it is probably a main sequence star located in the lmc .
its location on the main sequence branch of the cmd means that this may potentially be a candidate for a blue bumper .
however , while many blue bumpers exhibit secondary brightening episodes , the baseline of this event proved to be constant over more than 15 years .
therefore , we believe that it is unlikely to be a blue bumper . with @xmath88 at the innermost one .
dashed boxes mark ogle ii fields
. positions of the sources in the two ogle events are marked in red and with a number .
remaining symbols represent macho candidates with binary event # 9 ( green star ) , candidates for self lensing ( yellow squares ) , confirmed thick - disk lenses # 5 and # 20 ( dark blue filled squares ) and remaining marked with blue triangles.,title="fig:",width=321 ] + with @xmath88 at the innermost one .
dashed boxes mark ogle ii fields .
positions of the sources in the two ogle events are marked in red and with a number .
remaining symbols represent macho candidates with binary event # 9 ( green star ) , candidates for self lensing ( yellow squares ) , confirmed thick - disk lenses # 5 and # 20 ( dark blue filled squares ) and remaining marked with blue triangles.,title="fig:",width=321 ]
microlensing collaborations have deliberately chosen to monitor the densest fields on the sky , namely the galactic center and magellanic clouds .
this is done in order to increase the number of potentially detectable microlensing events , as the chances of seeing an events are of the order of @xmath89 ( i.e.,one out of 10 million monitored stars is undergoing a microlensing episode at any given time ) . however , in such crowded stellar fields observed with a medium - sized ground - based telescope , virtually every object detected on the image consists of several stars blended together .
neglecting blending can have severe consequences on the optical depth measurements ( e.g.,@xcite ) , as the amplifications and time - scales of microlensing events may be determined incorrectly , sometimes by a factor of 2 or more . on the other hand , @xcite showed that in the case of the galactic bulge the effect of overestimation of the events time - scales cancels out with the underestimate of the real number of monitored stars to lowest order . dealing with blending
is a delicate and difficult task . adding a blending parameter to the standard microlensing light curve model may cause degeneracies in the parameter space of the microlensing events ( e.g.,@xcite ) .
therefore , the favoured approach in recent optical depth determinations towards the galactic bulge and magellanic clouds is to neglect blending by narrowing the star sample to the brightest stars only ( @xcite , @xcite ) .
however , in the case of magellanic clouds the number of possible events is very small and it is even more unlikely to detect a microlensing event occurring in the limited number of bright stars . in our analysis , we decided to use all stars available in the ogle ii database . to deal with the blending issue which is most serious for faint stars
we make use of the archival images from the hubble space telescope ( hst ) located by chance in our fields .
the hst images have superior resolution when compared to the ground - based ogle images ( i.e. they are nearly blending - free ) and are usually much deeper than the ogle data . therefore these data are a valueable source of information regarding the amount of blending . in the hst archive , we identified two deep acs @xmath16-band images coinciding with the ogle fields , located in ogle fields lmc_sc6 and lmc_sc19 , as typical representatives of the dense and sparse parts of the lmc .
the location of these fields is shown in fig .
[ fig : fields ] .
the positions of the hst stars were transformed onto the ogle images and then we identified which hst stars contributed to each ogle object by calculating their distances from the ogle blend s centre .
a blend s radius was estimated using the gaussian profile for the given blend s magnitude at a chosen level above the background ( @xmath90 mag ) with an additional 1.3 pixel added in quadrature .
we identified hst stars that were not more than 3.5 mag fainter than the ogle blend ( i.e.,@xmath91 per cent ) and considered stars located within the blend radius as components of that blend .
this method allowed us to create an empirical distribution of the blending fractions for dense and sparse fields .
we divided these distributions into three magnitude bins : 14 [email protected] mag , 17.5 mag@xmath9219 mag and 19 [email protected] mag .
they are shown in fig .
[ fig : fs ] . then , every ogle lmc field was classified as dense or sparse according to its average stellar density down to 20.4 mag ( with 300 stars per square arcmin as a division boundary ) on the ogle template image ( see table [ tab : fields ] ) .
we then adopted the corresponding distributions in the detection efficiency calculations and in our recalculation of the number of monitored stars ( see section [ sec : eff ] ) . due to blending in ground - based observations
, we can not resolve individual stars that are separated by less than the typical seeing ( of order of 1 arcsec ) .
even though the ogle template images , with respect to the typical images , have usually high resolution thanks to stacking of @xmath80a dozen of the best - seeing images , we still have limited ability to resolve individual stars . since each ogle template object may consist of several stars , any of which could be microlensed , the number of stars we consider as potential sources for microlensing events is actually larger than number of objects detected on the template image . in order to estimate the actual number of monitored stars which enters directly into eq .
( [ eq : tau ] ) we derived the distributions of the number of hst stars hidden in a single ogle template object ( see fig .
[ fig : nhst ] ) using the same hst images and photometry .
then , for each ogle field we convolved its luminosity function with the distribution in each magnitude bin ( for a corresponding density level ) .
the estimated numbers of monitored stars in each field are provided in table [ tab : fields ] . on average ,
the correction factor for dense fields was about 2.3 and for sparse about 1.7 .
the total number of monitored stars was estimated to be about 11.8 million , compared to about 5.5 million objects detected on the ogle
ii template images in cut 0 .
it is worth pointing out that in our method of counting the total number of monitored stars we did not count all stars that would be observed with the hst down to 23.9 mag ( 20.4 mag + 3.5 mag ) .
we counted only those stars which were blended with other stars such that they formed a blend of magnitude brighter than 20.4 mag on the ogle template image .
for instance , a single 21 mag star might not be detected on the template image .
even if microlensed , we would not detect this event as in our analysis we did not search for transient events with no baseline ( i.e.,with no object on the template ) .
if such star , however , was blended with another 21 mag star they would form a 20.2 mag object very likely detectable on the template . in this case , both components would have blending factor of @xmath94 and both of them would be counted as a separate potential source for microlensing events .
the task of calculating the detection efficiency for microlensing events can be a complicated procedure . in general , this requires us to simulate events and then determine the fraction that we recover with our selection criteria .
it may seem that the most accurate method is to inject microlensing events into the series of real frames or to create completely new mock images and then put them through the entire photometric pipeline .
however , this is obviously very intensive computationally .
furthermore , it is not clear how to deal with the fact that any star within the blended object can be microlensed or whether the mock frame is indeed reproducing all characteristics of the real one .
some of these problems can be also encountered when simulating microlensing events at the so - called ` catalogue - level ' ( which only involves generating series of light curves ) .
however , the computational effort required for this procedure is significantly reduced and some difficulties can be resolved more easily .
the ` image - level ' simulations were performed by macho and moa collaborations in their calculation of the optical depth towards the galactic bulge @xcite .
the ` catalogue - level ' approach has been more commonly used for the bulge ( e.g.,@xcite ) and in the eros collaboration s study of microlensing towards the magellanic clouds @xcite . in this study
, we decided to use the existing photometric light curves as a base for the simulations .
however , as we discuss below , we have attempted to incorporate all of the advantages of the ` image - level ' simulations .
as shown in @xcite such an approach can be very successful when all factors are carefully taken into account .
the detection efficiency was determined individually for the fields containing both events ( sc2 and sc8 ) . for a wide range of time - scales between 1 and 1000 days
, we performed monte carlo simulation of about @xmath95 events . from the field s photometric database ,
a star was picked randomly , provided that the star satisfied the basic cut 0 ( i.e.,it had the required number of data points and sufficient mean brightness ) . for a given @xmath84 ,
the other microlensing parameters were drawn from flat distributions ( @xmath29 and @xmath36 ) and from the empirically derived distributions for the blending parameter ( @xmath62 ; see section [ sec : blending ] ) .
the latter was chosen with respect to the stellar density level of the simulated field and the magnitude bin in which the simulated star was located .
the original flux of the input star was apportioned between the source and blend in the ratio @xmath96 .
effectively , the lensed flux was added to already existing stars in the database , so no new stars were ` created ' . also , the procedure preserved any variability and non - gaussianity present in the original photometry , allowing for effects of decreasing or increasing of the variability amplitude during microlensing @xcite .
photometric errors - bars were scaled according to the empirically derived relation between the error bars and the magnitude ( see eq .
[ eq : simerr ] ) .
such simulated events were then analysed with our search pipeline described in section [ sec : search ] . mag .
lower panel : luminosity functions for the same fields recovered after applying the blending correction ( see section [ sec : blending ] ) .
their shapes follow the prototype lfs of two hst fields ( dense and sparse ) used for the blending correction ( thick dash - dotted lines ) .
, width=321 ] in their lmc data , the macho collaboration found one event ( macho - lmc-9 ) to be a clear caustic crossing binary lens .
our detection pipeline discriminated against binary and other exotic events , as we only fitted a point - source point - lens microlensing model .
however , we performed a visual inspection of the several thousands of high signal - to - noise light curves that passed cut 1 , and found no candidates for any non - standard type of events
. our detection efficiency did not take into account the influence of binary lenses because monte carlo simulations of binary events and their automated recovery is a difficult and computationally demanding task . therefore we followed the approach adopted by the eros group @xcite and reduced the efficiency by a factor of 0.9 to compensate for the 10 per cent possible binary lenses .
we performed our simulations on both bright and all stars samples with blending included .
in addition , we repeated the analysis assuming no blending ( @xmath97 @xmath98 ) in order to compare our results with other studies .
we defined the bright stars sample as stars with @xmath99 and brighter than @xmath100 mag , where @xmath101 is the red clump stars centre on the colour - magnitude diagram . in the reference field
lmc_sc1 , this threshold was at 18.8 mag .
the all stars sample comprised of all stars used in the search for microlensing events , i.e.,stars brighter than @xmath15 mag and with at least 80 data points .
[ fig : eff ] shows the derived detection efficiencies for a wide range of time - scales for a typical dense field .
most sparse fields were observed for a shorter period of time ( observations started about a year later than for dense fields ) .
however , in our detection efficiency calculations we simulated microlensing events within the full time - span of the ogle ii experiment , namely for @xmath29 within the range of @xmath102 . therefore , the efficiency for fields observed for a shorter time is reduced by about 30 per cent , owing to the events which fall outside the simulated light curve .
only one field classified as sparse ( lmc_sc12 ) was monitored since the very beginning of the project and the microlensing detection efficiency for this field is comparable to the efficiencies for dense fields .
[ fig : eff ] also shows the detection efficiencies for the non - blended analysis of all and bright stars samples ( with @xmath103 for all simulated events ) .
the efficiency for bright stars with and without blending is higher than the corresponding efficiencies for all stars sample , as it is generally easier to recover events occurring on brighter stars .
the observed luminosity functions ( lfs ) of dense and sparse ogle ii lmc fields are shown in the upper panel of fig .
[ fig : lfs ] .
blending is more severe in dense fields and it causes faint stars to ` merge ' and form brighter objects , populating brighter parts of the luminosity functions .
the bottom panel of fig .
[ fig : lfs ] shows blending - free luminosity functions for a dense and sparse field , recovered using the blending distributions obtained by comparing ogle and high - resolution hst images .
the luminosity functions obtained from hst are shown as well .
it is noticeable that the effect of over - populating the brighter parts and under - populating the faint end in blending - free lfs of dense fields has virtually disappeared after applying the correction .
the recovered dense and sparse luminosity functions follow their prototypes from the hst everywhere except at the bright and very faint end . for stars brighter than @xmath104 mag
the hst images become saturated , which is reflected in the steeper luminosity functions at the bright end .
the large scatter in the bright range is due to low number statistics in the hst s small field - of - view ( as compared to ogle ii fields ) .
the broadening of the red clump may also be caused by this effect , or could be due to the distance gradient along the ogle field .
the hst s lfs of dense and sparse areas differ significantly , which is likely caused by the different mix of lmc bar and disk populations in these two fields . in our monte carlo simulations
, we drew an object from the original ogle luminosity functions .
therefore we were not affected by the incompleteness of the bright end of the hst luminosity functions .
incompleteness in the faint end of the ogle luminosity functions is partially recovered in the correction process , but also is negligible as the sensitivity for such faint stars is very low .
although our blending analysis successfully recovers the underlying stellar brightness distribution , it is still approximate as we used only two hst images to represent dense and sparse fields . in future
, it would be interesting to expand this study with a detailed @xmath14band high - resolution follow - up , using hst and/or other high - resolution instruments .
we determined the optical depth using the all stars sample .
for the two events found in the ogle ii data we used the following standard equation for calculating the optical depth : @xmath105 where @xmath106 days is the time - span of all observations , @xmath107 is the total number of monitored stars ( see section [ sec : blending ] ) , @xmath108 is the total number of events , @xmath109 is the time - scale of each event detected with the efficiency of @xmath110 . for the time - scales of these events we used the microlensing fit where blending was a free parameter ( 5-parameters model ) .
the error in @xmath111 was calculated with the formula given by @xcite . for the original detection efficiency ( i.e.,not corrected for binary events ) , the optical depth was derived to be @xmath112 . if the efficiency is corrected for non detectability of binary lenses , the optical depth becomes @xmath113 .
table [ tab : tau ] presents all calculations of the optical depth .
cccc event & @xmath84 & @xmath114 & @xmath115 + + & & & + ogle - lmc-01 & @xmath116 & 0.212174 & 0.25 + & & & + ogle - lmc-02 & @xmath117 & 0.165536 & 0.13 + total @xmath118 & & & @xmath119 + + & & & + ogle - lmc-01 & @xmath116 & 0.190957 & 0.28 + & & & + ogle - lmc-02 & @xmath117 & 0.148982 & 0.15 + total @xmath118 & & & @xmath120 +
the ogle ii lmc data analysed in this study covered about 4.5 sq . deg . of the sky over 4 years , which is clearly smaller than the data used by macho or eros groups .
however , it provides a completely independent data set to test the hypotheses concerning the presence of machos in the galactic halo .
based on two the events we found the optical depth towards the lmc to be @xmath121 . on one hand , this value is in 2-@xmath69 agreement with the @xmath122 obtained by macho @xcite .
if both ogle ii events were caused by machos , then they would contribute @xmath1 per cent to the total mass of the halo , according to model * _ s _ * from @xcite .
on the other hand , our value of the optical depth is more consistent with a scenario of lensing due to the visible stellar lmc component alone ( self lensing ) , for which various estimates of the optical depth are present in the literature : @xmath123 @xcite , @xmath124 @xcite , @xmath125 @xcite .
@xcite presented a detailed theoretical study of the halo and self - lensing optical depth and compared it with the results of macho .
they simulated the disk and bar of the lmc and quantitatively derived a contribution from the lmc s own stars to the optical depth as @xmath126 . in the region of the lmc bar , where most ogle - ii fields are concentrated , they estimated the self - lensing optical depth to vary between @xmath127 and @xmath88 .
the optical depth derived for two events we found in the ogle - ii data and averaged over all fields is in a perfect agreement with these estimates .
also the time - scales of our two events seem to belong to the regime of self lensing , especially the 57.2-days event , which lies exactly on the line for the self lensing events in fig .
10 of @xcite . from the cmd of the lmc
, it is apparent that both microlensed sources are coincident with an lmc population , indicating that they probably belong to the lmc .
the locations of both events in the bar of the lmc also supports the self lensing scenario .
if we assume that our two events are indeed due to self lensing ( i.e.,we have zero events caused by the halo objects ) , we can derive an estimate for the upper limit of the macho abundance in the halo of the milky way .
we performed this analysis following the procedure similar to the one used by eros in @xcite , first estimating the number of expected events in the ogle ii data set if the halo was entirely made of machos , using model * _ s _ * @xcite . in this model
the optical depth towards the lmc equals @xmath128 , which also includes a contribution from self lensing and galaxy disk lensing .
dark matter lensing solely had @xmath129 .
for a so - called ` typical ' mass of @xmath130 , in ogle ii we should expect 16 events . at the 95 per cent poisson confidence level
there are up to 3 events consistent with a zero detection .
this divided by a number of expected events gives the upper limit on the mass fraction of about 19 per cent .
this translates to an upper limit on the optical depth due to machos of @xmath131 .
if blending is neglected in the all stars sample the numbers are somewhat larger : about 9 expected events would put a limit of @xmath132 .
[ fig : nexp ] shows the dependence of the number of expected events on the mass of the lensing objects for the blended and non - blended approach .
[ fig : upperlimit ] shows the 95 and 90 per cent confidence limits on the halo macho mass fraction for a wide range of masses .
it also compares our results with the results from eros and macho .
this figure clearly shows that the ogle ii project is the most sensitive to events caused by machos with masses between 0.01 and 0.2 @xmath3 . for this range of masses
, we can place an even tighter upper - limit of about 10 per cent ( at 95 per cent confidence ) on the abundance of compact dark matter objects in the galactic halo .
a more detailed theoretical analysis will be presented in a forthcoming paper ( calchi novati , mancini , scarpetta et al .
, in preparation ) . for the sake of comparison with the results of the eros group , we also performed the above analysis narrowing the stars sample only to the bright stars . in each field , our bright sample consisted of about 20 30 per cent of stars present in all stars sample , with a total over all fields of about 1.9 million .
after correction for blending , the number of monitored stars was about 3.6 million .
in the bright sample of ogle ii , we have not found any events . according to the full @xmath133 model of @xcite
, we should expect about 9 events to be caused by lenses of mass @xmath134 , including events due to self lensing .
if we neglect blending , then we expect 6 events . the total optical depth towards the lmc is therefore estimated to be @xmath132 ( @xmath135 ) with ( without ) the full blending correction .
the calculated upper - limit is very high due to the fact that ogle ii has much smaller sky coverage and shorter time - baseline compared to the combined eros i and eros ii projects .
however , a combined study of ogle ii and ogle iii data will be conducted in the near future and should be able to yield constraints of even stronger magnitude than those given by the eros collaboration ( @xcite , wyrzykowski et al .
, in preparation ) .
as has been discussed above , we did not have hst data for all ogle ii fields .
therefore we chose to use the two selected hst fields and consider these as being representative of two different density levels . in this way
we recover only an approximation to the underlying real luminosity function for each field . in order to check
how our selection of hst stars affects the result , we repeated the whole analysis for a different cut on the hst stars magnitude , namely 2.5 mag ( instead of 3.5 mag ) below the brightness of the ogle objects .
this translates to a minimal allowed blending fraction of 10 per cent .
we then rederived the blending distributions and the distributions for the number of hst stars in a single ogle object .
these distributions were obviously different compared to the original 3.5 mag cut ; for both sparse and dense fields , there was nearly no blending in the brightest magnitude bin . for the fainter stars , the distributions indicated a level of blending similar to that which was found previously , but with slightly less pronounced features for small @xmath62 . in a similar manner to that described in section [ sec : blending ] , we estimated the total number of monitored stars and found about @xmath136 stars .
then we also repeated the efficiency calculation including these new blending distributions .
it was similar to the original efficiency to within 2 per cent .
the detection efficiencies ( corrected for binary events ) for the two events were found to be @xmath137 and @xmath138 . since these efficiencies are similar to previous values while the number of monitored stars is reduced , we find that this yields a somewhat larger optical depth of @xmath139 .
this is , however , still consistent with our original value , indicating little sensitivity of the optical depth on the depth of blending stars magnitude limit . as an extreme case we also considered neglecting blending completely in our all stars sample . in this case
, we find that there are only about 5.5 million sources with @xmath103 .
the detection efficiency was obtained as described above and is shown on fig .
[ fig : eff ] .
it was larger than in the case when blending is taken into account by about 20 per cent .
however , this does not cancel out entirely with the reduction in the number of monitored sources . using these numbers with non - blended model time - scales ( table [ tab : models ] )
leads to a larger value for the optical depth of @xmath140 , where we have used the detection efficiencies corrected for lack of binary events .
this indicates that neglecting blending favours larger values of the optical depth , emphasising the importance of carrying out a careful analysis of blending when dealing with all stars samples in order to obtain reliable optical depth results towards the lmc .
in the search for microlensing events in ogle ii data towards the large magellanic cloud , we found two events .
the total optical depth for all 21 ogle ii lmc fields , covering about 4.5 square degrees , was derived for the all stars sample ( with a limit of @xmath141 mag ) as @xmath142 .
if both events were caused by machos , this would imply that their comprise @xmath1 per cent of the halo by mass .
however , the value of the optical depth and the characteristics of the events are consistent with the self lensing scenario , in which lensing towards the lmc occurs only between sources and lenses both located in the cloud . from this
, we can constrain the presence of compact dark matter objects in the milky way s halo and we find that our results are compatible with no detection .
the upper limit is found to be 19 per cent for @xmath143 and 10 per cent for masses between 0.01 and 0.2 @xmath3 .
it is worth emphasising that in our study we probed mainly the bar of the lmc , similarly to the macho project .
however , our results and conclusions are closer to the ones of the eros collaboration , which probed different regions of the lmc .
the sensitivity of the ogle ii experiment is limited to a relatively short time span and sky coverage .
however , the results presented in this paper will be soon strengthened and supplemented with the analysis of the 8-year ogle
iii data .
future photometric and astrometric surveys may confirm the lensing origin of most of events detected so far and the new ones by high - resolution imaging or precise astrometry , allowing for measuring the distances to the sources and lenses or their velocities .
the authors would like pay tribute to the late bohdan paczyski , as the subject of this work was always a subject of interest for him .
we are grateful to the referee , andy gould for his comments and assistance in combining macho and ogle data sets for measure the colour of ogle - lmc-01 .
we also would like to thank for their help at various stages of this work to drs wyn evans , subo dong , luigi mancini , sebastiano calchi - novati , gaetano scarpetta , hongsheng zhao and daniel faria .
w , sk , mcs and js acknowledge generous support from the european community s fr6 marie curie programme , contract no .
mrtn - ct-2004 - 505183 `` angles '' .
mcs also acknowledges support from the stfc - funded `` galaxy formation and evolution '' program at the institute of astronomy , university of cambridge .
the ogle project is partially supported by the polish mnisw grant n20303032/4275 .
js also acknowledges support through the polish mnisw grant no .
n20300832/070 .
n. , nguyen h. t. , werner m. w. , alcock c. , patten b. m. , stern d. , 2004 , in bulletin of the american astronomical society vol .
36 of bulletin of the american astronomical society , spitzer space telescope observations of the aftermath of microlensing event macho - lmc-5 .
pp 725 . , kozowski s. , belokurov v. , smith m. c. , skowron j. , udalski a. , 2008 , in proceedings of the manchester microlensing conference : the 12th international conference and angles microlensing workshop , eds
e. kerins , s. mao , n. rattenbury and l. wyrzykowski pos(gmc8)011 , the ogle search for microlensing events towards the lmc | we present the results from the ogle ii survey ( 1996 - 2000 ) towards the large magellanic cloud ( lmc ) , which has the aim of detecting the microlensing phenomena caused by dark matter compact objects in the galactic halo ( machos ) .
we use high resolution hst images of the ogle fields and derive the correction for the number of monitored stars in each field .
this also yield blending distributions which we use in catalogue level monte carlo simulations of the microlensing events in order to calculate the detection efficiency of the events .
we detect two candidates for microlensing events in the all stars sample , which translates into an optical depth of @xmath0 .
if both events were due to macho the fraction of mass of compact dark matter objects in the galactic halo would be @xmath1 per cent .
this optical depth , however , along with the characteristics of the events , seems to be consistent with the self lensing scenario , i.e. , self lensing alone is sufficient to explain the observed microlensing signal .
our results indicate a non - detection of machos lensing towards the lmc with an upper limit on their abundance in the galactic halo of 19 per cent for @xmath2 and 10 per cent for masses between 0.01 and 0.2 @xmath3 .
[ firstpage ] cosmology : dark matter , gravitational lensing , galaxy : structure , halo , galaxies : magellanic clouds |
the braneworld scenario provides an interesting alternative to the standard kaluza - klein compactification of the extra dimensions .
the simplest phenomenological models describing such a scenario are the five - dimensional randall - sundrum type braneworld models ( for a review see @xcite ) . from the point of view of embedding these models into a more fundamental theory , such as string / m - theory
, one may expect that a more complete version of the scenario must admit the presence of additional extra dimensions compactified on an internal manifold . from a phenomenological point of view , the consideration of more general spacetimes offer a richer geometrical structure and may provide interesting extensions of the randall - sundrum mechanism for the geometric origin of the hierarchy .
more extra dimensions also relax the fine - tunings of the fundamental parameters .
these models can provide a framework in the context of which the stabilization of the radion field naturally takes place .
in addition , a richer topological structure of the field configuration in transverse space provides the possibility of more realistic spectrum of chiral fermions localized on the brane .
several variants of the randall - sundrum scenario involving cosmic strings and other global defects of various codimensions have been investigated in higher dimensions ( see , for instance , @xcite and references therein ) . motivated by the problems of the radion stabilization and the generation of cosmological constant ,
the role of quantum effects in braneworlds has attracted great deal of attention @xcite-@xcite . a class of higher dimensional models with the topology @xmath1 , where @xmath2 is a one - parameter compact manifold , and with two branes of codimension one located at the orbifold fixed points , is considered in refs .
@xcite . in both cases of the warped and unwarped internal manifold ,
the quantum effective potential induced by bulk scalar fields is evaluated and it has been shown that this potential can stabilize the hierarchy between the planck and electroweak scales without fine tuning .
in addition to the effective potential , the investigation of local physical characteristics in these models is of considerable interest .
local quantities contain more information on the vacuum fluctuations than the global ones and play an important role in modelling a self - consistent dynamics involving the gravitational field . in the previous papers
saha06a , saha06b we have studied the wightman function , the vacuum expectation values of the field square and the energy - momentum tensor for a scalar field with an arbitrary curvature coupling parameter obeying robin boundary conditions on two codimension one parallel branes embedded in the background spacetime @xmath3 with a warped internal space @xmath2 . for an arbitrary internal space @xmath4 , the application of the generalized abel - plana formula @xcite allowed us to extract form the vacuum expectation values the part due to the bulk without branes and to present the brane induced parts in terms of exponentially convergent integrals for the points away from the branes .
the braneworld corresponds to a manifold with boundaries and the physical quantities , in general , receive both volume and surface contributions . in particular , the contributions located on the visible brane are of special interest as they are direct observables in the theory . in ref .
saha04surf the vacuum expectation value of the surface energy - momentum tensor is evaluated for a massive scalar field subject to robin boundary conditions on two parallel branes in @xmath0-dimensional ads bulk ( for surface densities induced by quantum fluctuations of a bulk scalar field on a single brane in rindler - like spacetimes see @xcite ) .
it has been shown that for large distances between the branes the induced surface densities give rise to an exponentially suppressed cosmological constant on the brane . in the randall - sundrum braneworld model , for the interbrane distances solving the hierarchy problem between the gravitational and electroweak mass scales ,
the cosmological constant generated on the visible brane is of the right order of magnitude with the value suggested by the cosmological observations . in the present paper
we investigate similar issues within the framework of higher dimensional braneworld models with warped internal spaces .
the paper is organized as follows . in the next section
we describe the corresponding bulk geometry and outline the structure of the eigenmodes and the kaluza - klein spectrum .
an integral representation is constructed for the zeta function related to the vacuum expectation value of the surface energy - momentum tensor .
this function is presented in the form of the sum of single brane and second brane induced parts .
the analytic continuation of the zeta function for the model with a single brane is done in section sec:1brane .
the surface energy - momentum is investigated on both sides of the branes and the renormalization procedure is discussed . in section sec:2brane
we study the surface energy density induced by the presence of the second brane .
various limiting cases are considered and the cosmological constant induced on the brane is estimated .
the section [ sec : enbal ] is devoted to the evaluation of the total vacuum energy in the two - brane setup on the base of the zeta function technique .
the relation between the bulk and surface energies is discussed .
the last section contains a summary of the work .
consider a massive scalar field @xmath5 with curvature coupling parameter @xmath6 .
the corresponding field equation has the form ( we adopt the conventions of ref .
@xcite for the metric signature and the curvature tensor ) @xmath7where @xmath8 is the covariant derivative operator and @xmath9 is the scalar curvature for a @xmath0-dimensional background spacetime . for the most important special cases of minimally and conformally coupled scalars one has @xmath10 and @xmath11 , respectively .
we will assume that the background spacetime has the topology @xmath12 and is described by the line element @xmath13where @xmath14 is the metric tensor for @xmath15-dimensional minkowski spacetime , @xmath16 is the inverse ads radius , and the coordinates @xmath17 , @xmath18 , cover the internal manifold @xmath2 , @xmath19 .
the ricci scalar corresponding to line element ( metric ) is given by formula @xmath20 , with @xmath21 being the scalar curvature for the metric tensor @xmath22 . in the discussion below , in addition to the radial coordinate @xmath23 we will also use the coordinate @xmath24 , in terms of which the line element is written in the form conformally related to the metric in the direct product spacetime @xmath25 by the conformal factor @xmath26 .
we are interested in one - loop vacuum effects induced by quantum fluctuations of the bulk field @xmath5 on two parallel branes of codimension one , located at @xmath27 and @xmath28 , @xmath29 .
we assume that on the branes the field obeys robin boundary conditions @xmath30with constant coefficients @xmath31 , @xmath32 .
this type of conditions is an extension of dirichlet and neumann boundary conditions and appears in a variety of situations , including the considerations of vacuum effects for a confined charged scalar field in external fields , spinor and gauge field theories , quantum gravity and supergravity .
robin boundary conditions naturally arise bulk fields in braneworld models .
the branes divide the space into three regions corresponding to @xmath33 , @xmath34 , and @xmath35 . in general , the coefficients in the boundary conditions ( [ boundcond ] ) can be different for separate regions . in a higher dimensional generalization of the randall - sundrum braneworld model
the coordinate @xmath23 is compactified on an orbifold @xmath36 , of ltngth @xmath37 , @xmath38 , and the orbifold fixed points @xmath39 and @xmath40 are the locations of two branes .
the corresponding line - element has the form ( [ metric ] ) with the warp factor @xmath41 . in these models the region between the branes
is employed only . for an untwisted bulk scalar with brane mass terms @xmath42 and @xmath43 , the corresponding ratio of the coefficients in the boundary condition ( [ boundcond ] )
is determined by the expression ( see , e.g. , refs . @xcite for the case of the bulk @xmath44 and refs .
@xcite for the geometry under consideration ) @xmath45 in the supersymmetric version of the model @xcite one has @xmath46 and the boundary conditions are the same for both branes . for a twisted scalar ,
dirichlet boundary conditions are obtained on both branes . for the geometry of two parallel branes in @xmath1 with boundary conditions
( [ boundcond ] ) , the wightman function and the vacuum expectation values ( vevs ) of the field square and the bulk energy - momentum tensor are investigated in refs .
@xcite . on manifolds with boundaries
the energy - momentum tensor in addition to the bulk part contains a contribution located on the boundary . for an arbitrary smooth boundary @xmath47 with the inward - pointing unit normal vector @xmath48 ,
the surface part of the energy - momentum tensor for a scalar field is given by the formula @xcite @xmath49where the `` one - sided '' delta - function @xmath50 locates this tensor on @xmath47 and @xmath51 in eq .
( [ tausurf ] ) , @xmath52 is the induced metric on the boundary and @xmath53 is the corresponding extrinsic curvature tensor . by expanding the field operator over a complete set of eigenfunctions @xmath54 obeying the boundary conditions and using the standard commutation rules , for the vev of the operator @xmath55 one finds @xmath56where
the bilinear form @xmath57 on the right is determined by the classical energy - momentum tensor ( [ tausurf ] ) and the collective index @xmath58 can contain both discrete and continuous parts . below in this section , we will consider the region between the branes .
the corresponding quantities for the regions @xmath59 and @xmath60 are obtained as limiting cases and are investigated in the next section . as we have mentioned before , in the orbifolded version of the model ,
the only bulk is the one between the branes . in this region
the inward - pointing normal to the brane at @xmath61 , @xmath62 , and the corresponding extrinsic curvature tensor are given by the relations@xmath63where @xmath64 is defined in formula ( [ ajbjrs ] ) . by using relations ( [ njmkjmn ] ) and the boundary conditions , the vev of the surface energy - momentum tensor on the brane at @xmath61 is presented in the form @xmath65with the notation@xmath66from the point of view of physics on the brane , eq .
( [ tauj ] ) corresponds to the gravitational source of the cosmological constant type with the surface energy density @xmath67 ( surface energy per unit physical volume on the brane at @xmath61 or brane tension ) , stress @xmath68 , and the equation of state @xmath69it is remarkable that this relation takes place for both subspaces on the brane .
it can be seen that this result is valid also for the general metric @xmath70 instead of @xmath14 in line element ( metric ) . for an untwisted bulk scalar in the higher dimensional generalization of the randall - sundrum braneworld based on the bulk @xmath71 the coefficient in eq .
( [ tauj ] ) is given by the formula @xmath72 in particular , the corresponding surface energy vanishes for minimally and conformally coupled scalar fields with zero brane mass terms . in order to evaluate the expectation value ( [ modesumform ] ) we need the corresponding eigenfunctions .
these functions can be taken in the decomposed form @xmath73where
@xmath74and @xmath75 are separation constants .
the modes @xmath76 are eigenfunctions for the internal subspace:@xmath77 \psi _ { \beta } ( x)=-\lambda _ { \beta } ^{2}\psi _ { \beta } ( x ) , \label{psibeteq}\]]with the eigenvalues @xmath78 and the orthonormalization condition @xmath79 in eq .
( [ psibeteq ] ) , @xmath80 is the laplace - beltrami operator for the metric @xmath81 . in the consideration below
we will assume that @xmath82 .
the dependence on @xmath83 and @xmath84 in the separation constants is factorized:@xmath85where the mass spectrum @xmath86 is determined by the radial equation together with the boundary conditions . from the field equation ( [ fieldeq ] ) , one obtains the equation for the function @xmath87 with the solution @xmath88where the normalization coefficient @xmath89 is determined below , and @xmath90here @xmath91 , @xmath92 are the bessel and neumann functions of the order @xmath93and the @xmath94-coordinates of the branes are denoted by @xmath95 , @xmath62 . in formula ( [ gnuj ] ) and in what follows for a given function @xmath96 we use the notation @xmath97with the coefficients @xmath98 in the discussion below we will assume values of the curvature coupling parameter for which @xmath99 is real . for imaginary @xmath99
the ground state becomes unstable @xcite .
note that for conformally and minimally coupled massless scalar fields one has @xmath100 and @xmath101 , respectively .
the function ( [ fny ] ) satisfies the boundary condition on the brane at @xmath102 . imposing the boundary condition on the brane @xmath28 we find that the eigenvalues @xmath86 are solutions to the equation @xmath103where the function on the left
is defined by@xmath104equation ( [ cnu ] ) determines the kaluza - klein ( kk ) spectrum along the transverse dimension .
we denote by @xmath105 , @xmath106 , the positive zeros of the function @xmath107 , arranged in the ascending order , @xmath108 .
the eigenvalues for @xmath86 are related to these zeros as @xmath109from the orthonormality condition of the radial functions,@xmath110for the coefficient @xmath89 in eq .
( [ fny ] ) one finds @xmath111note that , as we consider the quantization in the region between the branes , @xmath112 , the modes defined by ( [ fny ] ) are normalizable for all real values of @xmath99 from eq .
( [ nu ] ) .
as it follows from formula ( [ tauj ] ) , the vev of the surface energy - momentum tensor can be obtained from the vev of the field square evaluated on the branes .
the vev of the field square on the bulk is derived from the corresponding wightman function in the coincidence limit . by using the generalized abel - plana formula @xcite for the summation over the zeros @xmath113 , in ref .
@xcite this vev is presented in the form@xmath114where @xmath115 and @xmath116 provide two equivalent representations and@xmath117here and in what follows we use the notations @xmath118with the modified bessel functions @xmath119 , @xmath120 , and @xmath121the first term on the right of eq .
( [ phi2bulk ] ) , @xmath122 , corresponds to the vev in @xmath1 spacetime without branes , @xmath123 is induced by a single brane located at @xmath61 when the second brane is absent , and the last term is induced by the presence of the second brane .
after the standard renormalization of the boundary free part , the vev given by eq .
( [ phi2bulk ] ) is finite in the bulk and diverges on the branes . as a result , the vevs of the field square on the branes can not be obtained directly from this formula and an additional renormalization procedure is necessary .
one possibility is to remove from ( [ phi2bulk ] ) the terms which diverge for the points on the branes .
an alternative way is to restrict from the begining to the fluctuations on the brane and to use the dimensional regularization or the generalized zeta function technique ( for a discussion of the relation between different methods see , for instance , ref .
@xcite ) . here
we will follow the last method . substituting the eigenfunctions ( [ eigfunc1 ] ) into the corresponding mode sum and integrating over the angular part of the vector @xmath124 , for the expectation value of the energy density on the brane at @xmath61 we obtain @xmath125where @xmath126 , and @xmath127 ,
@xmath128 is defined by formulas ( [ omega ] ) , ( [ mbeta2 ] ) , and we have used the relation @xmath129 . to regularize the divergent expression on the right of this formula we define the function @xmath130where an arbitrary mass scale @xmath131
is introduced to keep the dimension of the expression .
after the evaluation of the integral over @xmath132 , this expression can be presented in the form@xmath133where the generalized partial zeta function @xmath134is introduced with the notation@xmath135the computation of the vev of the surface energy - momentum tensor requires the analytic continuation of the function @xmath136 to the value @xmath137 ( here and below @xmath138 is understood in the sense of the analytic continuation ) , @xmath139 in order to obtain this analytic continuation we will follow the procedure multiply used for the evaluation of the casimir energy ( see , for instance , @xcite ) . the starting point is the representation of the function ( [ zetsx ] ) in terms of the contour integral @xmath140where @xmath141 is a closed counterclockwise contour in the complex @xmath142 plane enclosing all zeros @xmath86 .
the location of these zeros enables one to deform the contour @xmath141 into a segment of the imaginary axis @xmath143 and a semicircle of radius @xmath9 , @xmath144 , in the right half - plane .
we will also assume that the branch points @xmath145 are avoided by small semicircles @xmath146 in the right half - plane with small radius @xmath147 .
here we assume that there is no zero mode corresponding to @xmath148 .
the changes introduced by the presence of the zero mode will be discussed below in this section . for negative @xmath149 with sufficiently large absolute value the integral over the large semicircle in eq .
( [ intzetsx1 ] ) tends to zero in the limit @xmath150 , and the expression on the right can be transformed to@xmath151 .
\label{zetajbets4}\end{aligned}\]]for @xmath152 the second integral on the right is finite in the limit @xmath153 , whereas the first integral vanishes as @xmath154 . as a result , by using the formula @xmath155 , in this strip of complex plane @xmath156 we have the following integral representation @xmath157where @xmath158 in the last formula we use the notation @xmath159 for @xmath115 and @xmath160 for @xmath116 .
the contribution of the second term on the right of eq .
( [ zet12 ] ) is finite at @xmath137 and vanishes in the limits @xmath161 or @xmath162 .
the first term corresponds to the contribution of a single brane at @xmath163 when the second brane is absent .
the surface energy density corresponding to this term is located on the surface @xmath164 for the brane at @xmath27 and on the surface @xmath165 for the brane at @xmath28 . to distinguish
on which side of the brane is located the corresponding term we use the superscript j = l for the left side and j = r for the right side .
as we consider the region between the branes , in formula ( [ fls ] ) j = r for @xmath115 and j = l for @xmath116 .
the further analytic continuation is needed for the function @xmath166 only and this is done in the next section . in the discussion
above we have assumed that there is no zero mode corresponding to @xmath148 .
for special values of the parameters of the model , satisfying the condition@xmath167the value @xmath148 is a solution to the eigenvalue equation ( [ cnu ] ) and the zero mode is present
. we will denote this mode by @xmath168 .
the corresponding contribution to the vev of the surface energy - momentum tensor and to the zeta function defined above is taken into account if we assume that the summation over @xmath169 in the corresponding formulas includes the summand with @xmath170 as well .
the expression for this term is obtained from the formula for general @xmath169 taking the limit @xmath171 .
in particular , the contribution of the zero mode to the partial zeta function ( [ zetsx ] ) is given by@xmath172as before , the contribution of the nonzero modes , can be presented as the contour integral given by eq .
( [ intzetsx1 ] ) , but now the contour @xmath141 contains an additional semicircle @xmath173 of small radius which avoids the origin in the right half - plane .
consequently , this contribution is given by the right hand side of eq .
( [ zetajbets4 ] ) plus the integral over the contour @xmath173 . evaluating the latter integral
we can see that it cancels the contribution coming from the zero mode presented by eq .
( [ zeromodezet ] ) .
hence , we conclude that the integral representation for the partial zeta function given by eqs .
( [ zet12 ] ) , ( [ fls ] ) is valid for the case of the presence of the zero mode as well .
in this section we will consider the geometry of a single brane placed at @xmath174 . the orbifolded version of this model corresponds to the higher dimensional generalization of the randall - sundrum 1-brane model with the brane location at the orbifold fixed point @xmath39 .
the corresponding partial zeta function is given by eq .
( [ fls ] ) . now in this formula @xmath175 for the left and right sides of the brane , respectively , and @xmath159 for @xmath176 and @xmath160 for @xmath177 .
in addition , the replacement @xmath178 should be done with @xmath179 and @xmath180 .
the integral representation ( [ fls ] ) for a single brane partial zeta function is valid in the strip @xmath181 and under the assumption that the function @xmath182 has no real zeros . for the analytic continuation to @xmath137
we employ the asymptotic expansions of the modified bessel functions for large values of the argument @xcite . for @xmath183 from these expansions one has @xmath184where the coefficients @xmath185 are combinations of the corresponding coefficients in the expansions for the functions @xmath186 and @xmath187 .
note that one has the relation@xmath188assuming that the coefficients in the boundary conditions are the same for both sides of the brane .
for the nonzero modes along the internal space @xmath189 we subtract and add to the integrand in ( [ fls ] ) the @xmath190 leading terms of the corresponding asymptotic expansion and exactly integrate the asymptotic part : @xmath191 .
\label{zetalabet}\end{aligned}\]]for the zero mode we first separate the integral over the interval @xmath192 and apply the described procedure to the integral over @xmath193 . by using these formulas for @xmath166 , the corresponding function @xmath194is written in the form @xmath195 - \sum_{l=1}^{n}\frac{|\psi _ { 0}(x)|^{2}v^{(f , j)}_{l}}{d_{1}+s - l+1 } \nonumber \\ & & + \frac{1}{2}\gamma ( \alpha _ { s}+1)\sum_{l=1}^{n}\frac{v^{(f , j)}_{l}z_{j}{}^{d_{1}+s - l+1}}{\gamma ( l/2)}\gamma \left ( \frac{l}{2}-\alpha _ { s}-1\right)\zeta _ { \sigma } \left ( \frac{l}{2}-\alpha _ { s}-1,x\right ) \bigg\ } , \label{fls1}\end{aligned}\ ] ] where @xmath196 and @xmath197 is defined by formula ( [ alfas ] ) . in eq . ( [ fls1 ] )
we have introduced the local spectral zeta function associated with the massless laplacian defined on the internal subspace @xmath2 : @xmath198where the prime on the summation sign means that the zero mode should be omitted . both integrals in eq .
( [ fls1 ] ) are finite at @xmath137 for @xmath199 . for large values @xmath200
the second integral behaves as @xmath201 and the series over @xmath202 in eq .
( [ fls1 ] ) is convergent at @xmath137 for @xmath203 . for these values
@xmath190 the poles at @xmath137 are contained only in the last two terms on the right .
the zero mode part has a simple pole at @xmath137 presented by the summand @xmath204 of the second sum in figure braces .
the pole part corresponding to the nonzero modes is extracted from the pole structure of the local zeta function ( [ localzeta ] ) .
the latter is given by the formula@xmath205where @xmath206 is a half integer , the coefficients @xmath207 are related to the seeley - dewitt or heat kernel coefficients for the corresponding non - minimal laplacian , and the dots denote the terms vanishing at @xmath208 . in the way similar to that used in ref .
@xcite , it can be seen that the coefficients @xmath209 are related to the corresponding coefficients @xmath210 for the massive zeta function@xmath211by the formula@xmath212the vev of the energy density on a single brane is derived from@xmath213by using relation ( [ zetapoles ] ) , the energy density is written as a sum of pole and finite parts : @xmath214laurent - expanding the expression on the right of eq .
( [ fls1 ] ) near @xmath137 , one finds@xmath215 \label{phi2lpf1}\end{aligned}\]]for the pole part , and @xmath216 \nonumber \\ & & -k_{d}^{d}z_{j}^{d_{2}}c_{j}\beta _ { d_{1}}|\psi _ { 0}(x)|^{2}\bigg\ { \sideset{}{'}{\sum } _ { l=1}^{n}\frac{2v^{(f , j)}_{l}}{d_{1}-l}% -v^{(f , j)}_{d_{1}}\left [ 2\ln ( \mu z_{j})+\psi \left ( \frac{d_{1}}{2}\right ) -\psi \left ( \frac{1}{2}\right ) \right ] \bigg\ } \nonumber \\ & & + \frac{k_{d}^{d}c_{j}}{(4\pi ) ^{d_{1}/2}}\sum_{l=1}^{n}\frac{% v^{(f , j)}_{l}}{\gamma ( l/2)}z_{j}^{d - l}\left\ { c_{(d - l)/2}(x)\left [ 2\ln \mu -\psi \left ( \frac{1}{2}\right ) \right ] + \omega _ { ( l - d_{1})/2}(x)\right\ } \label{phi2lpf2}\end{aligned}\ ] ] for the finite part , with @xmath217 being the diagamma function . in this formula
the prime on the summation sign means that the term with @xmath218 should be omitted and it is understood that @xmath219 for @xmath220 . in the pole part the second term in the square brackets comes from the zero mode along @xmath2 and this term is cancelled by the delta term on the right of eq .
( [ cpx ] ) .
for a one parameter internal manifold @xmath189 with the length scale @xmath221 one has @xmath222 and @xmath223 . in this case instead of the zeta function ( [ localzeta ] ) we could introduce the dimensionless function @xmath224 .
the corresponding dimensionless coefficients @xmath225 and @xmath226 in the formula analog to eq .
( [ zetapoles ] ) , are related to the coefficients in eq .
( [ zetapoles ] ) by the formulas@xmath227 , \end{split}%\]]and do not depend on @xmath221 .
now we see that the term with @xmath228 is combined with the term @xmath229 of eq .
( [ phi2lpf2 ] ) in the form @xmath230 .
the renormalization of the surface energy density can be done modifying the procedure used previously for the renormalization of the casimir energy in the randall - sundrum model @xcite and in its higher - dimensional generalizations with compact internal spaces flac03,flac03b .
the form of the counterterms needed for the renormalization is determined by the pole part of the surface energy density given by eq .
( [ phi2lpf1 ] ) .
for an internal manifold with no boundaries , this part has the structure @xmath231}a_{l}^{\mathrm{(s)}% } ( z_{j}/l)^{2l}$ ] . by taking into account that the intrinsic scalar curvature @xmath232 for the brane at @xmath61 contains the factor @xmath233
, we see that the pole part can be absorbed by adding to the brane action counterterms of the form@xmath234}b_{l}^{\mathrm{(s)}}r_{j}^{l } , \label{counterterms}\]]where the square brackets in the upper limit of summation mean the integer part of the enclosed expression . by taking into account that there is the freedom to perform finite renormalizations
, we see that the renormalized surface energy density on a single brane is given by the formula@xmath235}c_{l}^{\mathrm{(s)}}(z_{j}/l)^{2l}. \label{epsjren}\]]the coefficients @xmath236 in the finite renormalization terms are not computable within the framework of the model under consideration and their values should be fixed by additional renormalization conditions .
the total surface energy density for a single brane at @xmath61 is obtained by summing the contributions from the left and right sides:@xmath237 in formulas ( [ phi2lpf1 ] ) , ( [ phi2lpf2 ] ) we should take @xmath160 for @xmath238 and @xmath159 for @xmath239 .
now we see that , assuming the same boundary conditions on both sides of the brane , the coefficients @xmath240 enter into the sum of pole terms in the form@xmath241}\frac{v_{2l}^{(i , j)}}{\gamma ( l)}% z_{j}{}^{d-2l}c_{d/2-l}(x,0 ) .
\label{cinsumpole}\]]if the internal manifold contains no boundaries and @xmath242 is an odd number , one has @xmath243 and , hence , the pole parts coming from the left and right sides cancel out . in this case for the total surface energy density one obtains the formula @xmath244}\frac{v^{(i , j)}_{2l}}{u^{2l}}% \bigg ) \bigg ] \nonumber \\ & & -2|\psi _ { 0}(x)|^{2}\sum_{l=1}^{[n/2]}\frac{v^{(i , j)}_{2l}}{%
d_{1}-2l}+\gamma \left ( \frac{d_{1}}{2}\right ) \sum_{l=1}^{[n/2]}\frac{% v^{(i , j)}_{2l}}{\gamma ( l)}\frac{\omega _ { l - d_{1}/2}(x)}{% z_{j}^{2l - d_{1}}}\bigg\ } , \label{epslr1}\end{aligned}\ ] ] where the coefficients @xmath245 are defined by relation ( [ inuratas ] ) . note that this quantity does not depend on the renormalization scale @xmath131 . in the orbifolded version of the model with a single brane at @xmath39 , which corresponds to the higher dimensional generalization of the randall - sundrum 1-brane model ,
the bulk is symmetric under the reflection @xmath246 . in this model
the surface densities on the left and right sides of the brane are the same and coincide with @xmath247 . in particular , here the abovementioned cancellation of the pole parts from
left and right sides does not take place .
for a one parameter internal space of size @xmath221 the surface energy density on the brane at @xmath61 is a function on the ratio @xmath248 only .
note that in the case of the ads bulk the corresponding quantity does not depend on the brane position . to discuss the physics from the point of view of an observer residing on the brane , it is convenient to introduce rescaled coordinates@xmath249with this coordinates the warp factor in the metric is equal to 1 on the brane and they are physical coordinates for an observer on the brane .
for this observer the physical size of the subspace @xmath2 is @xmath250 and the corresponding kk masses are rescaled by the warp factor : @xmath251 .
now we see that the surface energy density is a function on the ratio @xmath252 of the physical size for the internal space ( for an observer residing on the brane ) to the ads curvature radius . as an application of the general results presented above
, we can consider a simple example with @xmath253 . in this case
the bulk corresponds to the @xmath44 spacetime with one compactified dimension @xmath254 .
the corresponding normalized eigenfunctions and eigenvalues are as follows @xmath255where @xmath221 is the length of the compactified dimension . the surface energy density induced on the brane is obtained from general formulas by the replacements @xmath256where the prime means that the summand @xmath257 should be taken with the weight 1/2 .
for the local zeta function from eq .
( [ localzeta ] ) one has@xmath258where @xmath259 is the riemann zeta function . now
the only poles of the function @xmath260 are the points @xmath261 . by using the standard formulas for the gamma function and the riemann zeta function ( see , for instance , @xcite )
, it can be seen that one has@xmath262for the residues appearing in ( [ zetapoles ] ) and @xmath263 for the finite parts , with @xmath264 being the euler constant .
as it has been shown in section [ sec : zetafunc ] , the partial zeta function related to the surface energy density on the brane at @xmath61 is presented in the form ( [ zet12 ] ) , where the second term on the right is finite at the physical point @xmath137 . by taking into account formulas ( fjs2 ) ,
( [ ifs0 ] ) , for two - brane geometry the vev of the surface energy density on the brane at @xmath61 is presented as the sum @xmath265the first term on the right is the energy density induced on a single brane when the second brane is absent .
the second term is induced by the presence of the second brane and is given by the formula @xmath266by taking into account relation ( [ tauj ] ) between the vev of the field square and the surface energy - momentum tensor , this formula can aslo be obtained from the last term of eq .
( [ phi2bulk ] ) evaluated at the brane @xmath267 . as we consider the region @xmath112 , the energy desnity ( [ emt2pl3 ] ) is located on the surface @xmath164 for the left brane and on the surface @xmath165 for the right brane .
consequently , in formula ( [ emt2pl2 ] ) we take @xmath239 for @xmath115 and @xmath177 for @xmath116 .
the energy densities on the surfaces @xmath268 and @xmath269 are the same as for the corresponding single brane geometry .
the expression on the right of eq . ( [ emt2pl3 ] ) is finite for all nonzero distances between the branes and is not touched by the renormalization procedure . for a given value of the ads energy scale @xmath16 and one parameter manifold @xmath2 with the length scale @xmath221 , it is a function on the ratios @xmath270 and @xmath271 .
the first ratio is related to the proper distance between the branes , @xmath272and the second one is the ratio of the size of the internal space , measured by an observer residing on the brane at @xmath27 , to the ads curvature radius @xmath273 .
the expression ( [ emt2pl3 ] ) for the surface energy density can be presented in another equivalent form : @xmath274which will be used below in the discussion of the relations between the bulk and surface energy densities . for the comparison with the case of the bulk spacetime @xmath275
when the internal space is absent , it is useful in addition to the vev ( [ emt2pl3 ] ) to consider the corresponding quantity integrated over the subspace @xmath2 : @xmath276where @xmath277 is defined by the relation @xmath278comparing this integrated vev with the corresponding formula from ref .
saha04surf , we see that the contribution of the zero kk mode ( @xmath279 ) in eq .
( [ phi2integrated ] ) differs from the vev of the energy density in the bulk @xmath280 by the order of the modified bessel functions : for the latter case @xmath281 with @xmath282 defined by eq .
( [ nu ] ) with the replacement @xmath283 .
note that for @xmath284 one has @xmath285 .
in particular , this is the case for minimally and conformally coupled scalar fields .
now we turn to the investigation of the part ( [ emt2pl3 ] ) in the surface energy density in asymptotic regions of the parameters . for large values of ads radius compared with the interbrane distance , @xmath286 , the main contribution to the integral on the right of eq .
( [ emt2pl3 ] ) comes from large values of @xmath287^{-1}$ ] . assuming that @xmath288 and @xmath289 are fixed
, we see that the order of the bessel modified functions is large . replacing these functions by their uniform asymptotic expansions for large values of the order @xcite , one obtains@xmath290 } , \label{epssmallkd}\end{aligned}\]]where @xmath291 and we have introduced the notation @xmath292the expression on the right of eq .
( [ epssmallkd ] ) is the corresponding surface energy on the brane in the bulk geometry @xmath293 . for large kk masses along @xmath2 , @xmath294 , @xmath295 , we can replace the modified bessel functions by the corresponding asymptotic expansions for large values of the argument . for the contribution of a given kk mode to the leading order
this gives@xmath296where @xmath297if in addition one has the condition @xmath298 , the dominant contribution into the @xmath142-integral comes from the lower limit and we have the formula@xmath299 in particular , for sufficiently small length scale of the internal space this formula is valid for all nonzero kk masses and the main contribution to the surface densities comes from the zero kk mode . in the opposite limit of large internal space , to the leading order we obtain the corresponding result for parallel branes in @xmath300 bulk @xcite . for large values of the mass with @xmath301 , @xmath302 , and @xmath303 ,
one has @xmath304 . introducing a new integration variable @xmath305 and using the uniform asymptotic expansions for the modified bessel functions
, it can be seen that the corresponding asymptotic formula for @xmath277 is obtained from ( [ epsjlargekk1 ] ) by the replacement @xmath306 .
if both masses @xmath307 and @xmath200 are of the same order , the asymptotic behavior for @xmath308 is described by eq .
( [ epsjlargekk1 ] ) replacing @xmath309 . as we could expect in this case the induced surface densities are exponentially suppressed . for small interbrane distances , @xmath286 , which is equivalent to @xmath310 , the main contribution into the integral in eq .
( emt2pl3 ) comes from large values @xmath142 and to the leading order we obtain formula ( [ phi22pllargelamb ] ) .
if in addition one has @xmath311 or equivalently @xmath312 , we can put in this formula @xmath313 .
assuming @xmath314 , for @xmath315 to the leading order one finds@xmath316where @xmath317 for @xmath318 , and @xmath319 for @xmath320 and @xmath321 , with @xmath322 for @xmath115 and @xmath323 for @xmath116 .
we see that for small interbrane distances the sign of the induced surface energy density is determined by the coefficient @xmath324 and this sign is different for two cases of @xmath325 .
now we consider the limit @xmath326 assuming that @xmath327 . using the asymptotic formulas for the bessel modified functions containing in the argument @xmath328 , for the contribution of a given nonzero kk mode we find the following results@xmath329this limit corresponds to the interbrane distances much larger compared with the ads curvature radius and with the inverse kk masses measured by an observer on the left brane : @xmath330 . for a single parameter manifold @xmath2 with length scale @xmath221 and
@xmath331 these conditions are satisfied for all nonzero kk modes . in the limit @xmath332 for fixed @xmath333 , by using the asymptotic formulas for the modified bessel functions for small values of the argument and assuming @xmath334 , one finds@xmath335where we have introduced the notation@xmath336for @xmath337 we should take into account the next terms in the corresponding expansions of the modified bessel functions .
the integral in eq .
( [ epssmallza1 ] ) is negative for small values of the ratio @xmath338 and is positive for large values of this ratio . as it follows from eq .
( [ epssmallza2 ] ) , for large interbtane separations the sign of the quantity @xmath339 is determined by the combination @xmath340 of the coefficients in the boundary conditions . in the limit under consideration
the kk masses measured by an observer on the brane at @xmath27 are much less than the ads energy scale , @xmath341 , and the interbrane distance is much larger than the ads curvature radius .
in particular ,
substituting @xmath342 , from these formulas we obtain the asymptotic behavior for the contribution of the zero mode to the surface energy density induced by the second brane in the limit @xmath343 . now combining the corresponding result with formulas ( [ epslargezb1 ] ) , ( [ epslargezb2 ] )
, we see that under the conditions @xmath344 , and @xmath345 , the contribution of the nonzero kk modes along @xmath4 is suppressed with respect to the contribution of the zero mode by the factor @xmath346 . from the analysis given above it follows that in the limit when the right brane tends to the ads horizon , @xmath347 , the energy density @xmath348 vanishes as @xmath349 for the nonzero kk mode along @xmath189 and as @xmath350 for the zero mode .
the energy density on the right brane , @xmath351 , vanishes as @xmath352 for the nonzero kk mode and behaves like @xmath353 for the zero mode . in the limit when the left brane tends to the ads boundary , @xmath161
, the contribution of a given kk mode vanishes as @xmath354 for @xmath355 and as @xmath356 for @xmath357 . for small values of the ads
curvature radius corresponding to strong gravitational fields , assuming @xmath358 and @xmath298 , we can estimate the contribution of the nonzero kk modes to the induced energy densities by formula ( [ epsjlargekk1 ] ) .
in particular , for the case of a single parameter internal space with the length scale @xmath221 , under the assumed conditions the length scale of the internal space measured by an observer on the brane at @xmath27 is much smaller compared to the ads curvature radius , @xmath359 . if @xmath360 one has @xmath361 and to estimate the contribution of the induced surface densities we can use formulas ( [ epslargezb1 ] ) and ( [ epslargezb2 ] ) , and the suppression is stronger compared with the previous case . for the zero kk mode , under the condition @xmath362 we have @xmath363 and to the leading order the corresponding energy densities are described by relations ( [ epssmallza1 ] ) and ( [ epssmallza2 ] ) . from these formulae
it follows that the induced energy densities integrated over the internal space behave as @xmath364 $ ] for the brane at @xmath61 and are exponentially suppressed .
note that in the model without the internal space we have similar behavior with @xmath99 replaced by @xmath282 and for a scalar field with @xmath365 the suppression is relatively weaker . in fig .
[ fig1 ] and fig .
[ fig2 ] we have plotted the energy densities induced on the left and right branes , respectively , by the presence of the second brane as functions on the size of the internal space @xmath253 and interbrane distance for a @xmath366 minimally coupled massless scalar field with the robin coefficients @xmath367 , @xmath368 .
the corresponding expressions are obtained from the general formulas of this section by the subtitutions ( [ replacefors1 ] ) .
introducing the rescaled coordinates defined by eq .
( [ scaledcoord ] ) , after the kaluza - klein reduction of the higher dimensional hilbert action , by the way similar to that in the randall - sundrum braneworld , it can be seen that effective @xmath15-dimensional newton s constant @xmath369 measured by an observer on the brane at @xmath61 and the fundamental @xmath0-dimensional newton s constant @xmath370 are related by the formula @xmath371 } e^{(d-2)k_{d}(b - j ) } , \label{gdj}\]]where@xmath372is the volume of the internal space measured by the same observer . in the orbifolded version of the model an additional factor 2 appears in the denominator of the expression on the right .
formula ( [ gdj ] ) explicitly shows two possibilities for the hierarchy generation by the redshift and large volume effects .
note that the ratio of the newton constants on the branes , @xmath373 , is the same as in the model without an internal space .
for large interbrane distances one has @xmath374 , @xmath375 , and the gravitational interactions on the brane @xmath28 are exponentially suppressed .
this feature is used in the randall - sundrum model to address the hierarchy problem . as we will see below this mechanism
also allows to obtain a naturally small cosmological constant generated by the vacuum quantum fluctuations ( for the discussion of the cosmological constant problem within the framework of braneworld models see references given in @xcite ) .
as we have already mentioned , surface energy density ( [ emt2pl3 ] ) corresponds to the gravitational source of the cosmological constant type induced on the brane at @xmath61 by the presence of the second brane . for an observer living on the brane at @xmath61 the corresponding effective @xmath15-dimensional cosmological constant
is determined by the relation @xmath376where @xmath377 is the @xmath15-dimensional effective planck mass scale for the same observer and @xmath378 is defined by eq .
( [ phi2integrated ] ) . denoting by @xmath379 the fundamental @xmath0-dimensional planck mass , @xmath380 , from eq .
( [ gdj ] ) one has the following relation@xmath381for the ratio of the effective and fundamental planck scales . from the asymptotic analysis for the induced vacuum densities given above it follows that for large interbrane distances , @xmath382 , the contribution of the modes with @xmath326 , @xmath361 is suppressed by the factor @xmath383 with respect to the contribution of the zero mode . for the contribution of the modes with @xmath384 , @xmath385 from eqs
( [ epssmallza1 ] ) , ( [ epssmallza2 ] ) one has @xmath386 . by using these relations ,
one obtains the following estimate for the ratio of the induced cosmological constant ( [ effcc ] ) to the corresponding planck scale quantity in the brane universe : @xmath387by taking into account relation ( [ planckhierarchy ] ) , this can also be written in the form@xmath388 .
\label{lambdjest1}\end{aligned}\]]for the model without an internal space this ratio is of the same order of magnitude for both branes . in the higher dimensional version of the randall - sundrum braneworld the brane at @xmath389 corresponds to the visible brane . for large interbrane distances , by taking into account eq .
( [ epssmallza2 ] ) , for the ratio of the induced cosmological constant ( [ effcc ] ) to the planck scale quantity in the corresponding brane universe one obtains@xmath390where the function @xmath391 is defined by eq .
( [ fnubet ] ) . in fig .
[ fig3 ] we have plotted the coefficient in this formula as a function of @xmath392 and @xmath393 for a @xmath366 minimally coupled massless scalar field in the model with the internal space @xmath253 and @xmath394 .
we recall that @xmath395 , where @xmath396 is the physical size of the internal space for an observer living on the brane @xmath28 .
using relation ( [ planckhierarchy ] ) with @xmath116 , we can express the corresponding interbrane distance in terms of the ratio of the planck scales @xmath397 ^{1/(d-2)}. \label{dist}\]]substituting this into eq .
( [ lambdb ] ) , for the ratio of the cosmological constant on the brane at @xmath116 to the corresponding planck scale quantity one finds@xmath398with @xmath399 .
the higher dimensional planck mass @xmath379 and ads inverse radius @xmath16 are two fundamental energy scales in the theory which in the randall - sundrum model are usually assumed to be of the same order , @xmath400 ( see , e.g. , @xcite ) . in this case one obtains the induced cosmological constant which is exponentially suppressed compared with the corresponding planck scale quantity on the visible brane . in the model with @xmath401 , @xmath402 tev , @xmath403 tev , assuming that the compactification scale on the visible brane is close to the fundamental planck scale , @xmath404 , for the ratio of the induced cosmological constant to the planck scale quantity on the visible brane we find the estimate @xmath405 . from ( [ dist ] )
one has @xmath406 and the corresponding interbranes distances generating the required hierarchy between the electroweak and planck scales are smaller than those for the model without an internal space . in the model proposed in ref .
@xcite , a separation between the fundamental planck scale and curvature scale is assumed : @xmath407 tev . under the assumption @xmath408 , in this model we have @xmath409}$ ] and @xmath410 .
on background of manifolds with boundaries the total vacuum energy is splitted into bulk and boundary parts . in the region between two branes the bulk energy per unit coordinate volume in the @xmath15-dimensional subspace is obtained by the integration of the @xmath411-component of the volume energy - momentum tensor over this region : @xmath412the surface energy per unit coordinate volume in the @xmath15-dimensional subspace , @xmath413 , is related to the surface densities evaluated in previous sections by the formula @xmath414now by making use of the formula for the volume energy - momentum tensor from ref . @xcite and the formula ( [ phi2j ] ) for the surface densities
, it can be seen that the following formal relation takes place for the unrenormalized vevs : @xmath415where @xmath416is the total vacuum energy per unit coordinate volume of the @xmath15-dimensional subspace , evaluated as the sum of zero - point energies of elementary oscillators .
the total vacuum energy within the framework of the randall - sundrum braneworld is evaluated in refs .
@xcite by the dimensional regularization method and in ref .
@xcite by the zeta function technique
@xcite consider the case of a minimally coupled scalar field in @xmath417 , and the case of arbitrary @xmath6 and @xmath242 with zero mass terms @xmath42 and @xmath43 is calculated in ref .
for the orbifolded version of the model under consideration with @xmath418 and zero mass terms on the branes , the vacuum energy is investigated in @xcite by using the dimensional regularization .
here we briefly outline the zeta function approach in the general case .
we consider the zeta function related to the vacuum energy ( [ toten1 ] ) : @xmath419where , as before , the parameter @xmath131 with dimension of mass is introduced by dimensional reasons .
the vacuum energy in the region between the branes is obtained by the analytic continuation of this zeta function to the value @xmath420:@xmath421after the evaluation of the integral over @xmath422 one obtains the formula @xmath423with the partial zeta function@xmath424to evaluate the vacuum energy we need to perform the analytic continuation of the function ( [ zetatot2 ] ) to the neighborhood of @xmath137 .
this corresponds to the analytic continuation for @xmath425 to the point @xmath426 . for this
we present the partial zeta function ( zetaparten ) as the contour integral @xmath427where the integration contour is the same as in formula ( [ intzetsx1 ] ) .
now by making use of the standard properties of the bessel functions , we see that the parts of the integrals over @xmath428 cancel and we find the following integral representation@xmath429for the further discussion it is convenient to write the partial zeta function in the decomposed form@xmath430with the separate terms@xmath431 , \label{zetapartsepar } \\
\delta \zeta _ { \beta } \left ( s\right ) & = & \frac{1}{\pi } \sin \frac{\pi s}{2}% \int_{\lambda _ { \beta } } ^{\infty } du\,(u^{2}-\lambda _ { \beta } ^{2})^{-s/2 } \notag \\ & & \times \frac{d}{du}\ln \left [ 1-\frac{\bar{i}_{\nu } ^{(a)}(uz_{a})\bar{k}% _ { \nu } ^{(b)}(uz_{b})}{\bar{k}_{\nu } ^{(a)}(uz_{a})\bar{i}_{\nu } ^{(b)}(uz_{b})}\right ] , \label{zetapartsepar1}\end{aligned}\]]where @xmath159 for @xmath115 and @xmath160 for @xmath116 .
the part @xmath432 is finite at @xmath433 and the analytic continuation is necessary for the parts @xmath434 , @xmath62 .
the latter are the partial zeta functions for the geometry of a single brane in the regions @xmath435 and @xmath436 , respectively . on the base of eq .
( [ zetanint2 ] ) similar decomposition can be given for the total zeta function@xmath437where separate parts are related to the corresponding partial zeta functions by the relations similar to eq .
( [ zetatot2 ] ) with @xmath438 defined by eq .
( [ zetaparten ] ) . in deriving the integral representation ( [ zetapartg ] ) we have assumed that there is no zero mode corresponding to @xmath148 . the contribution of the possible zero mode to the partial zeta function @xmath439 is @xmath440 . as regards to the nonzero modes , their contribution , as before , is given by the contour integral on the right of eq .
( [ zetapart2 ] ) , but now the point @xmath441 has to be avoided by the small semicircle in the right half - plane .
the integrals over the parts of the imaginary axis are transformed to the expression given by the right hand side of eq .
( [ zetapartg ] ) , whereas the integral over the small semicircle is equal to @xmath442 and cancels the contribution from the zero mode .
hence , the integral representation ( zetapartg ) for the partial zeta function is valid in the case of presence of the zero mode as well . in order to obtain an analytic continuation for the function @xmath443 , in the integrand for the non - zero modes we subtract and add the term which is obtained replacing the function @xmath444 by the first @xmath190 terms of the corresponding asymptotic expansion for large values @xmath142 . in the zero mode part we split the integral in eq .
( zetapartsepar ) into the integrals over @xmath192 and @xmath193 .
the contribution of the first integral to the zeta function is finite at @xmath445 and the part with the second integral can be treated in the way similar to that used for the non - zero modes .
after the explicit integration of the asymptotic part and introducing the notation @xmath446with @xmath447 and @xmath448 , for the corresponding zeta function the following formula is obtained @xmath449 -\gamma \left(1-\frac{s_{1}}{2}\right ) \sum_{l=-1}^{n}\frac{\tilde{w}_{l}^{(f , j)}}{z_{j}^{l+s_{1}}}% \zeta _ { \sigma } \left ( \frac{l+s_{1}}{2}\right ) \gamma \left ( \frac{l+s_{1}}{% 2}\right ) \nonumber \\ & & -\sum_{l=1}^{n}l\frac{w_{l}^{(f , j)}}{s_{1}+l } \bigg\ } \label{zetab}\end{aligned}\ ] ] where , as before , @xmath450 , and we have used the relation @xmath451 for the gamma function . in formula ( [ zetab]),@xmath452 with @xmath453 being the coefficients in the asymptotic expansion of the function @xmath454 $ ] for large values of the argument : @xmath455\sim \sum_{l=1}^{\infty } \frac{w_{l}^{(f ,
j)}}{% u^{l}},\quad j = a , b , \label{defwnul}\]]and we have defined the global zeta function @xmath456note that one has the relation @xmath457 , assuming that the coefficients in the boundary conditions are the same , and these coefficients are related to the coefficients in the similar expansions for the modified bessel functions ( see , for instance , @xcite ) . for @xmath458 both integrals on the right of formula ( [ zetab ] ) are finite at the physical point @xmath137 . for large values of @xmath200
the second integral behaves as @xmath459 and , hence , the series over @xmath83 is convergent at the point @xmath426 if @xmath460 . as a result , for these values of @xmath190 the only poles in the expression ( zetab ) are those for the function @xmath461 and the pole in the last sum corresponding to the term with @xmath218 .
note that taking @xmath462 we obtain the expansion of the zeta function over @xmath463 . in order to separate the pole and finite parts of the zeta function @xmath464 at the physical point @xmath137
, we employ the global analog of formula ( [ zetapoles ] ) : @xmath465as it has been shown in ref .
@xcite , the coefficients @xmath466 in this formula are related to the integrated seeley - dewitt coefficients @xmath467 for the massive zeta function @xmath468 by the formula@xmath469which is obtained from ( [ cpx ] ) by the integration over the internal space . for the model with the internal space @xmath253 the zeta function @xmath470
is related to the corresponding local zeta function discussed in section [ sec:1brane ] by the formula @xmath471 and the expressions for the coefficients @xmath472 and @xmath473 are directly obtained from ( [ ress1 ] ) , ( [ omegas1 ] ) . by making use of formula ( [ gamzetglob ] ) , near the point @xmath137 the zeta function @xmath474 is presented in the form of the sum of pole and finite parts:@xmath475with the residue@xmath476 .
\label{rese}\]]in formula ( [ rese ] ) the part with the first term in the square brackets comes from the nonzero kk modes along @xmath2 and the part with the second term comes from the zero mode .
now by taking into account formula ( cpcp0 ) , we see that the zero mode part is cancelled by the delta term in eq .
( [ cpcp0 ] ) . for the finite part of the zeta function
we obtain the formula @xmath477 \bigg\}\nonumber \\ & & + \frac{2\beta _ { d_{1}}}{d_{1}z_{j}^{d_{1}}}% \bigg\ { \sideset{}{'}{\sum}_{l=1}^{n}\frac{lw_{l}^{(f , j)}}{l - d_{1}}+ d_{1}w_{d_{1}}^{(f , j)}\left [ \frac{1}{2}\psi \left(\frac{d_{1}}{2}+1\right)-\frac{1}{2}\psi \left ( -\frac{1}{2}\right ) + \ln ( \mu z_{j})\right ] \bigg\ } \nonumber \\ & & + \frac{1}{(4\pi ) ^{d_{1}/2 } } \sum_{l=-1}^{n}\frac{\tilde w_{l}^{(f , j)}}{z_{j}^{l}}\left [ 2c_{(d - l)/2}\left ( \ln \mu -\frac{1}{2}\psi \left ( -\frac{1}{2}\right ) \right ) + \omega _ { ( l - d_{1 } ) /2}\right ] . \label{zetajnear-1}\end{aligned}\ ] ] where the prime on the summation sign means that the term @xmath478 should be omitted .
note that taking in this formula @xmath462 we obtain the expansion of the finite part over @xmath479 . for an internal space with the length scale
@xmath221 we can introduce the dimensionless zeta function @xmath480 with the coefficients @xmath481 and @xmath482 defined by the relation similar to eq .
( [ gamzetglob ] ) .
now it can be seen that in terms of these coefficients the last term on the right of formula ( zetajnear-1 ) is rewritten as@xmath483 .
\label{elast}\end{aligned}\]]this explicitly shows that the combination @xmath484 is a function on the ratio @xmath248 only@xmath485 similar to the case of the surface energy , for a single brane at @xmath61 we denote by @xmath486 the total vacuum energy in the region @xmath487 for @xmath239 and in the region @xmath488 for @xmath238 .
this energy is related to the zeta function considered above by the formula@xmath489where in the formulas for the function @xmath474 it should be taken @xmath490 for @xmath177 and @xmath159 for @xmath239 . in accordance with eq .
( [ polefin ] ) the regularized vacuum energy contains pole and finite parts .
the structure of the residue @xmath491 determines the form of the counterterms which should be added to the action for the renormalization of the vacuum energy . from formula ( [ rese ] )
it follows that for an internal space without boundaries the pole terms can be absorbed by adding to the brane action the counterterm @xmath492}a_{l}r_{j}^{l } , \label{countej}\]]where @xmath232 is the intrinsic curvature scalar for the brane at @xmath61 . note that this counterterm has the same structure as that for the surface energy .
the only difference is the additional summand corresponding to the term @xmath493 $ ] .
hence , by adding the counterterm ( [ countej ] ) we can absorb the pole terms for both total and surface energies . now by taking into account the possibility for the appearance of the finite renormalization terms , the renormalized vacuum energy for the geometry of a single brane is presented in the form@xmath494}c_{l}(z_{j}/l)^{2l } , \label{ejjren}\]]where @xmath495 and the coefficients @xmath496 in the finite renormalization terms should be fixed by imposing renormalization conditions . as in the case of the surface energies ,
now we see that for the internal manifold with no boundaries in the calculation of the total vacuum energy for a single brane at @xmath61 , including the contributions from the left and right regions , the pole parts of the energies cancel out for odd values @xmath242 ( assuming that the coefficients in the boundary conditions ( [ boundcond ] ) on the right and left surfaces are the same ) and we obtain a finite result . for this energy one receives @xmath497 \notag \\ & & + \frac{4\beta _ { d_{1}}}{z_{j}^{d_{1}}}\sum_{l=1}^{m}\frac{% lw_{2l}^{(i , j)}}{d_{1}(2l - d_{1})}+\frac{1}{(4\pi ) ^{d_{1}/2}}\bigg [ \sum_{l=1}^{m}\frac{4w_{2l}^{(i , j)}}{z_{j}^{2l}\gamma ( l)}\omega _ { l - d_{1}/2}-\omega _ { -d_{1}/2}\bigg ] .
\label{ejfinite}\end{aligned}\ ] ] with @xmath498 .
note that the energy per unit physical volume in the @xmath15-dimensional subspace is determined as @xmath499 and is a function on the ratio @xmath248 only . on the base of the analysis given above , for the geometry of two branes
the total energy in the region @xmath112 is presented in the form @xmath500where the interference term @xmath501 in this formula is finite for all nonzero values of the interbrane separation and is obtained directly from the part @xmath502 substituting @xmath137 .
after the integration by parts this term is presented in the form@xmath503and is not affected by finite renormalizations .
now we briefly discuss the behavior of the interference part in the vacuum energy in the asymptotic regions of the parameters . for large values of the ads radius compared with the interbrane distance , @xmath286 , by making use the uniform asymptotic expansions for the modified bessel functions , to the leading order
one finds the result for two parallel branes on the bulk @xmath504 : @xmath505where @xmath291 and @xmath506 is defined by formula ( [ cj ] ) . for large kk masses along the internal space , @xmath294 , for the contribution of the mode with a given @xmath83 one has@xmath507with @xmath508 defined by eq .
( [ cj1 ] ) .
this contribution is exponentially suppressed if @xmath298:@xmath509for small interbrane distances , satisfying the conditions @xmath286 and @xmath510 to the leading order we have the formula@xmath511where @xmath512 for @xmath513 and @xmath514 for @xmath515 .
noting that the renormalized vacuum energies for single branes are finite in the limit @xmath516 , we see that for sufficiently small interbrane distances the totall vacuum energy is dominated by the interference part . in particular
, it follows from here that for the case @xmath512 any fixation of the interbrane distance can be stable only locally . in the limit
@xmath326 assuming that @xmath361 , for the contribution of a given kk mode one has the estimate@xmath517and the contribution from this part of kk spectrum is exponentially small . for @xmath518 and for fixed @xmath519
we find@xmath520from the asymptotic analysis given above it follows that when the right brane tends to the ads horizon , @xmath347 , one has the estimate @xmath521 for the nonzero kk modes and @xmath522 for the zero kk mode .
when the left brane tends to the ads boundary , @xmath523 , the interference part of the vacuum energy behaves as @xmath524 . for small values of
the ads curvature radius corresponding to strong gravitational fields the main contribution comes from the zero mode and the interference part is suppressed by the factor @xmath525 . in figures
we have presented the dependence of the interference part of the vacuum energy on the size of the internal space @xmath253 and interbrane distance for a @xmath366 minimally coupled massless scalar field with the robin coefficients @xmath367 , @xmath368 ( fig .
[ fig4 ] ) and @xmath526 , @xmath368 ( fig .
[ fig5 ] ) .
note that , in accordance with the asymptotic analysis , for small interbrane distances this part is negative for the first case and positive for the second one .
in the second example the interference part of the vacuum energy has minimum with respect to both variables now let us check that for the separate parts of the vacuum energy the standard energy balance equation takes places .
we denote by @xmath527 the perpendicular vacuum stress on the brane integrated over the internal space .
this stress is determined by the vacuum expectation value of the @xmath528-component of the bulk energy - momentum tensor : @xmath529 .
we expect that in the presence of the surface energy the energy balance equation will be in the form @xmath530where @xmath531 is the @xmath0-volume in the bulk and @xmath532 is the @xmath242-volume on the brane @xmath61 per unit coordinate volume in the @xmath15-dimensional subspace , @xmath533combining equations ( [ enbalance])-([sj ] ) , one obtains @xmath534with @xmath535 being the perpendicular vacuum stress on the brane at @xmath61 .
the vacuum stresses normal to the branes can be presented in the form @xmath536 , @xmath62 , where @xmath537 is the stress for a single brane at @xmath61 when the second brane is absent and @xmath538 is induced by the presence of the second brane .
the latter determines the interaction forces between the branes and is defined by the formula given in ref .
@xcite . here
we consider the corresponding quantity integrated over the internal space : @xmath539with the notation @xmath540as in the case of the induced surface energy density , the formula for the interaction force can also be presented in the form@xmath541 \notag \\ & & \times \frac{\partial } { \partial z_{j}}\ln \left\vert 1-\frac{\bar{i}_{\nu } ^{(a)}(uz_{a})\bar{k}_{\nu } ^{(b)}(uz_{b})}{\bar{k}_{\nu } ^{(a)}(uz_{a})% \bar{i}_{\nu } ^{(b)}(uz_{b})}\right\vert .
\label{pjint}\end{aligned}\]]now by taking into account expressions ( [ deltepsjnew ] ) , ( [ deltaehigh ] ) , ( [ pjint ] ) , it can be explicitly checked that the interference parts in the vacuum energies and effective pressures on the branes obey the energy balance equation @xmath542the second term in the nominator on the right of this formula corresponds to the additional pressure acting on the brane due to the nonzero external curvature .
continuing our previous work @xcite , in the present paper we have investigated the expectation value of the surface energy - momentum tensor induced by the vacuum fluctuations of a bulk scalar field with an arbitrary curvature coupling parameter satisfying robin boundary conditions on two parallel branes in background spacetime @xmath1 with a warped internal space @xmath2 .
vacuum stresses on the brane are the same for both subspaces and the energy - momentum tensor on the brane corresponds to the source of the cosmological constant type in the brane universe .
it is remarkable that the latter property is valid also for the more general model with the mertic @xmath543 instead of @xmath544 in line element ( [ metric ] ) . as an regularization procedure for the surface energy density
we employ the zeta function technique . by using the residue theorem
we have constructed an integral representations for the partial zeta function corresponding to a given kk mode along the internal subspace .
this function is presented as the sum of single brane and second brane induced parts [ see formula ( [ zet12 ] ) ] .
the latter is finite at the physical point and the further analytical continuation is necessary for the first term only .
we have done this in section [ sec:1brane ] . as the first step we subtract and add to the integrand the leading terms of the corresponding asymptotic expansion for large values of the argument and explicitly integrate the asymptotic part .
further , for the regularization of the sum over the modes along the internal space we use the local zeta function related to these modes . by making use of the formula for the pole structure of this function , we have presented the energy density on a single brane as the sum of the pole and finite parts.these separate parts are determined by formulas ( [ phi2lpf1 ] ) , ( phi2lpf2 ) .
the pole parts in the surface energy density are absorbed by adding to the brane action the counterterms having the structure given by eq .
( [ counterterms ] ) .
the renormalized energy density on the corresponding surface of a single brane is determined by formula ( epsjren ) , where the second term on the right presents the finite renormalization part .
the coefficients in this part can not be determined within the model under consideration and their values should be fixed by additional renormalization conditions which relate them to observables .
unlike to the single brane part , the surface energy density induced by the presence of the second brane contains no renormalization ambiguities and is investigated in section [ sec:2brane ] .
this part is given by formula ( emt2pl3 ) , or equivalently by formula ( [ deltepsjnew ] ) .
we have investigated the induced energy density in various asymptotic regions for the parameters of the model . in the limit @xmath545
we have obtained the result for the geometry of two boundaries in the bulk @xmath546 . for the modes along @xmath2 with large kk masses , the induced energy density is exponentially small . in particular
, for sufficiently small length scales of the internal space this is the case for all nonzero kk modes and the main contribution comes from the zero mode . for small interbrane distances , to the leading order , the induced energy density is given by formula ( [ epsjsmalldist ] ) and the contribution of the induced part dominates the single brane parts . when the right brane tends to the ads horizon , @xmath347 , the induced energy density on the left brane vanishes as @xmath547 for the nonzero kk mode and like @xmath548 for the zero mode . in the same limit the corresponding energy density on the right brane behaves as @xmath549 for the nonzero kk mode and like @xmath550 for the zero mode . in the limit when the left brane tends to the ads
boundary the contribution of a given kk mode into the induced energy density vanishes as @xmath354 and as @xmath356 for the left and right branes , respectively . for small values of
the ads curvature radius corresponding to strong gravitational fields , for nonzero kk modes under the conditions @xmath551 and @xmath552 , the contribution to the induced energy density is suppressed by the factor @xmath553 .
for the zero kk mode , the corresponding energy density integrated over the internal space behaves as @xmath554 $ ] for the brane at @xmath61 and is exponentially small . in the model without the internal space we have similar behavior with @xmath99 replaced by @xmath282 and for a scalar field with @xmath555 the suppression is relatively weaker . in the model under discussion
the hierarchy between the fundamental planck scale and the effective planck scale in the brane universe is generated by the combination of redshift and large volume effects .
the corresponding efective newton s constant on the brane at @xmath61 is related to the higher - dimensional fundamental newton s constant by formula ( [ gdj ] ) ( see also eq . ( [ planckhierarchy ] ) for the ratio of the corresponding planck masses ) and for large interbrane separations is exponentially small on the brane @xmath28 .
we show that this mechanism also allows obtaining a naturally small cosmological constant generated by the vacuum quantum fluctuations of a bulk scalar . for large interbrane distances
the ratio of the induced cosmological constant to the the corresponding planck scale quantity in the brane universe is estimated by formula ( [ lambdjest1 ] ) and is exponentially small .
for the visible brane in the higher dimensional generalization of the randall - sundrum two - brane model , this ratio is given in terms of the effective and fundamental planck masses by eq .
( [ lambonb ] ) .
we have considered two classes of models with the compactification scale on the visible brane close to the fundamental planck scale .
for the first one the higher dimensional planck mass and the ads inverse radius are of the same order and in the second one a separation between these scales is assumed . in both cases
the corresponding interbrane distances generating the hierarchy between the electroweak and planck scales are smaller than those for the model without an internal space and the required suppression of the cosmological constant is obtained without fine tuning .
in section [ sec : enbal ] we have considered the total vacuum energy in the region between the branes , evaluated as the sum of zero - point energies for elementary oscillators .
we have shown that this energy is equal to the sum of surface and volume energies .
the latter is evaluated as the integral of the bulk energy density .
similar to the case of the surface energy , for the evaluation of the total vacuum energy we have used the zeta function regularization technique .
this energy is presented as the sum of the single brane and interference parts .
we have extracted the pole parts of the single brane vacuum energies and have discussed the corresponding renormalization procedure .
the divergences can be absorbed by adding the counterterms to the brane action which have the structure similar to that for the surface energy .
the interference part in the vacuum energy is given by formula ( deltaehigh ) and is regular for all nonzero interbrane distances .
we have investigated the asymptotic behavior of this part in various limiting cases . for sufficiently small interbrane distances
the total vacuum energy is dominated by the interference part given by formula ( [ deltesmalldist ] ) .
in particular , from this formula it follows that in the case @xmath512 any minimum of the vacuum energy is local and the fixation of the interbrane distance by the corresponding effective potential is stable only locally .
further , we have shown that the induced vacuum densities and vacuum effective pressures on the branes satisfy the standart energy balance equation ( [ enbalance ] ) with the inclusion of the surface terms .
the work was supported by the armenian ministry of education and science grant no . 0124 and by pve / capes program .
r. gregory , phys .
lett . * 84 * , 2564 ( 2000 ) ; t. gherghetta and m.e .
shaposhnikov , phys .
lett . * 85 * , 240 ( 2000 ) ; i. olasagasti and a. vilenkin , phys .
d * 62 * , 044014 ( 2000 ) ; t. gherghetta , e. roessl , and m.e .
shaposhnikov , phys . lett . *
b491 * , 353 ( 2000 ) ; s. randjbar - daemi and m.e .
shaposhnikov , phys .
b * 491 * , 329 ( 2000 ) ; s. randjbar - daemi and m.e .
shaposhnikov , phys .
b * 492 * , 361 ( 2000 ) ; z. chacko and a.e .
nelson , phys .
d * 62 * , 085006 ( 2000 ) ; i. oda , phys .
d * 64 * , 026002 ( 2001 ) ; p. kanti , r. madden , and k.a .
olive , phys .
d * 64 * 044021 ( 2001 ) ; e. roessl and m.e .
shaposhnikov , phys .
d * 66 * , 084008 ( 2002 ) ; z. chacko , p.j .
fox , a.e .
nelson , and n. weiner , j. high energy phys . *
03 * , 001 ( 2002 ) ; t. multamki and i. vilja , phys .
b * 545 * , 389 ( 2002 ) ; j. louko and d.l .
wiltshire , j. high energy phys .
, * 02 * , 007 ( 2002 ) ; h. davoudiasl , j.l .
hewett , and t.g .
rizzo , j. high energy phys .
, * 04 * , 001 ( 2003 ) ; s. randjbar - daemi and v. rubakov , j. high energy phys . ,
* 10 * , 054 ( 2004 ) ; g. kofinas , phys .
b * 633 * , 141 ( 2006 ) ; e. papantonopoulos , gr - qc/0601011 ; b.m.n .
carter , a.b .
nielsen , and d. wiltshire , j. high energy phys .
, * 07 * , 034 ( 2006 ) ; m. peloso , l. sorbo , and g. tasinato , phys . rev .
d * 73 * , 104025 ( 2006 ) ; m. minamitsuji , w. naylor , and m. sasaki , hep - th/0606238 ; y. sendouda , sh .
kinoshita , and sh .
mukohyama , hep - th/0607189 ; s.l .
parameswaran , s. randjbar - daemi , and a. salvio , hep - th/0608074 .
m. bordag , u. mohideen , and v.m .
mostepanenko , phys . rep . * 353 * , 1 ( 2001 ) ; k. kirsten , _ spectral functions in mathematics and physics _
( crc press , boca raton , fl , 2001 ) ; a.a .
bytsenko , g. cognola , e. elizalde , v. moretti , and s. zerbini , _ analytic aspects of quantum fields _ ( world scientific , singapore , 2003 ) . | we investigate the vacuum expectation value of the surface energy - momentum tensor for a massive scalar field with general curvature coupling parameter obeying the robin boundary conditions on two codimension one parallel branes in a @xmath0-dimensional background spacetime @xmath1 with a warped internal space @xmath2 .
these vacuum densities correspond to a gravitational source of the cosmological constant type for both subspaces of the branes . using the generalized zeta function technique in combination with contour integral representations ,
the surface energies on the branes are presented in the form of the sum of single brane and second brane induced parts .
for the geometry of a single brane both regions , on the left and on the right of the brane , are considered . at the physical point
the corresponding zeta functions contain pole and finite contributions .
for an infinitely thin brane taking these regions together , in odd spatial dimensions the pole parts cancel and the total zeta function is finite . the renormalization procedure for the surface energies and the structure of the corresponding counterterms are discussed .
the parts in the surface densities generated by the presence of the second brane are finite for all nonzero values of the interbrane separation and are investigated in various asymptotic regions of the parameters . in particular , it is shown that for large distances between the branes the induced surface densities give rise to an exponentially suppressed cosmological constant on the brane .
the total energy of the vacuum including the bulk and boundary contributions is evaluated by the zeta function technique and the energy balance between separate parts is discussed . |
looking around the world that we live in , we will find that there are complex systems everywhere .
the study of these systems has become a hot research area in different disciplines such as physics , applied mathematics , biology , engineering , economics , and social sciences . in particular ,
agent - based models have become an essential part of research on complex adaptive systems ( cas ) @xcite .
for example , self - organized phenomena in an evolving population consisting of agents competing for a limited resource , have potential applications in areas such as engineering , economics , biology , and social sciences @xcite .
the famous el farol bar attendance problem proposed by arthur @xcite constitutes a typical example of such a system in which a population of agents decide whether to go to a popular bar having limited seating capacity .
the agents are informed of the attendance in the past weeks , and hence share common information , make decisions based on past experience , interact through their actions , and in turn generate this common information collectively .
these ingredients are the key characteristics of complex systems @xcite .
the proposals of the binary versions of models of competing populations , either in the form of the minority game ( mg ) @xcite or in a binary - agent - resource ( b - a - r ) game @xcite , have led to a deeper understanding in the research in agent - based models . for modest resource levels in which there are more losers than winners , the minority game proposed by challet and zhang @xcite represents a simple , yet non - trivial , model that captures many of the essential features of such a competing population .
the mg , suitably modified , can be used to model financial markets and reproduce the stylized facts .
the b - a - r model , which is a more general model in which the resource level is taken to be a parameter , has much richer behaviour .
in particular , we will discuss the model and report the emergence of plateaux - and - jump structures in the average success rates in the population as the resource level is varied .
we analyze the results within the ideas of the trail of histories in the history space and the strategy performance ranking patterns .
the binary - agent - resource ( b - a - r ) model @xcite is a binary version of arthur s el farol bar attendance model @xcite , in which a population of @xmath0 agents repeatedly decide whether to go to a bar with limited seating based on the information of the crowd size in recent weeks . in the b - a - r model
, there is a global resource level @xmath3 which is not announced to the agents . at each timestep @xmath4
, each agent decides upon two possible options : whether to access resource @xmath3 ( action ` 1 ' ) or not ( action ` 0 ' ) .
the two global outcomes at each timestep , ` resource overused ' and ` resource not overused ' , are denoted by @xmath5 and @xmath6 .
if the number of agents @xmath7 choosing action ` @xmath6 ' exceeds @xmath3
( i.e. , resource overused and hence global outcome @xmath5 ) then the @xmath8 abstaining agents win .
by contrast if @xmath9 ( i.e. , resource not overused and hence global outcome @xmath6 ) then the @xmath7 agents win .
in order to investigate the behaviour of the system as @xmath3 changes , it is sufficient to study the range @xmath10 .
the results for the range @xmath11 can be obtained from those in the present work by suitably interchanging the role of ` 0 ' and ` 1 ' @xcite . in the special case of @xmath12 ,
the b - a - r model reduces to the minority game . in the b - a - r model
, each agent shares a common knowledge of the past history of the most recent @xmath13 outcomes , i.e. , the winning option in the most recent @xmath13 timesteps .
the agents are essential identical , except for the details in the strategies that they are holding .
the full strategy space thus consists of @xmath14 strategies , as in the mg .
initially , each agent randomly picks @xmath15 strategies from the pool of strategies , with repetitions allowed .
the agents use these strategies throughout the game . at each timestep
, each agent uses his momentarily best performing strategy with the highest virtual points .
the virtual points for each strategy indicate the cumulative performance of that strategy : at each timestep , one * virtual point * ( vp ) is awarded to a strategy that would have predicted the correct outcome after all decisions have been made , whereas it is unaltered for those with incorrect predictions .
notice that in the literature , sometimes one vp is deducted for an incorrect prediction .
the results reported here , however , come out to be the same .
a random coin - toss is used to break ties between strategies . in the b - a - r model , the agents in the population may or may not be connected , by some kind of network . in the case of a networked population @xcite ,
each agent has access to additional information from his connected neighbours , such as his neighbours strategies and/or performance . here
, we will focus our discussion on the b - a - r model in a non - networked population and report some numerical results for networked population .
the b - a - r model thus represents a general setting of a competing population in which the resource level can be tuned . from a governmental management point of view , for example , one would like to study how a population may react to a decision on increasing or decreasing a certain resource in a community .
will such a change in resource level lead to a large response in the community or the community will be rather insensitive ? to evaluate the performance of an agent , one * ( real ) point *
is awarded to each winning agent at a given timestep .
a maximum of @xmath3 points per turn can therefore be awarded to the agents per timestep .
an agent has a success rate @xmath16 , which is the mean number of points awarded to the agent per turn over a long time window . the mean success rate @xmath17 among the agents
is then defined to be the mean number of points awarded per agent per turn , i.e. , an average of @xmath16 over the agents .
we are interested in investigating the details of how the success rates , including the mean success rate and the highest success rate among the agents , change as the resource level @xmath3 varies in the efficient phase , where the number of strategies ( repetitions counted ) in play is larger than the total number of distinct strategies in the strategy space .
the effects of varying @xmath3 were first studied by johnson _ et al .
these authors reported the dependence of the fluctuations in the number of agents taking a particular option , on the memory size @xmath13 for different values of @xmath3 . for the mg ( i.e. , @xmath12 ) in the efficient phase ( i.e. , small values of @xmath13 ) the number of agents making a particular choice varies from timestep to timestep , with additional stochasticity introduced via the random tie - breaking process .
the corresponding period depends on the memory length @xmath13 .
the underlying reason is that in the efficient phase for @xmath18 , no strategy is better overall than any other .
hence there is a tendency for the system to restore itself after a finite number of timesteps , thereby preventing a given strategy s vps from running away from the others . as a result ,
the outcome bit - string shows the feature of anti - persistency or double periodicity @xcite .
since a maximum of @xmath18 points can be awarded per turn , the mean success rate @xmath17 over a sufficiently large number of timesteps is bound from above by @xmath19 .
we have carried out extensive numerical simulations on the b - a - r model to investigate the dependence of the success rate on @xmath3 for @xmath10 . unless stated otherwise , we consider systems with @xmath20 agents and @xmath21 . figure [ fig : w_m1m3s2 ] shows the results of the mean success rate ( @xmath17 , dark solid line ) as a function of @xmath3 in a typical run for ( a ) @xmath22 and ( b ) @xmath23 , together with the range corresponding to one standard deviation about @xmath17 in the success rates among the @xmath0 agents ( @xmath24 , dotted lines ) and the spread in the success rates given by the highest and the lowest success rates ( @xmath25 and @xmath26 , thin solid lines ) in the population . by taking a larger value of @xmath0 than most studies in the literature , we are able to analyze the dependence on @xmath3 and @xmath13 in greater detail and discover new features . in particular , these quantities all exhibit _ abrupt transitions _
( i.e. , jumps ) at particular values of @xmath3 . between the jumps ,
the quantities remain essentially constant and hence form steps or ` plateaux ' .
we refer to these different plateaux as states or phases , since it turns out that the jump occurs when the system makes a transition from one type of state characterizing the outcome bit - string to another .
the origins of the plateaux are due to ( i ) the finite number of states allowed in each @xmath13 , and ( ii ) insensitive to @xmath3 inside each state .
the success rates within each plateau ( state ) come out to the the same . for different runs ,
the results are almost identical . at most , there are tiny shifts in the @xmath3 values at which jumps arise due to ( i ) different initial strategy distributions among the agents in different runs , and ( ii ) different random initial history bit - strings used to start the runs . in other words ,
a uniform distribution of strategies among the agents ( e.g. , in the limit of a large population ) gives stable values of @xmath3 at the jumps .
the results indicate that a community does not react to a slight change in resource level that lies within a certain plateau , but the response will be abrupt if the change in resource level swaps through critical values between transitions .
.the states characterized by @xmath25 for the b - a - r model with @xmath20 , @xmath22 and @xmath21 .
the results are obtained from the numerical data shown in fig .
[ fig : w_m1m3s2](a ) .
[ cols="^,^,^,^",options="header " , ] as the game proceeds , the system evolves from one history bit - string to another .
this can be regarded as transitions between different nodes ( i.e. , different histories ) in the history space . for @xmath18 in the efficient phase ,
it has been shown @xcite that the conditional probability of an outcome of , say , @xmath6 following a given history is the same for all histories . for @xmath27 ,
the result still holds for states characterized by @xmath28 . note that a history bit - string can only make transitions to history bit - strings that differ by the most recent outcome , e.g. , 111 can only be make transitions to either 110 or 111 , and thus many transitions between two chosen nodes in the history space are forbidden . in addition , these allowed transitions do not in general occur with equal probabilities
this leads to specific outcome ( and history ) bit - string statistics for a state characterized by @xmath25 .
@xmath29 & 000 & 1 & 1 + & 001 & 1 & 1 + @xmath30 & 010 & 1 & 1 + & 100 & 1 & 1 + & 011 & 1 & 1 + @xmath31 &
101 & 1 & 1 + & 110 & 1 & 1 + @xmath32 & 111 & 1 & 1 + & 8 & 8 + @xmath331 + 000 & 0 & 0 + 001 & 0 & 1 + 010 & 0 & 1 + 100 & 0 & 1 + 011 & 1 & 2 + 101 & 1 & 2 + 110 & 1 & 2 + 111 & 2 & 3 + & 5 & 12 + @xmath331 + 000 & 0 & 0 + 001 & 0 & 1 + 010 & 0 & 1 + 100 & 0 & 1 + 011 & 1 & 3 + 101 & 1 & 3 + 110 & 1 & 3 + 111 & 3 & 5 + & 6 & 17 + 0.5 cm @xmath331 + 000 & 0 & 0 + 001 & 0 & 0 + 010 & 0 & 0 + 100 & 0 & 0 + 011 & 0 & 1 + 101 & 0 & 1 + 110 & 0 & 1 + 111 & 1 & 2 + & 1 & 5 + @xmath331 + 000 & 0 & 0 + 001 & 0 & 0 + 010 & 0 & 0 + 100 & 0 & 0 + 011 & 0 & 1 + 101 & 0 & 1 + 110 & 0 & 1 + 111 & 1 & 3 + & 1 & 6 + @xmath331 + 000 & 0 & 0 + 001 & 0 & 0 + 010 & 0 & 0 + 100 & 0 & 0 + 011 & 0 & 1 + 101 & 0 & 1 + 110 & 0 & 1 + 111 & 1 & 4 + & 1 & 7 + @xmath331 + 000 & 0 & 0 + 001 & 0 & 0 + 010 & 0 & 0 + 100 & 0 & 0 + 011 & 0 & 0 + 101 & 0 & 0 + 110 & 0 & 0 + 111 & 0 & 1 + & 0 & 1 + we have carried out detailed analysis of the outcomes following a given history bit - string for @xmath22 , and for each of the possible states over the whole range of @xmath3 , i.e. , we obtain the chance of getting 1 ( or 0 ) for every 3-bit history by counting from the outcome bit - string .
table [ tab : tm_m3 ] gives the _ relative numbers of occurrences _ of each outcome for every history bit - string .
for the state with @xmath34 , for example , the outcomes @xmath5 and @xmath6 occur with equal probability for every history bit - string , as in the mg .
for the other states , the results reveal several striking features .
it turns out that @xmath25 is simply given by the _ relative frequency _ of an outcome of @xmath6 in the outcome bit - string , which in turn is governed by the resource level @xmath3 .
for example , a 0 to 1 ratio of @xmath35 in the outcome bit - strings corresponds to the state with @xmath36 . in table
[ tab : tm_m3 ] , we have intentionally grouped the history bit - strings into rows according to the label @xmath37 in fig . [
fig : hs_m3 ] .
we immediately notice that for every possible state in the b - a - r model , the relative frequency of each outcome is a property of the _ group _ of histories having the same label @xmath37 rather than the individual history bit - string , i.e. , all histories in a group have the same relative fraction of a given outcome .
this observation is important in understanding the dynamics in the history space for different states in that it is no longer necessary to consider each of the @xmath38 history bit - strings in the history space .
instead , it is sufficient to consider the four groups of histories ( for @xmath22 ) as shown in fig . [
fig : hs_m3](a ) .
analysis of results for higher values of @xmath13 show the same feature .
for the state characterized by @xmath39 , the outcome bit - string is persistently @xmath6 and the path in the history space is repeatedly 111@xmath331 ( in the format of _ history _ @xmath33 _ outcome _ ) .
therefore , the path simply corresponds to an infinite number of loops around the history node 111 .
since the path is restricted to the @xmath32 node , we will also refer to this state as @xmath40 state .
the system is effectively frozen into one node in the history space . in this case
, there are effectively only _ two _ kinds of strategies in the whole strategy set , which differ by their predictions for the particular history 111 .
the difference in predictions for the other ( @xmath41 ) history bit - strings become irrelevant . obviously , the ranking in the performance of the two effective groups of strategies is such that the group of strategies that suggest an action ` 1 ' for the history 111 , outperforms the group that suggests an action ` 0 ' . for a uniform initial distribution of strategies ,
there are @xmath42 agents taking the action ` 0 ' and @xmath43 agents taking the action ` 1 ' , since half of the strategies predict 0 and half of them predict 1 . to sustain a winning outcome of 1
, the criterion is that the resource level @xmath3 should be higher than the number of agents taking the action ` 1 ' .
therefore , we have for the state with @xmath44 that @xmath45 and @xmath46 these results are in agreement with numerical results . for @xmath21 , @xmath47 for @xmath48 .
note that eqs . and are valid for _ any _ values of @xmath13 .
table [ tab : tm_m3 ] shows that the states with @xmath49 , @xmath50 , @xmath51 have very similar features in terms of the bit - string statistics .
they differ only in the frequency of giving an outcome of @xmath6 following the history of @xmath52 .
note that the @xmath53 and @xmath29 histories do not occur .
the results imply that as the system evolves , the path in history space for these states is restricted to the two groups of histories labeled by @xmath54 and @xmath31 .
the statistics show that the outcome bit - strings for the states with @xmath55 , @xmath50 and @xmath51 exhibit only one @xmath5-bit in a period of @xmath56 , @xmath57 and @xmath58 bits , respectively .
we refer to these states collectively as @xmath59 states , since the portion of allowed history space is bounded by the @xmath31 histories .
graphically , the path in history space consists of a few self - loops at the node 111 , i.e. , from 111 to 111 , then passing through the @xmath31 group of histories once and back to 111 , as shown in fig .
[ fig : hs_m3](b ) .
the states with @xmath60 , @xmath61 , @xmath62 involve the other groups of histories and exhibit complicated looping among the histories .
we refer to them collectively as higher ( i.e. , @xmath63 ) states .
the observed values of @xmath25 for @xmath64 states can also be derived by following the evolution of the strategy performance ranking pattern as the game proceeds .
we summarize the main ideas here .
details can be found in ref.@xcite .
the major result is that for given @xmath13 , the values of @xmath25 in a @xmath64 state can only take on @xmath65 where @xmath66 .
taking @xmath22 for example , we have @xmath67 and hence @xmath68 .
the values of @xmath25 are @xmath69 , @xmath50 , and @xmath51 , exactly as observed in the numerical simulations .
( [ eq : wmax_z1 ] ) is a result of the collective response of the agents stemming from their strategy selection and decision - making processes , which in turn are coupled to the strategy ranking pattern . to close the feedback mechanism ,
the strategy ranking pattern must evolve in such a way so as to be consistent with the path in the history space . recall that the @xmath59 states are those in which the system only visits the @xmath54 and @xmath70 histories , as shown in fig .
[ fig : hs_m3](b ) .
in this situation , only @xmath71 history bits in a strategy are relevant , despite each strategy has @xmath38 entries , i.e. , strategies that only differ in their predictions for the histories which do _ not _ occur are now effectively identical . therefore , many strategies have tied performance .
the key point is that a complete path of the @xmath64 states in history space corresponds to one in which there are @xmath72 turns around the @xmath32 history , i.e. , from @xmath73 for @xmath72 turns , then breaks away to visit each of the @xmath31 histories once and returns to the @xmath32 history ( see fig .
[ fig : hs_m3](b ) ) .
the value of @xmath72 that is consistent with the condition @xmath74 turns out to be @xmath66 .
we have argued that for @xmath75 , the system loops around the @xmath32 history indefinitely , since the number of agents ( @xmath76 ) persistently taking the option 1 is smaller than @xmath3 , in the large population limit . for @xmath74 ,
the situation is as follows .
as the system starts to loop around the @xmath32 history , the strategies start to split in performance , with the group of strategies predicting @xmath73 ( @xmath77 ) becomes increasingly better ( worse ) .
there is an overlap in performance between these two groups , as a result of their predictions when the system passes through the @xmath78 histories . as the number of loops at the @xmath32 history increases , the overlap in performance decreases and the number of agents taking option 1 increases towards @xmath76 , as a result of the strategy selection process .
the condition @xmath74 , thus imposes an upper bound on @xmath72 .
when the number of turns at the @xmath79 history exceeds a certain value fixed by @xmath3 , the option 1 becomes the _ losing _ option and the system breaks away to the @xmath70 histories as the system makes transition from the history 111 to the history 110 .
the lower bound on @xmath72 is set by the restriction that the strategies in the group predicting @xmath73 must perform better on the average than the group predicting @xmath80 , as @xmath81 . with this understanding on the allowed paths in the history space
, we can evaluate the number of winners in each turn of the path and obtain eq .
( [ eq : wmax_z1 ] ) for @xmath25 , with @xmath72 being the number of loops that the system makes at the @xmath32 history and @xmath66 .
we have also studied the b - a - r model in a connected population .
the agents are assumed to be connected in the form of a classical random graph ( crg ) , i.e. , random network . in a crg
, each node ( representing an agent here ) has a probability @xmath82 of connecting to another node .
the agents interact in the following way . when an agent is linked to other agents , he checks if the strategies of his connected neighbours perform better than his own strategies .
if so , he will choose the one with the highest vp among his neigbhours and then use it for decision in that turn .
if not , he will use his own best strategy .
we have carried out detailed numerical calculations of @xmath17 and @xmath83 .
figures [ fig : wmcrg_m3s2 ] and [ fig : wxcrg_m3s2 ] show how @xmath17 and @xmath83 depend on @xmath3 for various connecting probability @xmath82 , with the @xmath84 results corresponding to those in a non - networked population .
we observe that the features of plateaux - and - jumps persist in networked populations .
there are differences in the details .
while @xmath25 takes on the same set of values , the threshold in the resource level ( @xmath85 ) to sustain a particular level of success rate is found to be higher in a networked population than in a non - networked one , as shown in fig .
[ fig : wmcrg_m3s2 ] and fig .
[ fig : wxcrg_m3s2 ] . for a given resource level ,
allowing agents to share information on strategy performance may actually worsen the global performance of the system .
this is particularly clear if we inspect the results of @xmath86 in the range of @xmath87 in fig .
[ fig : wmcrg_m3s2 ] .
the addition of a small number of links will lower @xmath17 .
this behaviour is consistent with the crowd - anticrowd theory @xcite in that the links enhance the formation of crowds
. the situation can be quite different in a high resource limit @xmath88 , where a small number of links may actually be beneficial .
the reason is that for high resource level , some resource is left unused as @xmath89 agents will not have access to a strategy that predicts the winning option . using the links , some of these agents switch from losers to winners and thus enhance the overall performance of the population @xcite
we have studied numerically and analytically the effects of a varying resource level @xmath3 on the success rate of the agents in a competing population within the b - a - r model .
we found that the system passes through different states , characterized either by the mean success rate @xmath17 or by the highest success rate in the population @xmath25 , as @xmath3 decreases from the high resource level limit .
transitions between these states occur at specific values of the resource level .
we found that different states correspond to different paths covering a subspace within the whole history space . just below the high resource level
is a range of @xmath3 that gives @xmath13 states corresponding to the fractions @xmath90 , with @xmath66 .
this result is in excellent agreement with that obtained by numerical simulations .
the paths of these states in the history space are restricted to those @xmath13-bit histories with at most one - bit of 0 and with @xmath72 loops around the 111 history .
this result is derived by considering the coupling of the restricted history subspace that the system visits , the strategy performance ranking pattern , and the strategy selection process . while our analysis can also be applied to the @xmath91 states , the dynamics and the results are too complicated to be included here .
our analysis also serves to illustrate the sensitivity within multi - agent models of competing populations , to tunable parameters . by tuning an external parameter , which we take as the resource level in the present work
, the system is driven through different paths in the history space which can be regarded as a ` phase space ' of the system .
the feedback mechanism , which is built - in through the decision making process and the evaluation of the performance of the strategies , makes the system highly sensitive to the resource level in terms of which states the system decides to settle in or around .
these features are quite generally found in a wide range of complex systems .
the ideas in the analysis carried out in the present work , while specific to the b - a - r model used , should also be applicable to other models of complex systems .
while we have focused on the b - a - r model in our discussions , many of the underlying ideas are general to a wider class of complex systems . for example
, one may regard the resource level @xmath3 as a handle in controlling a driving force in the system . with @xmath18 ,
i.e. , in the mg , and the random initial distribution of strategies and random initial history , the system is allowed to diffuse from an initial node in the history space to visit all the possible histories . a deviation of the resource level from @xmath92 acts like a driving force in the history space .
thus , there is always a competition between diffusive and driven behaviour , resulting in the non - trivial behaviour in the b - a - r model and its variations .
for this reason , the present b - a - r system provides a fascinating laboratory for studying correlated , non - markovian diffusion on a non - trivial network ( i.e. , history space ) . | we aim to study the effects of controlling the resource level in agent - based models .
we study , both numerical and analytically , a binary - agent - resource ( b - a - r ) model in which @xmath0 agents are competing for resources described by a resource level @xmath1 , where @xmath2 with @xmath3 being the maximum amount of resource per turn available to the agents .
each agent picks the momentarily best - performing strategy for decision with the performance of the strategy being a result of the cumulative collective decisions of the agents .
the agents may or may not be networked for information sharing .
detailed numerical simulations reveal that the system exhibits well - defined plateaux regions in the success rate which are separated from each other by abrupt transitions . as @xmath3 increases , the maximum success rate forms a well defined sequence of simple fractions .
we analyze the features by studying the outcome time series , the dynamics of the strategies performance ranking pattern and the dynamics in the history space .
while the system tends to explore the whole history space due to its competitive nature , an increasing @xmath3 has the effect of driving the system to a restricted portion of the history space .
thus the underlying cause of the observed features is an interesting self - organized phenomena in which the system , in response to the global resource level , effectively avoids particular patterns of history outcomes .
we also compare results in networked population with those in non - networked population . * paper to be presented in the 10th annual workshop on economic heterogeneous interacting agents ( wehia 2005 ) , 13 - 15 june 2005 , university of essex , uk . * |
over the next five years the present generation of oscillation experiments at accelerators with long - baseline @xmath3 beams , k2k at kek @xcite , minos at numi beam of fnal @xcite and icarus and opera @xcite at cngs cern to gran sasso neutrino beams @xcite is expected to confirm the @xmath4 transitions observed in the atmospheric @xmath5 and measure @xmath6 and @xmath7 within @xmath8 of accuracy if @xmath9 ev@xmath10 .
k2k and minos are looking for neutrino disappearance , by measuring the @xmath3 survival probability as a function of neutrino energy while icarus and opera will search for evidence of @xmath11 interactions in conventional @xmath3 beams .
even not explicitly optimized for @xmath12 searches these experiments can explore @xmath13 oscillation parameter beyond the chooz limit @xcite .
in particular icarus and opera @xcite can reach a combined sensitivity on @xmath13 ( cp violation and matter effects not accounted for ) a factor @xmath14 better than chooz in the allowed parameter region of atmospheric neutrino oscillations for five years exposure to the cngs beam at nominal intensity @xcite .
the @xmath12 oscillations will be searched for as @xmath15 charge current events excess over the @xmath15 contamination of the beam : the key issue will be the knowledge of the neutrino beam composition and spectrum .
a strong impact on @xmath13 sensitivity is expected from the systematics of the @xmath16 ratio , and in particular from its normalization error .
the cngs beam - line design was accomplished on the basis of the previous wanf @xmath17 beam experience at cern sps @xcite for chorus and nomad experiments @xcite which allowed for a powerful study of the neutrino beam providing a strong benchmark for conventional neutrino beams and in particular for the cngs . in this sense
the cngs beam systematics on @xmath16 ratio can be derived and predicted from the previous measurements and studies performed with the wanf .
in the wanf @xmath17 facility 450 gev / c protons were extracted by the sps and sent into a segmented target composed by 11 be rods , 10 cm of length and 3 mm diameter each .
the produced positive ( negative ) mesons , essentially @xmath18 and @xmath19 , were focused ( defocused ) by two pulsed magnetic lenses ( horn and reflector ) into a 290 m vacuum decay tunnel were neutrinos were produced .
the resulting @xmath17 beam in nomad at 840 m from the be target , was characterized by an energy @xmath20 gev and a contamination of @xmath21 , @xmath22 and @xmath23 of @xmath24 , @xmath15 and @xmath25 respectively .
four main processes as sources of neutrinos were recognized : * 450 gev / c proton interactions in the be target which produce @xmath26 , @xmath27 , @xmath28 , ... originating @xmath29 of the @xmath17 flux in nomad ; * primary protons non interacting or missing the target which produce in the horn and reflector walls , windows and beam - dump , mesons only weakly or not focused by the beam - optics ( @xmath30 of @xmath31 ) ; * `` prompt neutrinos '' generated in the decay of charmed particles and kaons in the target and dump yielding @xmath32 of @xmath25 ; * particles reinteractions along the beam - line affecting @xmath33 of @xmath17 and @xmath15 but @xmath34 of defocused component @xmath35 .
-1.5 cm -0.5 cm -5.0 cm -1.3 cm -0.5 cm -0.7 cm in order to predict energy spectra , intensity and composition of the @xmath5 beam a sound knowledge of meson production in the target is required .
indeed the accurate description of the @xmath36 and @xmath28 relative to the @xmath37 is essential to calculate the @xmath15 initial content in the beam .
direct measurements of @xmath18 and @xmath19 production in be by 400 gev / c protons were performed by na20 collaboration @xcite and by the spy collaboration @xcite at 450 gev / c .
accuracy of monte carlo generators of hadronic interactions for the meson production could limit the sensitivity to neutrino oscillation searches .
the best agreement between predictions and data was found with fluka standalone code @xcite which reproduces the measured @xmath18 and @xmath19 yields at the level of @xmath38 in the momentum region @xmath39 gev / c which is expected to contribute most to @xmath5 flux @xcite .
furthermore an accurate description of the primary proton beam spot , focusing system as well as of the materials inserted in the beam line from the target to the dump ( hadronic reinteraction processes ) is also mandatory especially for the neutrino beam minor components .
a complete analysis of this @xmath5 beam was performed by a beam line simulation from the be target up to the nomad detector including the fluka generator with further corrections to the meson production based on the residual differences between the predicted and measured meson yield in be .
these reweighting functions were calculated as a function of the meson momentum by integrating the particle production over 10 mrad of angular acceptance and including also the transport efficiency as a function of production angle along the beam - line .
the @xmath40 reweighting was obtained from the corresponding for @xmath27 using a quark - counting model with 15 % of uncertainty @xcite .
the agreement with the measured @xmath5 interactions in nomad was at the few percent level ( fig .
[ nu - cc ] ) with a systematic normalization error of @xmath41 on @xmath3cc / p.o.t . essentially determined by the accuracy on the @xmath37 and @xmath36 production in the be target ( @xmath42 ) and on the proton beam spot position on
the be target .
the corresponding energy dependent error ranged between 2 and @xmath43 in the @xmath44 gev neutrino energy . due to correlations between the sources of the @xmath17 and @xmath15 fluxes , the normalization uncertainties on
@xmath45 ratio was smaller the the uncertainties on the individual @xmath17 and @xmath15 fluxes : @xmath46 @xcite .
the corresponding energy - dependent uncertainty ranged from 4 to 6 @xmath47 . -0.9
cm -7.1 cm
in the cngs neutrino beam facility ( fig . [ cngslayout ] ) the primary protons at 400 gev / c of momentum with nominal intensity of @xmath48 pot / year ( proton beam shared operations ) , will be extracted from the cern sps and sent to a carbon target , 13 graphite rods of 10 cm length , 5 and 4 mm of diameter , the first eight spaced by 9 cm .
similarly to previous wanf neutrino beam a magnetic horn and a reflector will allow the focusing ( defocusing ) of positive ( negative)charged secondaries into a @xmath49 1 km long decay tunnel where an intense @xmath17 neutrino beam is produced .
a large graphite and iron dump will absorb the residual hadrons at the end on the beam - line .
a particular attention was devoted to have the full containment of the proton beam spot in the target and to reduce as well as possible the quantity of material in the beam line in order to maximize the neutrino flux in the gran sasso site .
a full detailed simulation of the cngs beam line was performed within fluka framework . at the nominal proton beam intensity @xmath48 pot / year ( proton beam shared operations ) roughly 2900 @xmath17cc / kt / year are expected at the gran sasso site .
the increase the proton beam intensity has been studied , resulting in a possible upgrade bigger than @xmath50 @xcite .
the muon neutrino flux will be characterized by an average energy of @xmath51 gev and @xmath52 @xmath15 to @xmath3 contamination for @xmath53 gev ( fig .
[ cngs - beam ] ) .
the @xmath24 and @xmath25 component are below @xmath54 and @xmath55 respectively
. the @xmath56 intrinsic level in the beam will be below @xmath57 allowing for clean @xmath0 appearance experiments with both icarus and opera detectors .
due to the 732 km of baseline
the contribution to neutrino beam from the @xmath58 , @xmath28 is reduced by a factor @xmath59 with respect to the wanf , while those from high energy proton interactions downstream of the graphite target and from prompt charmed particles and kaons decay in the target and dump are negligible .
the @xmath15 component will be mainly produced in the @xmath60 decay instead from @xmath36 and @xmath28 as in the wanf .
the previous scheme used for the wanf neutrino beam calculation based on fluka and spy , na20 hadroproduction data in be is well suitable for the cngs neutrino neutrino beam .
the be - c scaling due to the different target material is expected to be almost independent on the secondary meson type , to depend only weakly on the transverse momentum of the meson , so contributing less than @xmath54 to the uncertainty on neutrino flux and less than @xmath61 to @xmath16 ratio .
the systematics of neutrino flux at gran sasso can be evaluated from the wanf by properly rescaling the contribution of each meson to neutrino flux according to the different production processes .
a total normalization error of @xmath62 can be predicted for both @xmath17 and @xmath15 where @xmath63 of uncertainty is the secondary production at the target and @xmath64 error is due to particle reinteractions along the beam - line .
owing to the target configuration and the 732 km base - line , the position of proton beam on the target section , the beam - line optics such as horn and reflector currents and the amount of material included in the simulation are not critical but contribute only @xmath65 to normalization error of @xmath17 and @xmath15 flux . due to correlations between the sources of the @xmath17 and @xmath15 fluxes , where both neutrinos are essentially produced from @xmath37 of different momenta , the @xmath17 by pion decay and @xmath15 from the subsequent muon decay , a large cancellation occurs in the @xmath16 error evaluation .
therefore the normalization uncertainty on @xmath16 is expected to be only @xmath66 .
the corresponding bin - to - bin energy dependent error is estimated to be @xmath67 in the relevant part of the spectrum for @xmath17 , @xmath15 and @xmath16 ratio . -1.0
the cern to gran sasso cngs neutrino beam project for @xmath0 oscillation search with icarus and opera detectors largely benefits of the wanf experience with chorus and nomad experiments . the neutrino flux can be predicted at gran sasso within a systematic uncertainty @xmath68 , the @xmath16 ratio will be known with @xmath66 normalization error and @xmath69 energy dependent error , better than quoted by icarus and opera , opening the possibility to search for @xmath1 oscillation exploring @xmath13 beyond the chooz limit .
however the sensitivity on both @xmath0 and @xmath1 oscillation channels are completely dominated by the statistics . in 5 years of cngs standard operations ,
@xmath48 pot / year , 12 @xmath56 cc events can be recognized in icarus and similarly in opera if @xmath70 ev@xmath10 . in the same time
an excess of only 21 @xmath1 cc events over 120 @xmath15cc from beam contamination and @xmath0 is expected in icarus ( 2.35 kton fiducial mass ) for @xmath71 gev if @xmath72 ( @xmath73 ) .
the foreseen increase of proton beam intensity is mandatory . | energy spectra , intensity and composition of the cern to gran sasso cngs neutrino beam for @xmath0 and @xmath1 oscillation searches are presented .
the associated beam systematics , which is the major ingredient for the @xmath2 search sensitivity , are obtained from the study of the previous cern wanf . |
we are grateful for conversations with r. akhoury , e. braaten , a. el - khadra , a. hoang , a. kronfeld , r. leigh , p. mackenzie , m. mangano , m. olsson , j. rosner , and j. stack .
this work was supported in part by department of energy grant de - fg02 - 91er40677 .
s. w. thanks the aspen center for physics where part of this work was performed .
we gratefully acknowledge the support of a gaann fellowship , under grant number de - p200a10532 , from the u. s. department of education for m. s. g. t hooft , in _ the why s of subnuclear physics _ , proceedings of the international school of subnuclear physics , erice , 1977 , ed .
a. zichichi ( plenum , new york , 1979 ) , p. 943 ; a. mueller , in _ qcd - 20 years later _ ,
proceedings of the workshop , aachen , germany , 1992 , eds .
p. zerwas and h. kastrup ( world scientific , singapore , 1993 ) , vol . 1 , | we show that the perturbation series for quarkonia energies diverges at large orders .
this results in a perturbative ambiguity in the energy that scales as @xmath0 where @xmath1 is the bohr radius of quarkonium and @xmath2 is the qcd scale parameter .
this ambiguity is associated with a nonperturbative contribution to the energy from distances of order @xmath3 and greater .
this contribution is separate from that of the gluon condensate .
-.5 in 6.5 in # 1 # 2 # 3 phys .
rev .
lett . * # 1 * , # 2 ( # 3 ) # 1 # 2 # 3 phys .
rev .
d * # 1 * , # 2 ( # 3 ) # 1 # 2 # 3 phys .
lett .
* b#1 * , # 2 ( # 3 ) # 1 # 2 # 3 nucl .
phys .
* b#1 * , # 2 ( # 3 ) * nonperturbative effects in quarkonia associated + with large orders in perturbation theory * + + department of physics + university of illinois + 1110 west green street + urbana , il 61801 + the states of quarkonia which have a size much less than the characteristic scale of nonperturbative qcd , @xmath3 , can be accurately described by qcd perturbation theory , just as positronium is accurately described by qed perturbation theory @xcite .
the quarkonium wave function is nonzero at long distances , of order @xmath3 and greater , so the states of quarkonia are also influenced by nonperturbative physics . in this article
we show that perturbation theory is aware of this nonperturbative effect via its large - order behavior .
nonperturbative effects in quarkonia have previously been paramaterized in terms of vacuum condensates . at short distances , the leading such effect is due to the gluon condensate @xcite .
we argue that this effect is separate from the nonperturbative effect we investigate , which is due to the nonzero wave function at long distances .
that perturbation theory is aware of ( some ) nonperturbative effects via its large - order behaviour is familiar from the study of infrared renormalons @xcite .
renormalons correspond to factorial growth of the coefficients of the perturbation series , and are associated with nonperturbative contributions of order @xmath4 , where @xmath5 is a characteristic momentum and @xmath6 is a positive integer .
our analysis is in the spirit of infrared renormalons .
we , however , find perturbative coefficients that grow less rapidly than factorially , and which are associated with a nonperturbative contribution of order @xmath7 , rather than a power of @xmath8 . at short distances
, quarkonia can be described by the qcd analogue of the coulomb potential @xcite , @xmath9 with a coupling that depends on the distance between the quark and the antiquark . at distances of order @xmath3 and greater
the coupling is large , and eq .
( [ pot ] ) is invalid . to study the effects of the running coupling perturbatively
, we expand the evolution equation for the coupling as an infinite sum of terms which depend on the coupling at an arbitrary fixed scale @xmath10 , @xmath11 .
@xmath12 where @xmath13 is the one - loop beta function for @xmath14 flavors of light quarks . inserting eq .
( [ run ] ) into eq .
( [ pot ] ) gives @xmath15 with @xmath16 diagrammatically , the potential in eq .
( [ potpert ] ) corresponds to the sum of vacuum - polarization insertions in the tree - level potential , as shown in fig . 1 .
in this sense , our analysis is similar to that of infrared renormalons . the first term in eq .
( [ potpert ] ) is the qcd coulomb potential with a fixed coupling .
we take this as the unperturbed potential .
its solution involves the usual coulomb energies and wave functions .
we regard the remaining terms as perturbing potentials of order @xmath17 , and evaluate their contribution using ordinary nonrelativistic time - independent perturbation theory . to evaluate the contribution of the @xmath18 perturbing potential to the energy of the state with principal quantum number @xmath19 , orbital quantum number @xmath20 , and azimuthal quantum number @xmath21 , one evaluates the matrix element @xmath22 for simplicity we momentarily restrict our attention to the ground state . for quarks of mass @xmath23 ,
the ground - state wavefunction and energy are @xmath24 @xmath25 where @xmath1 is the bohr radius , @xmath26 ( the calculation for arbitrary @xmath27 is neither more difficult nor more enlightening ; we present the result later . ) inserting eq .
( [ wave ] ) into eq .
( [ delta ] ) , dividing by @xmath28 , and integrating over angles yields @xmath29 making the change of variables @xmath30 and setting @xmath31 gives @xmath32 where @xmath33 for large @xmath34 , the integral of eq .
( [ int ] ) is dominated by the region where the argument of the exponential is a maximum .
we denote this maximum by @xmath35 , defined by the implicit equation @xmath36 we approximate the integral by the saddle - point method , which yields @xmath37 thus eq . ( [ ek ] ) becomes @xmath38 the series for the energy correction , whose @xmath18 term is given by eq .
( [ ekz0 ] ) , is divergent . for large @xmath34 ,
the size of the @xmath18 term is governed by the factor @xmath39 the term where the series begins to diverge is obtained by finding the minimum of this function with respect to @xmath34 ( keeping in mind that @xmath35 implicitly depends on @xmath34 via eq .
( [ z0 ] ) ) : @xmath40 using eq . ( [ z0 ] ) , we find that the last two terms cancel . solving for @xmath35 gives @xmath41 where the last relation follows from the definition of the qcd scale parameter @xmath2 : @xmath42 recalling that @xmath30 , eq .
( [ z0beta ] ) tells us that the series begins to diverge when the integral in eq .
( [ ekr ] ) is dominated by distances of order @xmath3 .
this feature is reminiscent of infrared renormalons . within the context of perturbation
theory , the best one can do is to truncate the series at the minimum term .
we take the size of this term as an estimate of the inherent ambiguity in the perturbative calculation of the energy . inserting eq .
( [ z0beta ] ) into eq .
( [ ekz0 ] ) yields @xmath43 the analogous calculation for the state with quantum numbers @xmath27 with wavefunction are associated laguerre polynomials . ]
@xmath44 and energy @xmath45 yields the result @xmath46 ^ 2\ , e^{-2/na\lambda } \label{dele}\ ] ] ( with @xmath1 given by eq .
( [ bohr ] ) ) .
this is the main result of the paper . in a full qcd calculation , the ambiguity in the perturbative calculation of the energy , eq .
( [ dele ] ) , would be cancelled by a related ambiguity in the nonperturbative contribution to the energy , resulting in an unambiguous expression .
this cancellation between perturbative and nonperturbative ambiguities is in the same spirit as the analogous cancellation which occurs for infrared renormalons @xcite .
thus we learn that quarkonia energies have a nonperturbative contribution which scales with @xmath47 like eq .
( [ dele ] ) . however , it is impossible to calculate the coefficient of this nonperturbative contribution within the context of perturbation theory . the ambiguity in the perturbative calculation of the energy , eq .
( [ dele ] ) , is nearly proportional to the probability density that the quarks in quarkonia are separated by a distance of @xmath3 , which is given by @xmath48 ^ 2 e^{-\rho}$ ] , with @xmath49 .
the average separation for the quarks is roughly @xmath50 .
the ambiguity increases as @xmath51 decreases , _
i.e. _ , as the size of the state increases , until @xmath52 , at which point it apparently begins to decrease . however , this value of @xmath51 corresponds to @xmath53 , where our perturbative approach is unreliable , and eq .
( [ dele ] ) should not be trusted . the ambiguity in the perturbative calculation of the energy , eq .
( [ dele ] ) , is proportional to @xmath54 rather than a power of @xmath55 ( as would arise from an infrared renormalon ) because the series for the energy correction , eq .
( [ ekz0 ] ) , diverges less rapidly than factorially .
an approximate solution to eq .
( [ z0 ] ) is @xmath56 ; thus the factor @xmath57 in eq .
( [ ekz0 ] ) is approximately @xmath58 , which grows less rapidly than factorially , @xmath59 . a similarly divergent series , resulting in a perturbative ambiguity proportional to @xmath60 , appears in the calculation of soft - gluon resummation in the production of an object of mass @xmath5 from a hadron collision of total energy @xmath61 ( @xmath62 ) @xcite .
a nonperturbative contribution to quarkonia energies from the gluon condensate has been previously studied @xcite .
this contribution is associated with an infrared renormalon arising from emission and absorption of soft gluons , as shown in fig .
2 .
the perturbative ambiguity from this renormalon is cancelled by a related ambiguity in the gluon condensate @xcite .
the gluon - condensate contribution to the quarkonia energies is @xcite @xmath63 where @xmath64 are numbers of order unity .
since @xmath65 is of order @xmath66 , this contribution is proportional to @xmath67 .
both the physical origin of eq .
( [ delcond ] ) and its functional dependence on @xmath55 show that it is a completely separate nonperturbative effect from the one considered in this paper .
the gluon condensate has been considered as the leading nonperturbative contribution to quarkonia in various analyses @xcite . for states which are small compared with @xmath3 ,
the gluon condensate , proportional to @xmath67 , is formally more important than the nonperturbative effect discussed in this paper , proportional to @xmath54 .
however , as @xmath50 approaches @xmath3 , the relative size of the gluon - condensate contribution and the nonperturbative contribution considered in this paper are unknown , since we do not know the normalization of the latter .
it is therefore important to establish the size of this nonperturbative effect . in this paper
we have shown that the calculation of quarkonia energies is divergent at large orders in perturbation theory . the associated ambiguity in the perturbative result , eq .
( [ dele ] ) , is proportional to @xmath54 , where @xmath1 is the bohr radius , @xmath19 is the principal quantum number , and @xmath2 is the qcd scale parameter .
this ambiguity is associated with a nonperturbative contribution to the energy , coming from distances of order @xmath3 and greater , which scales with @xmath47 like eq .
( [ dele ] ) .
this is a separate nonperturbative effect from the gluon - condensate contribution to quarkonia energies .
it is important to establish the size of this nonperturbative effect to judge its impact on quarkonia energies . |
recently , much attention has been paid to the investigation of the qcd phase diagram in terms of baryonic as well as isotopic ( isospin ) chemical potentials .
the reason is that dense baryonic matter which can appear in heavy - ion collision experiments has an evident isospin asymmetry .
moreover , the dense hadronic / quark matter inside compact stars is also expected to be isotopically asymmetric . to describe the above mentioned realistic situations , i.e. when the baryonic density is comparatively low , usually different nonperturbative methods or effective theories such as chiral effective lagrangians and especially nambu jona - lasinio ( njl ) type models @xcite are employed . in this way ,
the qcd phase diagram including chiral symmetry restoration @xcite , color superconductivity @xcite , and charged pion condensation ( pc ) phenomena @xcite were investigated under heavy - ion experimental and/or compact star conditions , i.e. in the presence of such external conditions as temperature , chemical potentials and possible external ( chromo)magnetic fields ( see the above references ) . obviously , the ( 3 + 1)-dimensional njl models depend on the cutoff parameter which is typically chosen to be of the order of 1 gev , so that the results of their usage are valid only at _ comparatively low energies , temperatures and densities ( chemical potentials)_. moreover , there exists also a class of renormalizable theories , the ( 1 + 1)-dimensional chiral gross neveu ( gn ) type models @xcite , model '' instead of `` chiral gn model '' for ( 1 + 1)-dimensional models with _ a continuous chiral symmetry _ , since the chiral structure of the lagrangian is the same as that of the ( 3 + 1)-dimensional njl model . ] that can be used as a laboratory for the qualitative simulation of specific properties of qcd at _ arbitrary energies_. renormalizability , asymptotic freedom , as well as the spontaneous breaking of chiral symmetry ( in vacuum ) are the most fundamental inherent features both for qcd and all gn type models . in addition , the @xmath1 phase diagram is qualitatively the same for the qcd and gn models @xcite ( here @xmath2 is the quark number chemical potential and @xmath3 is the temperature ) .
note also that the gn type models are suitable for the description of physics in quasi one - dimensional condensed matter systems like polyacetylene @xcite .
it is currently well understood ( see , e.g. , the discussion in @xcite ) that the usual _ no - go _ theorem @xcite , which generally forbids the spontaneous breaking of any continuous symmetry in two - dimensional spacetime does not work in the limit @xmath4 , where @xmath0 is the number of colored quarks .
this follows from the fact that in the limit of large @xmath0 the quantum fluctuations , which would otherwise destroy a long - range order corresponding to a spontaneous symmetry breaking , are suppressed by @xmath5 factors .
thus , the effects inherent for real dense quark matter , such as cooper pairing phenomenon ( spontaneous breaking of the continuous @xmath6 symmetry ) or charged pion condensation ( spontaneous breaking of the continuous isospin symmetry ) might be simulated in terms of a simpler ( 1 + 1)-dimensional gn - type model , though only in the leading order of the large @xmath0 approximation ( see , e.g. , @xcite and @xcite , respectively ) .
this paper is devoted to investigation of the charged pion condensation ( pc ) phenomenon in the framework of the ( 1 + 1)-dimensional njl model with two quark flavors and in the presence of the quark number ( @xmath2 ) as well as isospin ( @xmath7 ) chemical potentials .
the consideration is performed in the leading order of the @xmath5-expansion . in our previous papers @xcite the phase diagram of the above mentioned massless or massive njl@xmath8 model
was already investigated in the case of homogeneous , i.e. independent of space coordinate , order parameters ( chiral and charged pion condensates ) .
the situation corresponds to the conserved lorentz and spatial translational invariance and is adequate to physical systems in vacuum , i.e. at zero chemical potentials .
however , in dense baryonic matter , i.e. at nonzero quark number chemical potential , there might appear new phases with a spatially inhomogeneous chiral and/or charged pion condensates which destroy both chiral and/or isospin as well as spatial translational invariances of a system .
in particular , the possibility of the phase with inhomogeneous chiral condensate was discussed in the framework of both ( 1 + 1)-dimensional @xcite and ( 3 + 1)-dimensional @xcite models . at the same time the possibility of the phase with inhomogeneous charged pion condensation is less investigated .
model in @xcite .
it was claimed there that inhomogeneous pc phase is possible only at rather high values of an isotopic chemical potential , @xmath9 , where @xmath10 mev is a cutoff parameter .
so this result might be out of the scope of a model application .
moreover , the authors of @xcite made some technical simplifications in their research of the inhomogeneous pc phenomenon . ] thus , in this paper , in contrast to @xcite , we consider the phase portrait of the above mentioned massive ( 1 + 1)-dimensional njl model with two chemical potentials , @xmath2 and @xmath7 , in the leading order of the @xmath5-expansion taking into account the possibility that the charged pion condensate might become spatially inhomogeneous . the temperature is taken to be zero .
for simplicity , for the chiral condensate we use a spatially homogeneous ansatz .
( in contrast , in our previous paper @xcite the spatially inhomogeneous ansatz for the chiral condensate and homogeneous one for the charged pion condensate was used in the framework of the same massless njl@xmath8 model . )
notice once more , the isotopic asymmetry is an inevitable property of dense quark matter which might be created in heavy - ion collision experiments or inside compact stars .
so , we believe that such a simplified study in the framework of the two - dimensional njl model with isospin chemical potential could shed new light on the properties of real dense baryonic matter and will provide a deeper understanding of the charged pc phenomenon .
in particular , it is shown in our paper that in the framework of the model under consideration a pc phase with _
nonzero baryon density _ is realized just in the case of inhomogeneous charged pion condensate but not in the case of homogeneous one . in analogy , one can expect that in real ( 3 + 1)-dimensional dense hadronic / quark matter the charged pc phenomenon is realized rather with inhomogeneous pion condensate than with spatially homogeneous one .
the paper is organized as follows . in section
ii we derive , in the leading order of the large @xmath0-expansion , the general expression for the thermodynamic potential of the two - flavored massive njl@xmath8 model with quark number chemical potential @xmath2 and isospin chemical potential @xmath7 in the case of spatially homogeneous chiral condensate and inhomogeneous pc .
first , in sec .
iii we reduce our consideration to the case of homogeneous pc and find that in this case only a pc phase with _ zero density _ of quarks is possible in the model .
second , in sec .
iv it is shown that charged pc phase with _ nonzero quark density _ in the framework of the model is possible only with spatially inhomogeneous charged pion condensate .
final sec .
v presents a summary and some concluding remarks .
the discussion of some technical problems are relegated to two appendices .
we consider a ( 1 + 1)-dimensional njl@xmath8 model to mimic the phase structure of real dense quark matter composed of two massive quark flavors ( @xmath11- and @xmath12- quarks ) .
its lagrangian has the form : @xmath13q+ \frac { g}{n_c}\big [ ( \bar qq)^2+(\bar q\mathrm{i}\gamma^5\vec\tau q)^2 \big ] , \label{1}\end{aligned}\ ] ] where the quark field @xmath14 is a flavor doublet ( @xmath15 or @xmath16 ) and color @xmath0-plet ( @xmath17 ) as well as a two - component dirac spinor ( the summation in ( [ 1 ] ) over flavor , color , and spinor indices is implied ) ; @xmath18 ( @xmath19 ) are pauli matrices ; the quark number chemical potential @xmath2 in ( [ 1 ] ) is responsible for the nonzero baryonic density of quark matter , whereas the isospin chemical potential @xmath7 is taken into account in order to study properties of quark matter at nonzero isospin densities ( in this case the densities of @xmath11 and @xmath12 quarks are different ) .
the dirac gamma matrices in two - dimensional spacetime have the following form : @xmath20 evidently , the model ( [ 1 ] ) is a simple generalization of the original ( 1 + 1)-dimensional gross - neveu model @xcite with a single massless quark color @xmath0-plet to the case of two massive quark flavors and additional chemical potentials . as a result , in the case under consideration we have a modified flavor symmetry group , which depends essentially on whether the bare quark mass @xmath21 and isospin chemical potential @xmath7 take zero or nonzero values .
indeed , at @xmath22 the lagrangian ( [ 1 ] ) is invariant under transformations from the chiral @xmath23 group .
then , at @xmath24 this symmetry is reduced to @xmath25 , where @xmath26 is the third component of the isospin operator ( here and above the subscripts @xmath27 mean that the corresponding group acts only on left , right handed spinors , respectively ) .
evidently , this symmetry can also be presented as @xmath28 , where @xmath29 is the isospin subgroup and @xmath30 is the axial isospin subgroup .
quarks are transformed under these subgroups as @xmath31 and @xmath32 , respectively . in the case
@xmath33 the lagrangian ( [ 1 ] ) is invariant with respect to the @xmath34 , which is a diagonal subgroup of the chiral @xmath35 group .
finally , in the most general case with @xmath36 the initial model ( [ 1 ] ) is symmetric under the above mentioned isospin subgroup @xmath29 .
in addition , in all foregoing cases the model is invariant under color su(@xmath0)- , baryon charge @xmath37- and electric charge @xmath38 groups . the linearized version of the lagrangian ( [ 1 ] ) , which contains composite bosonic fields @xmath39 and @xmath40 @xmath41 , has the following form : @xmath42q -\frac{n_c}{4g}\big [ \sigma\sigma+\pi_a\pi_a\big ] .
\label{2}\end{aligned}\ ] ] from the lagrangian ( [ 2 ] ) one obtains the following constraint equations for the bosonic fields @xmath43 obviously , the lagrangian ( [ 2 ] ) is equivalent to the lagrangian ( [ 1 ] ) when using the constraint equations ( [ 200 ] ) .
furthermore , it is clear that the bosonic fields ( [ 200 ] ) are transformed under the isospin @xmath29 subgroup in the following manner : @xmath44 i.e the expression ( @xmath45 ) remains unchanged under transformations of the isospin subgroup @xmath29 . to avoid the _ no - go _
theorem @xcite , which forbids the spontaneous breaking of continuous symmetries in the considered case of one space dimension , we restrict the discussion only to the leading order of the large @xmath0 expansion ( i.e. to the case @xmath46 ) , where this theorem is not valid @xcite .
in particular , the effective action @xmath47 can be found in this approximation through the relation : @xmath48[dq]\exp\bigl(\mathrm{i}\int\tilde l\,d^2 x\bigr),\label{a1}\end{aligned}\ ] ] where @xmath49 is a normalization constant .
it is clear from ( [ 2 ] ) and ( [ a1 ] ) that @xmath50 where the quark contribution to the effective action , i.e. the term @xmath51 in ( [ a2 ] ) , is given as follows @xmath52[dq]\exp\bigl(\mathrm{i}\int\bar q [ \mathrm{i}\gamma^\rho\partial_\rho- m_0+\mu\gamma^0 + \nu\tau_3\gamma^0-\sigma -\mathrm{i}\gamma^5\pi_a\tau_a ] q~d^2 x\bigr ) .
\label{a3}\ ] ] here we used the notation @xmath53 .
the ground state expectation values @xmath54 and @xmath55 of the composite bosonic fields are determined by the saddle point equations , @xmath56 where @xmath57 . in vacuum ,
i.e. in the state corresponding to an empty space with zero particle density and zero values of the chemical potentials @xmath2 and @xmath7 , the quantities @xmath54 and @xmath55 do not depend on space coordinates .
however , in dense quark medium , when @xmath58 , @xmath59 , the ground state expectation values of bosonic fields might have a nontrivial dependence on @xmath60 .
in particular , in this paper we will use the following ansatz : @xmath61 where @xmath62 , and @xmath63 are constant quantities .
in fact , they are coordinates of the global minimum point of the thermodynamic potential ( tdp ) @xmath64 . ) , which lead to ground states with even lower values of the tdp . ] in the leading order of the large @xmath0-expansion it is defined by the following expression : @xmath65 which gives @xmath66[dq]\exp\bigl(\mathrm{i}\int d^2 x\bar q { \cal d } q \bigr)\right ) , \label{8}\ ] ] where @xmath67 ( remember , in this formula @xmath68 is indeed a flavor doublet , i.e. @xmath69 . ) to proceed , let us introduce in ( [ 8])-([9 ] ) the new quark doublets , @xmath70 and @xmath71 , namely : @xmath72 and @xmath73 . since this transformation of quark fields does not change the path integral measure in ( [ 8 ] ) , the expression ( [ 8 ] ) for the thermodynamic potential is easily transformed to the following one : @xmath74[d\psi]\exp\bigl(\mathrm{i}\int d^2 x\bar\psi d\psi \bigr)\right ) , \label{10}\end{aligned}\ ] ] where instead of the @xmath75dependent dirac operator @xmath76 a new @xmath75independent operator @xmath77 appears @xmath78 the expression ( [ 10 ] ) for the thermodynamic potential is easily transformed to the following one : @xmath79 where the tr - operation @xmath80 stands for the trace in spinor- ( @xmath81 ) , flavor- ( @xmath82 ) as well as two - dimensional coordinate- ( @xmath60 ) spaces , respectively , and @xmath83 is the respective trace without @xmath75space .
moreover , @xmath84 is the momentum space representation of the dirac operator @xmath77 ( [ 110 ] ) .
obviously , @xmath85 is a 4@xmath864 matrix in the direct product of the spinor and flavor spaces . since @xmath87
, one can evaluate the expression ( [ 11 ] ) with a help of any program of analytical calculations and find @xmath88 where @xmath89 and @xmath90 .
( in order to emphasize the fact that the expression ( [ 12 ] ) is divergent , i.e. unrenormalized , we use in this tdp notation the superscript un . ) obviously , the function @xmath91 is symmetric with respect to the transformation @xmath92 .
( at @xmath93 it is also invariant with respect to the @xmath94 transformation . )
moreover , it is invariant under each of the transformations @xmath95 , @xmath96 and @xmath97 . or @xmath96 transformations we perform in the integral ( [ 12 ] ) the @xmath98 or @xmath99 change of variables , respectively
, then one can easily see that the expression ( [ 12 ] ) remains intact . finally ,
if @xmath97 , we should transform @xmath100 in the integral ( [ 12 ] ) in order to be convinced that the tdp remains unchanged .
] hence , without loss of generality , we restrict ourselves to the constraints @xmath101 , @xmath102 , @xmath103 , and @xmath104 . in the following , we will investigate the global minimum point of the tdp ( [ 12 ] ) just on this region .
however , first of all let us consider the case of spatially homogeneous condensates , i.e. the @xmath105 case .
supposing that @xmath105 in ( [ 12 ] ) , we obtain after some technical calculations the tdp for the case of spatially homogeneous charged pion condensate , @xmath106\big [ ( p_0+\mu)^2-(e^-_{\delta})^2\big ] \big\ } , \label{120}\end{aligned}\ ] ] where @xmath107 the argument of the @xmath108-function in ( [ 120 ] ) is proportional to the inverse quark propagator in the energy - momentum space representation .
hence , its zeros are the poles of the quark propagator .
so , using ( [ 120 ] ) one can find the dispersion laws for quasiparticles , i.e. the momentum dependence of the quark ( @xmath109 , @xmath110 ) and antiquark ( @xmath111 , @xmath112 ) energies , in a medium ( the full expression of the quark propagator matrix is presented in appendix b of paper @xcite ) : @xmath113 integrating in ( [ 120 ] ) over @xmath114 ( see in @xcite for similar integrals ) , one obtains for the unrenormalized tdp of the system at zero temperature the following expression : @xmath115 where @xmath116 is the heaviside theta - function .
it is clear that the tdp ( [ 14 ] ) is an ultraviolet divergent quantity , so in order to get any physical information one should renormalize it , using a special dependence of such quantities as the bare coupling constant @xmath117 and the bare quark mass @xmath21 on the cutoff parameter @xmath118 ( @xmath118 restricts the integration region in the divergent integral in ( [ 14 ] ) , @xmath119 ) . the renormalization procedure for the simplest massive gn model
was already discussed in the literature , see , e.g. , in @xcite . in a similar way ,
it is easy to see that , cutting of the divergent integral in ( [ 14 ] ) and then using the substitution @xmath120 and @xmath121 , where @xmath122 and @xmath123 are new free finite renormalization group invariant massive parameters note , the quantity @xmath124 is not equal to the physical , or dynamical , quark mass @xmath125 . the last one is defined by the pole position of the quark propagator .
alternatively , it can be found as a gap , i.e. one of the coordinates of the global minimum point of the thermodynamic potential . however , parameter @xmath126 is equal to a dynamically generated quark mass in the vacuum and at @xmath93 ( a more detailed discussion on the physical essence of these parameters is given in @xcite ) . ]
( which do not depend on the cutoff @xmath118 ) , it is possible to obtain in the limit @xmath127 a finite renormalization group invariant expression for the tdp ( for details see , e.g. , the papers @xcite ) .
namely , @xmath128 where @xmath129 \label{23}\end{aligned}\ ] ] is a finite renormalization group invariant expression for the tdp ( [ 22 ] ) in vacuum , i.e. at @xmath130 and @xmath131 , taken in the chiral limit , i.e. at @xmath132 . in the following ,
when studying the phase structure , the quantity @xmath126 is still treated as a free parameter , however instead of the massive parameter @xmath124 of the model we will use a dimensionless parameter , @xmath133 . as a result , one can see that in the massive njl@xmath8 model the dimensional transmutation effect is absent formally . indeed , both before and after renormalization this massive model is parameterized by one massive- and one dimensionless quantity .
( before renormalization the model is characterized by a bare mass @xmath21 and a dimensionless bare coupling constant @xmath117 , while after renormalization the mass @xmath126 and dimensionless quantity @xmath134 are free model parameters . ) in contrast , in the massless gn - type models , i.e. at @xmath93 , the coupling constant @xmath117 is replaced after renormalization by the massive parameter @xmath126 ( it is the so - called dimensional transmutation phenomenon ) . in our subsequent calculations throughout the paper the quantity @xmath134 is fixed by @xmath135 . in this case
we have in the initial njl@xmath8 model the same relation between the pion mass and the dynamical quark mass in vacuum as in some njl - type models in the realistic case of the ( 3 + 1)-spacetime @xcite .
investigating the behavior of the global minimum point ( whose coordinates are just the gaps @xmath125 and @xmath63 ) of the tdp ( [ 22 ] ) vs chemical potentials , it is possible to establish the phase structure presented in fig .
1 . there , in the phases 1 , 2 and 3 the gap @xmath63 is vanishing , i.e. these are the normal quark matter phases with a nonzero gap @xmath125 . however , at the boundaries between phases the gap @xmath125 changes its value by a jump ( the details of the investigation , including the behavior of gaps , particle densities , meson masses etc , are presented in @xcite ) . for each point @xmath136 of the vacuum region of fig .
1 we have @xmath137 and @xmath138 ( the physical meaning of the parameter @xmath126 is described in footnote [ foot1 ] ) . finally , one can see in fig .
1 the homogeneous charged pion condensation phase in which both gaps are not equal to zero . what is more interesting for us is that all over this phase the quark number density @xmath139 is equal to zero , @xmath140 . therefore , in dense ( i.e. with nonzero @xmath141 ) quark matter , mimicked by the initial gn - type model , the phase with spatially homogeneous charged pion condensation can not be realized .
in this section , the possibility for the ground state of the initial njl@xmath8 model ( 1 ) with spatially inhomogeneous charged pion condensate is investigated .
we start with the unrenormalized tdp ( [ 12 ] ) which can be rewritten in the form @xmath143 where the notation @xmath144 as well as the identity @xmath145 with @xmath146 are used .
the argument of the @xmath147-function in ( [ 012 ] ) can be expanded into a product of four linear multipliers , @xmath148 , \label{013}\end{aligned}\ ] ] where @xmath149 are presented in appendix [ apa ] by the expressions ( [ aa1 ] ) . if the quantities @xmath150 and @xmath151 in ( [ 012 ] ) are defined by the relations ( [ 1200 ] )
, then numerical analysis shows that all the roots @xmath149 are real valued quantities vs gaps @xmath152 , chemical potentials @xmath153 and spatial momentum @xmath154 .
moreover , two of the roots , @xmath155 , are negative valued quantities .
taking into account the remark after formula ( [ 13 ] ) , it is possible to obtain immediately from ( [ 013 ] ) the quark - antiquark dispersion laws , @xmath156 note that at @xmath105 the quasiparticle energies ( [ e61 ] ) coincide with the corresponding expressions from ( [ e6 ] ) . at the particular values of the chemical potentials and gaps
the plots of the quasiparticle energies @xmath109 , ...
( [ e61 ] ) vs @xmath154 are presented in fig .
, it is possible to perform the @xmath114-integration in ( [ 013 ] ) using the general formula ( see appendix [ apb ] ) @xmath157 where @xmath158 is a real quantity . as a result
, we have @xmath159 to renormalize the expression ( [ 15 ] ) we must first regularize it . in this connection , it is necessary to make the following remark . in the case of homogeneous condensates ( see the previous section ) usually the momentum cutoff regularization scheme is used .
however it does not work in the case of spatially inhomogeneous condensates since three - momentum is no longer conserved , etc . as a result ,
an additional modification of the tdp is needed ( for details see in @xcite ) . ] . as discussed in the recent papers @xcite , an adequate regularization scheme in the case of spatially inhomogeneous phases
is that with an energy constraint equal for all quasiparticles .
so , dealing with spatial inhomogeneity , one can use , e.g. , the schwinger proper - time regularization , dimensional regularization etc . in particular , in our recent paper @xcite the _ symmetric energy cutoff regularization _
scheme was proposed in considering the behavior of chiral density waves in the presence of an external magnetic field in the framework of the four - dimensional nambu jona - lasinio model . there , for each quasiparticle the same ( finite ) interval of their energy values was allowed to contribute to the regularized thermodynamic potential . in the present investigation we will also use the symmetric energy cutoff regularization for the tdp , i.e. @xmath160 where the notations ( [ e61 ] ) for quasiparticle energies @xmath109 etc . are used . now , let us consider the identity @xmath161 clearly , at @xmath162 for all quantities @xmath163 one finds the relation @xmath164 , so that @xmath165 since the expression in parenthesis in ( [ 17 ] ) is an ultraviolet ( uv ) convergent one , i.e. it is a finite quantity in the @xmath166 limit , we see that in ( [ 17 ] ) all the uv divergences are located in the last term which is nothing but energy cutoff regularized vacuum thermodynamic potential of the system ( [ 18 ] ) .
hence , in order to renormalize the tdp @xmath167 ( [ 15 ] ) it is sufficient to remove uv divergences from the quantity ( [ 18 ] ) by substituting in ( [ 18 ] ) @xmath120 and @xmath168 by quantities with an appropriate behavior of @xmath169 vs @xmath118 . as a result
, we have @xmath170\bigg\}. \label{21}\end{aligned}\ ] ] where @xmath171 is given in ( [ 23 ] ) .
+ [ fig : condensates_massive ] we have studied numerically the tdp ( [ 21 ] ) as a function of @xmath125 , @xmath63 and @xmath172 for some physically motivated value of the massive parameter @xmath173 , where @xmath174 and @xmath126 is a free parameter of the model ( see the corresponding explanation at the end of the previous section iii ) .
the properties of its global minimum point vs chemical potentials give us the phase structure of the model which is presented in fig .
it is easy to see that the phase structure in the case of spatially inhomogeneous condensates ( see fig .
3 ) is the same as in the case of homogeneous ones ( see fig .
1 ) but with two essential exceptions . namely , the normal quark
matter phases 1 and 3 of fig .
1 are replaced by two inhomogeneous pion condensation phases ipc@xmath175 in fig .
the behavior of the gaps @xmath125 and @xmath63 as well as of the wave vector @xmath172 vs chemical potentials in these phases are presented in figs 4 - 6 .
+ the tdp ( [ 21 ] ) provides also the expressions for the quark number density @xmath141 and isospin density @xmath176 , @xmath177 in particular , as it follows from ( [ 33 ] ) and ( [ 21 ] ) @xmath178\nonumber\\&=&\int_{-\infty}^\infty\frac{dp_1}{2\pi}\left [ \theta ( \mu-\eta^{+}_1)+ \theta ( \mu-\eta^{-}_1)\right ] , \label{34}\end{aligned}\ ] ] where the last equality appeares due to the fact that @xmath179 . using these expressions
one can easily prove that in the inhomogeneous pion condensation phases ipc@xmath175 both isospin density @xmath176 and quark number density @xmath141 are nonzero ( see fig .
3 ) . in contrast , in the homogeneous pion condensation phase of fig .
3 the density @xmath141 is zero . the behavior of @xmath176 and @xmath141 at some particular values of chemical potentials are presented in figs 7 , 8 . hence , we have proved that in the framework of the initial model the charged pc phenomenon with nonzero @xmath141-density is possible only with spatially inhomogeneous charged pion condensate .
this paper is devoted to the investigation of the so - called charged pion condensation ( pc ) phenomenon which might be observed in dense baryonic matter with different contents of @xmath11 and @xmath12 quarks . to simplify the consideration ,
we have restricted ourselves to the ( 1 + 1)-dimensional njl - type model ( 1 ) with quark number @xmath2 and isospin @xmath7 chemical potentials at zero temperature .
special attention is paid to the influence of spatial inhomogeneity of different condensates on charged pc phenomenon .
our consideration is performed in the leading order of the large-@xmath0 expansion .
recall , the charged pc phenomenon was studied recently in the framework of some qcd - like effective theories such as njl models or chiral effective theories in the usual ( 3 + 1)-dimensional minkowski spacetime @xcite .
however , the existence of the charged pc phase with _ nonzero baryon or quark number density _ , denoted below as pcd phase , was there predicted without sufficient certainty .
indeed , for some values of model parameters ( the coupling constant @xmath117 , cutoff parameter @xmath118 etc . ) the pcd phase is allowed by njl models . however , for other physically interesting values of @xmath117 and @xmath118 the pcd phase is forbidden in the framework of njl models @xcite . moreover ,
if the electric charge neutrality constraint is imposed , the charged pion condensation phenomenon depends strongly on the bare ( current ) quark mass values .
in particular , it turns out that the pcd phase is forbidden in the framework of njl models if bare quark masses reach the physically acceptable values of @xmath180 mev ( see in @xcite ) . as for investigations of the charged pion condensation phenomenon in the framework of the ( 1 + 1)-dimensional massive / massless njl model ( 1 ) ,
the results of our recent papers show that the pcd phase is also absent there , if pc condensate is spatially homogeneous @xcite ( see also sec .
iii of the present paper ) .
however , earlier we have found one factor which promotes the creation of the pcd phase at least in ( 1 + 1)-dimensional spacetime .
it is the finiteness of the volume of a physical system @xcite .
since such a constraint with certain boundary conditions imposed for any system is equivalent to its consideration in a space with nontrivial topology , we have studied in @xcite the initial model ( 1 ) in the spacetime @xmath181 ( spatial coordinate is compactified ) and proved that at some boundary conditions for spinor fields the charged pcd phase is realized in the system . in this paper
, we have proved that a spatial inhomogeneity of pc condensate is also a factor which promotes the appearance of pcd phases on the phase diagram of the njl@xmath8 model ( 1 ) . indeed ,
if consideration of dense quark matter is performed in terms of homogeneous pion condensates , then pcd phase is absent on the phase diagram of the model ( 1 ) ( see fig .
however , if the spatial modulation of pion condensates is taken into account in the form ( 10 ) , then two pcd phases appear on the phase diagram of the model ( 1 ) ( those are ipc@xmath182 and ipc@xmath8 phases of fig .
3 ) . in summary , we conclude that charged pion condensation phenomenon of dense and isotopically asymmetric
quark / hadron matter is more preferable to be spatially inhomogeneous than homogeneous .
finally , we would like to discuss the reliability of the main result of our paper and try to predict what might happen with the pcd phase in the framework of a more general ansatz for condensates . to simplify the problem ,
let us consider the case of massless njl@xmath8 model ( 1 ) with @xmath93 , with an evident generalization of the ansatz ( [ 6 ] ) .
indeed , if @xmath93 , then it is possible to study the phase structure of the model in terms of the following simultaneous spatial modulations of the chiral and pion condensates @xmath183 evidently , there are three particular cases of ( [ 60 ] ) .
i ) the choice @xmath184 correspongs to spatially homogeneous condensates and the phase structure of the model for this parameterization was studied in @xcite .
ii ) then , the choice @xmath185 is really the ansatz ( [ 6 ] ) at @xmath93 .
iii ) finally , the phase structure of the model under the constraint @xmath105 was studied in @xcite .
recall that in cases i ) and iii ) the pcd phase does not appear at the phase diagram .
we have made preliminary estimations of the phase structure of the massless model ( 1 ) in the framework of the ansatz ( [ 60 ] ) and found that 1 ) the phase with @xmath186 , @xmath187 , @xmath188 , and @xmath142 is absent .
2 ) there is an absolute minimum of the tdp corresponding to spatially inhomogeneous pcd phase with @xmath188 , @xmath142 , @xmath189 , @xmath190 .
3 ) for the same values of chemical potentials there is an equivalent tdp extremum , corresponding to a chiral spiral phase , where @xmath186 , @xmath187 , @xmath191 , and @xmath192 .
it means that inside an inhomogeneous pcd phase , bubbles of the inhomogeneous phase with chiral spiral are allowed to exist and vice versa . by analogy
, one might expect that in the more physically interesting case with @xmath193 , the spatially inhomogeneous charged pcd phase would continue to be present at the phase diagram of the model ( 1 ) , even if an arbitrary more general parameterization of condensates is used .
let us denote the integral in the left hand side of ( [ int ] ) by @xmath197 ( recall , there @xmath158 is a real quantity ) .
it well - known that in quantum field theory any loop @xmath114-integration is performed in the supposition that @xmath114 is a shorthand notation for @xmath198 , where @xmath199 . in this case
the causality of the theory is preserved . taking this circumstance in mind
, we see that in ( [ int ] ) the integration contour at @xmath200 ( @xmath201 ) lies above ( below ) the singularity point @xmath202 of an integrand function .
hence , it is possible to perform in ( [ int ] ) the wick rotation of the integration contour and to direct it along the imaginary axis of the complex @xmath114-plane . in thus obtained integral one can change an integration variable , @xmath203 . as a result ,
we come to the relation @xmath204 in the last integral of ( [ b1 ] ) one can again change an integration variable , @xmath205 .
hence , @xmath206 the last integral in ( [ b2 ] ) can be easily taken using the integration by part method .
thus , up to an omitted infinite term which does not depend on @xmath158 , we obtain @xmath207 .
the authors are grateful to professor d. ebert for the support of our investigations and for many fruitful discussions and to t.g .
khunjua for his interest to our problem .
d. ebert , k.g .
klimenko , m.a .
vdovichenko and a.s .
vshivtsev , phys .
d * 61 * , 025005 ( 2000 ) ; d. ebert and k.g .
klimenko , nucl .
phys . a * 728 * , 203 ( 2003 ) .
menezes , m.b .
pinto , s.s .
avancini , a.p . martinez and c. providencia , phys .
c * 79 * , 035807 ( 2009 ) ; a. ayala , a. bashir , a. raya and a. sanchez , phys .
d * 80 * , 036005 ( 2009 ) ; n. sadooghi , arxiv:0905.2097 ; e.j .
ferrer , v. de la incera , j.p .
keith , i. portillo and p.p .
springsteen , phys .
c * 82 * , 065802 ( 2010 ) .
mizher , m.n .
chernodub and e.s .
fraga , phys .
d * 82 * , 105016 ( 2010 ) ; b. chatterjee , h. mishra and a. mishra , phys .
d * 84 * , 014016 ( 2011 ) .
f. preis , a. rebhan and a. schmitt , jhep * 1103 * , 033 ( 2011 ) ; arxiv:1109.6904 ; m. delia and f. negro , phys .
d * 83 * , 114028 ( 2011 ) ; e.v .
gorbar , v.a .
miransky and i.a .
shovkovy , arxiv:1111.3401 .
m. buballa , phys . rep . * 407 * , 205 ( 2005 ) ; i.a .
shovkovy , found .
phys . * 35 * , 1309 ( 2005 ) ; m.g .
alford , a. schmitt , k. rajagopal , and t. schfer , rev . mod .
phys . * 80 * , 1455 ( 2008 ) .
d. ebert , v.v .
khudyakov , v.c .
zhukovsky and k.g .
klimenko , jetp lett . * 74 * , 523 ( 2001 ) ; phys .
d * 65 * , 054024 ( 2002 ) ; d. blaschke , d. ebert , k.g .
klimenko , m.k .
volkov and v.l .
yudichev , phys .
d * 70 * , 014006 ( 2004 ) ; t. fujihara , d. kimura , t. inagaki and a. kvinikhidze , phys .
d * 79 * , 096008 ( 2009 ) .
ferrer and v. de la incera , phys .
d * 76 * , 045011 ( 2007 ) ; s. fayazbakhsh and n. sadooghi , phys .
d * 82 * , 045010 ( 2010 ) ; phys .
d * 83 * , 025026 ( 2011 ) ; d.t .
son and m.a .
stephanov , phys . atom .
nucl . * 64 * , 834 ( 2001 ) ; m. loewe and c. villavicencio , phys .
d * 67 * , 074034 ( 2003 ) ; arxiv:1107.3859 ;
l. he , m. jin , and p. zhuang , phys . rev .
d * 71 * , 116001 ( 2005 ) ; d.c .
duarte , r.l.s .
farias and r.o .
ramos , phys .
d * 84 * , 083525 ( 2011 ) ; d. ebert , k.g .
klimenko , a.v .
tyukov and v.c .
.zhukovsky , eur .
j. c * 58 * , 57 ( 2008 ) .
d. ebert and k.g .
klimenko , j. phys .
g * 32 * , 599 ( 2006 ) ; eur .
j. c * 46 * , 771 ( 2006 ) .
andersen and t. brauner , phys .
d * 78 * , 014030 ( 2008 ) ; j.o .
andersen and l. kyllingstad , j. phys .
g * 37 * , 015003 ( 2009 ) ; y. jiang , k. ren , t. xia and p. zhuang , arxiv:1104.0094 .
mu , l.y .
he and y.x . liu , phys . rev .
d * 82 * , 056006 ( 2010 ) .
h. abuki , r. anglani , r. gatto , g. nardulli and m. ruggieri , phys .
d * 78 * , 034034 ( 2008 ) ; h. abuki , r. anglani , r. gatto , m. pellicoro and m. ruggieri , phys .
d * 79 * , 034032 ( 2009 ) ; r. anglani , acta phys .
supp . * 3 * , 735 ( 2010 ) .
gross and a. neveu , phys .
d * 10 * , 3235 ( 1974 ) .
j. feinberg , annals phys .
* 309 * , 166 ( 2004 ) ; m. thies , j. phys .
a * 39 * , 12707 ( 2006 ) .
u. wolff , phys .
b * 157 * , 303 ( 1985 ) ; t. inagaki , t. kouno , and t. muta , int . j. mod .
phys . a * 10 * , 2241 ( 1995 ) ; s. kanemura and h .-
sato , mod .
lett . a * 10 * , 1777 ( 1995 ) .
a. chodos , h. minakata , f. cooper , a. singh , and w. mao , phys .
d * 61 * , 045011 ( 2000 ) ; k. ohwa , phys .
d * 65 * , 085040 ( 2002 ) .
a. chodos and h. minakata , phys .
a * 191 * , 39 ( 1994 ) ; h. caldas , j.l .
kneur , m.b .
pinto and r.o .
ramos , phys .
b * 77 * , 205109 ( 2008 ) ; h. caldas , nucl .
b * 807 * , 651 ( 2009 ) ; h. caldas , arxiv:1106.0948 .
v. schon and m. thies , arxiv : hep - th/0008175 ; phys .
d * 62 * , 096002 ( 2000 ) ; a. brzoska and m. thies , phys .
d * 65 * , 125001 ( 2002 ) .
n.d . mermin and h. wagner , phys .
* 17 * , 1133 ( 1966 ) ; s. coleman , commun .
phys . * 31 * , 259 ( 1973 ) .
abreu , a.p.c .
malbouisson and j.m.c .
malbouisson , europhys .
lett . * 90 * , 11001 ( 2010 ) ; phys .
d * 83 * , 025001 ( 2011 ) .
d. ebert , k.g .
klimenko , a.v .
tyukov and v.c .
zhukovsky , phys .
d * 78 * , 045008 ( 2008 ) .
d. ebert and k.g .
klimenko , phys .
d * 80 * , 125013 ( 2009 ) ; v.c .
zhukovsky , k.g .
klimenko and t.g .
khunjua , moscow univ .
bull . * 65 * , 21 ( 2010 ) .
d. ebert and k.g .
klimenko , arxiv:0902.1861 .
o. schnetz , m. thies and k. urlichs , annals phys .
* 314 * , 425 ( 2004 ) ; g. basar , g.v .
dunne and m. thies , phys .
d * 79 * , 105012 ( 2009 ) ; c. boehmer and m. thies , phys .
d * 80 * , 125038 ( 2009 ) ; j. hofmann , phys .
d * 82 * , 125027 ( 2010 ) .
f. correa , g.v .
dunne and m.s .
plyushchay , annals phys . *
324 * , 2522 ( 2009 ) ; g. basar and g.v .
dunne , jhep * 1101 * , 127 ( 2011 ) ; m.s .
plyushchay , a. arancibia and l.m .
nieto , phys .
d * 83 * , 065025 ( 2011 ) ; a. arancibia and m.s .
plyushchay , arxiv:1111.0600 ; t. kojo , arxiv:1106.2187 .
d. ebert , n.v .
gubina , k.g .
klimenko , s.g .
kurbanov and v.c .
zhukovsky , phys .
d * 84 * , 025004 ( 2011 ) .
deryagin , d.y .
grigoriev and v.a .
rubakov , int . j. mod
. phys . a * 7 * , 659 ( 1992 ) ;
m. sadzikowski and w. broniowski , phys .
b * 488 * , 63 ( 2000 ) ; w. broniowski , arxiv:1110.4063 .
e. nakano and t. tatsumi , phys .
d * 71 * , 114006 ( 2005 ) .
d. nickel , phys .
d * 80 * , 074025 ( 2009 ) ; s. carignano , d. nickel and m. buballa , phys .
d * 82 * , 054009 ( 2010 ) ; h. abuki , d. ishibashi and k. suzuki , arxiv:1109.1615 .
s. maedan , prog .
phys . * 123 * , 285 ( 2010 ) ; a. flachi , jhep * 1201 * , 023 ( 2012 ) .
i.e. frolov , k.g .
klimenko and v.ch .
zhukovsky , phys .
d * 82 * , 076002 ( 2010 ) ; moscow univ .
phys . bull .
* 65 * , 539 ( 2010 ) ; e.j .
ferrer , v. de la incera and a. sanchez , arxiv:1205.4492 .
t. kojo , y. hidaka , l. mclerran and r.d .
pisarski , nucl .
a * 843 * , 37 ( 2010 ) .
g. basar , g.v .
dunne and d.e .
kharzeev , phys .
* 104 * , 232301 ( 2010 ) .
gorbar , m. hashimoto and v.a .
miransky , phys .
lett . * 96 * , 022005 ( 2006 ) ; j.o .
andersen and t. brauner , phys .
d * 81 * , 096004 ( 2010 ) .
d. ebert , t.g .
khunjua , k.g .
klimenko and v.c .
.zhukovsky , arxiv:1106.2928 [ hep - ph ] . | the properties of two - flavored massive nambu jona - lasinio model in ( 1 + 1)-dimensional spacetime are investigated in the presence of isospin and quark number chemical potentials .
the consideration is performed in the large-@xmath0 limit , where @xmath0 is the number of colored quarks .
it is shown in the framework of this model that charged pion condensation phenomenon of dense quark / hadron isotopically asymmetric matter is rather a spatially inhomogeneous than a homogeneous one . |
mosaic ccd cameras are now becoming the wide - field imaging norm in astronomy , spurred by the practical device size fabrication limit for a reasonable yield of about 4k x 4k 15 @xmath0 m pixels or less .
the scientific prospects for these cameras are being exploited to greater degrees as the cameras and control electronics become simpler to fabricate .
programs that are already benefitting from the increased field of view include stellar population studies , galactic structure work , deep galaxy and star counts , searches for low surface brightness galaxies , and searches for gravitational lensing and microlensing .
however , inadequately aligned ccds in a mosaic detector can severely compromise the performance in these applications , especially in areas such as gravitational lensing where the lens potential must be reconstructed from the image points .
photometry on objects that span two ccds will also be limited in accuracy .
the work involved in deconvolving a large format ( 30 ) mosaic image with misaligned ccds can be daunting , injecting some degree of error into the results of the analysis .
the largest existing ccd mosaic ( 8k x 8k pixels ) uses a mounting scheme developed by luppino et al.@xmath2 .
it uses custom packages constructed for each ccd from a low expansion alloy , which are attached to a machined mounting block .
alignment screws provide the micro - adjustments to bring the ccds into position .
other mosaics have also been fabricated using a combination of precision machining and delicate assembly steps involving micromanipulation of the detectors under a high power microscope@xmath3 , or use of an alignment template@xmath4 . in the optimal case ,
the resulting assembled mosaic can have pixels along rows and columns aligned on the scale of one pixel width ( 15 @xmath0 m ) . in practice , the four element mocam detector ( canada - france - hawaii telescope ) fabricated using the luppino mounting scheme has angular misalignments between ccds of as much as 3 pixels ( greg fahlman , ubc , private communication ) .
the precision of alignment achievable using etched alignment sockets in a silicon substrate has been demonstrated previously@xmath5 , where angular misalignments are on the order of 20 ppm ( 1 @xmath0 m variation over a 5 cm substrate ) and column registration is to within a tenth of a pixel . in this technique ,
the varying atomic densities associated with the different crystal planes of silicon are exploited to achieve anisotropic chemical etching .
the resulting etched sockets have an extremely precise definition with respect to the cad designed masks .
this paper presents the first imaging results with a prototype mosaic ccd camera based on such an etched alignment technique .
an overview of the technique and construction of the prototype camera are presented in section 2 . in section 3 ,
the imaging results are discussed .
section 4 describes three techniques for assessing the alignment precision of the mosaic , and section 5 outlines future developments .
there are many issues involved , both in the lithographic fabrication of the etched substrate and the assembly of the mosaic , which are discussed in chapman et al .
1997 and chapman 1996 . here
we provide an overview as a context for the prototype camera . the silicon wafer is first etched with the desired socket layout , followed by the definition of aluminum bond pads and traces around the sockets .
semiconductor lithography techniques are used throughout to transfer mask patterns to the wafer and create etch stops .
the fabrication process is not without its difficulties .
problems with the chemical etching include residue formation , non - linear etch rates and hillocks ( small silicon growths ) , all of which can lead to misaligned or tilted ccds .
an optimized edp etchant and careful cleaning steps will alleviate most of these deleterious effects@xmath6 .
the large sockets can lead to imprecision with the metallization as they tend to funnel the photoresist used to define the etch stop .
this problem can be minimized through adjustments to some of the processing steps .
once the substrate has been fabricated and diced from the wafer , the mosaic is assembled by first bonding the fragile silicon substrate to an invar mounting plate with thermally conductive epoxy .
the substrate is heated and bonding wax applied .
the individual ccd detectors are then easily mounted and aligned against two orthogonal reference edges in the sockets , without micromanipulation under a high power microscope .
paraffin wax provides a reversible bond , flows well during application and results in a very even layer , which tends to act as a lubricant during the alignment process .
the detectors are then wire - bonded the to the substrate , or external circuit board .
the general alignment concept is shown in figure [ a ] .
the camera consists of two unthinned loral 3k x 1.5k , 15@xmath0 m pixel ccds wax bonded to an etched silicon substrate .
the flexible silicon substrate is epoxy - bonded to a polished invar36 plate to maintain flatness and prevent breakage .
the combination of bond layers , substrate and invar plate add a thermal resistance of approximately 0.1 @xmath7c / w .
the devices , substrate and mounting plate are bonded evenly over their entire area to form a solid structure . to minimize thermal expansion mismatch
stresses , the cte of the substrate matches that of the detectors ( both silicon ) , and is close to that of the invar mounting plate ( table 1 ) . at an operating temperature of -90@xmath7c , we estimate the invar
silicon thermal expansion mismatch to be about 3@xmath0 m over the 5 cm substrate length ; this has been modeled using finite element analysis ( ansys 5.3 ) which shows negligible bowing of the assembly . the precision of alignment is detailed in section 4 .
the completed prototype is shown in figure [ c ] .
the mosaic detector is mounted to a teflon ` spider ' and screwed to aluminum brackets in the detector head .
wiring for this prototype is done directly from pcb to the dewar outputs .
cooling is achieved with a cold finger attached to a liquid nitrogen ( ln@xmath8 ) tank .
the cold finger quickly cools an aluminum disk , which in turn cools the invar mounting plate .
as the thermal conductivity of the invar is not high , less robust cooling schemes ( such as thermo - electric ) may require a more complete invar surface contact with the cold finger to achieve the -80@xmath7c operating temperatures .
the hold time is roughly 9 hours .
no measurable increase in the outgassing rate was observed compared to the reference empty dewar during thermal cycling@xmath9 .
removal and replacement of ccds has proven extremely straightforward in practice .
replacement of marginal or faulty devices can be accomplished without directly handling or disturbing the other ccds .
the old wire - bonds are first removed under a microscope .
the substrate can then be heated to @xmath10 c and the problem ccd replaced . in practice
, it has taken as little as two hours to completely change two ccds and have the mosaic detector operational again .
of the seven engineering grade loral ccds were made available to us , most had nonfunctional output amplifiers or substrate shorts .
many of the devices had to be mounted and aligned in the mosaic substrate and then tested in order to find two that worked adequately .
the success of the modularity of the technique is assured by the fact that 4 ccds were replaced in the bottom socket before a working one was found .
the first images taken with this prototype camera are very promising for the continued and successful use of the etch alignment technique .
images of various extended objects and star frames were taken in the i - band filter using the ubc 42 cm telescope with the mosaic detector mounted at the cassegrain focus .
the f/13.5 focus provides an image scale of 0.5 arcsec / pixel with a total field of view of 30 arcminutes diameter .
standard reduction of the images was performed ( dark subtraction and flat fielding ) .
an existing ubc controller built by ron johnson was used for initial test purposes .
all clock lines were hardwired in parallel on the pc board .
the controller only has one channel , thus only the equivalent area of one ccd can be readout at a time .
micro - code has been written for readout of the upper , lower , or half of each ccd for any given exposure .
two frames are shown in figures [ m13b ] , and [ fielda ] , of m13 and a random star field .
the images show that cosmetics of the devices are not very good ( also see test pattern images , section 4 )
. numerous dead regions , blocked columns , and hot pixels limit the scientific uses .
nonetheless , these ccds provide a very adequate proof of concept and allow precise measurement of the degree of alignment possible .
the mechanical stability of the technique is also verified through repeated use of the camera in lab and on telescope .
in order to test the accuracy of alignment , test images and star field images were taken .
reduction of these images results in viable techniques for measurement of the relative ccd displacement . in addition
, measurements under a microscope were used as an external check on the precision .
the results of all three techniques are compared in table 2 .
the measurements are all in agreement within error estimates .
the star fields and optical measurements are considered more reliable than the test patterns for reasons discussed below . as an initial verification of the ccd angular alignment and assessment of the quality of the ccds
, various test patterns were imaged across the mosaic .
the camera was fitted with a short focal length lens capable of imaging a target a few feet away . as a test of alignment ,
the data has certain limitations : lens aberrations , non - flat test pattern , and test pattern quality .
the lens gives a vignetted field , only part of which is distortion free .
the central region of each image should give images with small aberration .
the data is reduced using the image reduction astronomical facility ( iraf ) .
the test pattern line to be reduced is scanned along each row for a well defined drop - off point in pixel intensity of 20 percent .
a linear fit to the central line of the test pattern for each ccd is performed .
the coefficients correspond to the formula @xmath11 and @xmath12 is the correlation coefficient .
the two sloped lines on each graph map the column and row of the test pattern on each ccd .
subtraction of the slope parameters , b , gives the angular misalignment of the two ccds in the small angle approximation , as shown in figure [ triangles ] .
the known distance between ccd imaging areas ( 2.13 mm ) is compared with the observed step in the y - intercept of the fit .
we find the column registration to be within half a pixel .
figure [ x ] shows the reduced test pattern data ( corresponding to figures [ xx1 ] , [ xx2 ] .
the results are as follows : test1 - 40ppm , test2 - 90ppm , test3 - 50ppm ( 20ppm corresponds to 1@xmath0 m displacement along the 5 cm substrate ) .
the spread in values for the 3 tests are more likely an indication of the tests themselves rather than of the actual misalignment .
the test patterns are all different and reduction of the data may not provide the same degree of accuracy in all cases . also , lens aberrations and image flatness may vary between the 3 tests .
we estimate the error for these results to be about @xmath133 @xmath0 m .
the images of star fields provide a powerful diagnostic of the relative ccd positions .
the same telescope and setup are used as described in section 3 . the same star field is first imaged on both ccds by moving the telescope between exposures . by calculating the transformation of one star frame to the other , the angular misalignment of the ccds can be deduced .
the column registration ( 1.6 @xmath0 m ) and inter - ccd gap ( 2.11 mm ) are found by comparison of features which overlap both ccds in some frames but not others . in principle
this technique can provide much more information about the relative placement of the ccds ( tip and tilt ) and the ccds themselves ( periodic non - uniformities and fringing across the ccd)@xmath14 .
the technique also applies equally well to mosaics of many ccds .
figure [ stars ] shows the two star fields in the same plane before mapping .
the images were reduced using iraf and the centroids of all stars were found with daophot .
a table of common sources in both frames is constructed and a transformation is found between the two frames using the routine geotran in iraf .
several transformations are found in this way using different numbers of stars in the two frames . as translation is due to movement of the telescope , the angular rotation of the frames
gives the ccd misalignment .
the average value is 1.3 @xmath0 m over the 5 cm substrate ( or 26 ppm ) .
the iraf routine geomap is used to map one star frame to the other to find an rms error of 1@xmath0 m in the transformation from the mapback accuracy .
a high magnification microscope , with a digitally metered movable eyepiece crosshair accurate to within 0.5@xmath0 m , was used to make measurements of the alignment between the detectors and the substrate sockets .
the etched socket edges were straight and aligned beyond our ability to measure them ( @xmath131@xmath0 m ) .
the angular misalignment of the ccd rows and columns with respect to the socket edge was measured to be less than 1@xmath0 m along the 5 cm long axis of each detector ( 20 parts per million ) .
the mechanical accuracy of the technique was also verified with a number of microscope measurements which are detailed in table 3 for the two - element prototype .
planar ( x , y ) measurements of the mosaic substrate were made using a photographic plate comparator . at low magnification ,
these were repeatable to about a micron .
height measurements with a calibrated z stage ( @xmath132@xmath0 m accuracy ) at the four corners of each device indicate a slight tilt to one of the devices along the long axis of the ccds , resulting in a @xmath133@xmath0 m overall flatness variation .
as this is likely due to excessive butting against the angled socket edge , it may be possible to improve the overall flatness .
there was no measurable tip to the ccds along the short axis of the ccds .
a 6 meter liquid mirror telescope ( lmt ) is being constructed near vancouver bc canada for a dedicated survey of large scale structure in the universe@xmath15 .
we are using our mosaic technique to fabricate a 4k x 4k ccd mosaic ( four 2kx2k 15@xmath0 m pixel ccds ) to operate in time - delay and integrate ( tdi ) mode@xmath16 .
a widefield corrector lens will be used to correct for distortion of star trails across the ccds resulting from the high latitude of the telescope .
alternatively , ccds can easily be angled with respect to each other using the etched socket technique , thereby minimizing the error .
the devices are 2-side buttable , but we are forced to arrange them all with the same orientation for use as a tdi imager .
we can therefore only minimize the ccd gap in the n - s orientation .
however , the performance of the mosaic is not reduced by more sizable gaps ( @xmath17 1 mm ) in the e - w direction ( readout direction ) for tdi readout .
although all 4 ccds could be aligned in a common socket , the favored design is to use thin barriers ( 100 @xmath0 m ) to maintain modularity of ccds .
maximum gaps between ccds of 100 @xmath0 m are the same as those proposed for the newest mosaic cameras being fabricated , such as an 8k x 8k pixel mosaic on the 3.5 m telescope at the apache point observatory , new mexico .
ccd spacing of 100 @xmath0 m is probably close to the fundamental limitation of the etch technique for ccd packing density .
the common socket approach would provide excellent alignment if the outer edges of the socket are used as the reference , but at the expense of maximum packing density .
alternatively , if the ccds are butted against each other , little is gained except in the wax bond and the reference point of the socket edge ; the rough ccd edge is not ideal for precision alignment . in both cases , it is very difficult to remove and replace a faulty ccd without disturbing the remaining ones .
other interests have been expressed in using the etch alignment technique to construct mosaic cameras consisting of two 2kx4k pixel ccds .
the technique appears to be adaptable to thinned devices and scaleable to the largest silicon wafer sizes . on an 8 `` wafer , eight 2k
x 4k devices can be aligned .
the 12 '' wafer awaits 3k x 6k , 15 @xmath0 m pixel ccds .
there are limitations to the technique .
both the precision of angular alignment and the registration of rows / columns depend on the wafer dicing process . since the device is butted against the socket edge , any angular misalignment of the device edge or inconsistency in the imaging area
detector edge distance , will show up as misalignment of the imaging area .
based on our measurements of our loral ccds , this appears to be a surmountable problem@xmath9 .
height variations of the individual ccd devices will show up as height variations of the composite mosaic .
the most realistic solution is to ensure from the manufacturer that the ccds destined for a mosaic all come from the same thickness wafer .
our successful construction of a prototype camera using the etch alignment technique provides a sufficient proof of concept to merit further development and use of the technique in future ccd mosaic cameras .
the fabrication process is relatively simple and economical using common lithography laboratory equipment .
three independent measurements assure the resulting composite device is flat , aligned and mechanically stable .
replacement of faulty ccds is fast and straightforward .
as far as we are aware , no other existing mosaic technique is comparable in terms of alignment achieved .
if the ccds can be diced accurately enough , there will be accurate registration between pixels and minimal angular displacement between ccds .
the overall flatness and alignment of the mosaic are within the tolerances of most astronomical observing projects .
software reduction of images should be fast and straightforward
. these devices may be suitable for other applications such as medical imaging or remote sensing where large , flat focal plane detectors are critically important .
with pleasure , we acknowledge the assistance and suggestions of mike jackson , paul hickson , and ron johnson . this research is partially supported by operating grants from the natural sciences and engineering research council of canada .
g.luppino and k.miller , a modular dewar design and detector mounting strategy for large format astronomical ccd mosaics , _ publications of the astronomical society of the pacific _ * 104 * , pp215222 ( 1992 ) .
c.l.chen , r.w.johnson , r.c.jaeger , m.b.cornelius , w.a.foster , multichip thin - film technology for low temperature packaging , _ ieee 40th electronic components and technology conference _ , pp571579 , las vegas ( 1990 ) .
technique & angular alignment & column registration + test pattern a & 2 @xmath13 3 @xmath0 m & 5 @xmath13 2 @xmath0 m + b & 4.5 @xmath0 m & 8 @xmath0 m + c & 2.5 @xmath0 m & 6 @xmath0 m + star transformations & 1.3 @xmath13 1 @xmath0 m & @xmath19 @xmath0 m + microscope & 1 @xmath13 1 @xmath0 m & @xmath20 @xmath0 m + average of 3 measures & 1.75 @xmath0 m & 3.13 @xmath0 m + parameter & value & units + device thickness a & 550 @xmath13 1 & @xmath0 m + b & 551 & + substrate thickness & 550 & @xmath0 m + socket depth & 80 & @xmath0 m + socket flatness & @xmath25 & @xmath0 m + composite device flatness & @xmath133 & @xmath0 m + device tilt a & @xmath26 & @xmath0 m + b & 0 & + device tip a , b & @xmath27 & @xmath0 m + | first imaging results are obtained with a new ccd mosaic prototype ( 3k x 3k , 15 @xmath0 m pixels ) .
the ccds are aligned using an etched socket alignment technique .
three different measurements of the alignment are made using star images , test pattern images , and microscope analysis .
the ccds have an angular misalignement of less than 30 ppm .
the composite device is flat to within @xmath1 m , with rows / columns oriented to within 20 ppm .
the use of an existing technology with built in precision reduces many of the difficulties and expenses typically encountered with mosaic detector construction .
a new camera being built for the ubc liquid mirror telescope is also described . |
the problem of compactifying the ( generalized ) jacobian of a singular curve has been studied since igusa s work @xcite around 1950 .
he constructed a compactification of the jacobian of a nodal and irreducible curve @xmath0 as the limit of the jacobians of smooth curves approaching @xmath0 .
igusa also showed that his compactification does not depend on the considered family of smooth curves .
an intrinsic characterization of the boundary points of igusa s compactification as the torsion free , rank 1 sheaves which are not line bundles is due to mumford and mayer .
the complete construction for a family of integral curves over a noetherian hensel local ring with residue field separably closed was carried out by dsouza @xcite .
one year later , altman and kleiman @xcite gave the construction for a general family of integral curves .
when the curve @xmath0 is reducible and nodal , oda and seshadri @xcite produced a family of compactified jacobians @xmath8 parameterized by an element @xmath9 of a real vector space .
seshadri dealt in @xcite with the general case of a reduced curve considering sheaves of higher rank as well . in 1994 , caporaso showed @xcite how to compactify the relative jacobian over the moduli of stable curves and described the boundary points of the compactified jacobian of a stable curve @xmath0 as invertible sheaves on certain deligne - mumford semistable curves that have @xmath0 as a stable model .
recently , pandharipande @xcite has given another construction with the boundary points now representing torsion free , rank 1 sheaves and he showed that caporaso s compactification was equivalent to his .
on the other hand , esteves @xcite constructed a compactification of the relative jacobian of a family of geometrically reduced and connected curves and compared it with seshadri s construction @xcite using theta funtions and alexeev @xcite gave a description of the jacobian of certain singular curves in terms of the orientations on complete subgraphs of the dual graph of the curve .
most of the above papers are devoted to the construction of the compactified jacobian of a curve , not to describe it .
moreover these constructions are only valid for certain projective curves .
however simpson s work @xcite on the moduli of pure coherent sheaves on projective schemes allows us to define in a natural way the jacobian of any polarized projective curve @xmath0 as the space @xmath10 of equivalence classes of stable invertible sheaves with degree @xmath2 with respect to the fixed polarization .
this is precisely the definition we adopt and we also denote by @xmath11 the space of equivalence classes of semistable pure dimension one sheaves with rank 1 on every irreducible component of @xmath0 and degree @xmath2 . in some recent papers about the moduli spaces of stable vector bundles on elliptic fibrations , for instance @xcite , @xcite , @xcite , @xcite , the jacobian in the sense of simpson of spectral curves appears .
beauville @xcite uses it as well in counting the number of rational curves on k3 surfaces .
this suggests the necessity to determine the structure of these simpson jacobians .
in this paper we give an explicit description of the structure of these simpson schemes , @xmath10 and @xmath11 of any degree @xmath2 , for @xmath0 a polarized curve of the following types : tree - like curves and all reduced and reducible curves that can appear as singular fibers of an elliptic fibration .
the paper is organized as follows .
we will work over an algebraically closed field @xmath12 of characteristic zero . in the second part
we define the simpson jacobian @xmath10 of any projective curve @xmath0 over @xmath12 , but the definition is also functorial . by simpson s work @xcite , the moduli space @xmath1 of equivalence classes of semistable pure dimension one sheaves on @xmath0 of ( polarized ) rank 1 and
degree @xmath2 is a projective scheme that contains @xmath10 and then it is a compactification of @xmath10 .
the subscheme @xmath11 is always a connected component of this compactification @xmath1 .
we show that @xmath11 is only one of the connected components that can appear in the moduli space @xmath1 ( see proposition [ p : otrascomponentes ] ) and it coincides with the compactifications constructed by other authors .
the description of the connected components of @xmath1 different to @xmath11 is still an open problem and , as we show ( see example [ e : fraccionario ] ) , it have to cover the study of moduli spaces of higher rational rank sheaves on reducible curves .
this study is not included in this paper , however when the curve @xmath0 is also reduced , as our curves , the connected component @xmath11 is projective so that it can be considered as a compactification of the simpson jacobian of @xmath0 as well .
in the third part we collect some general properties about semistable pure dimension one rank 1 sheaves on reducible curves and some results relating the picard groups of two reduced and projective curves if there is a birational and finite morphism between them . in the fourth part
we give the description ( theorem [ t : treelike ] ) of the simpson jacobian @xmath10 and of the connected component @xmath11 of its compactification when @xmath0 is a tree - like curve , that is , a projective reduced and connected curve such that the intersection points of its irreducible components are disconnecting ordinary double points , but the singularities lying only at one irreducible component can be arbitrary singularities .
the picard group of a tree - like curve is isomorphic to the direct product of the picard groups of its irreducible components .
then we use a teixidor s lemma @xcite that allows us to order the irreducible components of @xmath0 and to find subcurves @xmath13 of @xmath0 that are also tree - like curves and intersect their complements in @xmath0 at just one point . with this lemma ,
the determination of the stable rank 1 sheaves on @xmath0 which are locally free at all intersection points of the irreducible components of @xmath0 is analogous to this given by teixidor in @xcite for curves of compact type .
when the sheaf is not locally free at some intersection points , to study its stability we use a seshadri s lemma describing the stalk of a pure dimension one sheaf at the intersection points . finally for strictly semistable rank 1 sheaves , we give a recurrent algorithm that enables us to construct a jordan - hlder filtration and then to determine their @xmath14-equivalence classes .
the descriptions given here , and in particular the recurrent algorithm , will be essential in the next part for the analysis of the simpson schemes of kodaira reduced fibers of an elliptic fibration .
the aim of the fifth part is to describe the structure of the simpson jacobian and of the connected component @xmath11 of its compactification @xmath1 when @xmath0 is a reducible and reduced kodaira fiber of an elliptic fibration , that is , a fiber of type @xmath4 , @xmath5 or @xmath6 with @xmath7 . using the results of the second section ,
we first determine the picard group of these curves .
we find then necessary and sufficient conditions for a line bundle to be ( semi)stable . for the analysis of ( semi)stable pure dimension
one rank 1 sheaves which are not line bundles , the fundamental result is lemma [ l : nodelinea ] : if @xmath0 is a projective reduced and connected curve , @xmath15 a singular point at which @xmath0 is gorenstein and @xmath16 is a pure dimension one rank 1 sheaf on @xmath0 which is not invertible at @xmath15 , then there is a projective curve @xmath17 , a finite birational morphism @xmath18 and a pure dimension one rank 1 sheaf @xmath19 on @xmath17 such that @xmath20 .
hence we prove that there is an isomorphism @xmath21 between the border of the connected component @xmath11 of @xmath0 and the whole component @xmath22 of the curve @xmath17 .
when @xmath0 is a fiber of type @xmath4 or @xmath6 , the curve @xmath17 is a tree - like curve and then the structure of this border is determined by previous results . if @xmath0 is a fiber of type @xmath5 , that is , three rational smooth curves meeting at one point @xmath15 , the curve @xmath17 is also the union of three rational smooth curves meeting at one point
but in such a way that the curve @xmath17 can not be embedded ( even locally ) in a smooth surface ( see figure 3 ) .
the study of simpson schemes for this curve @xmath17 is in the subsection [ s : curvax ] .
lemma [ l : localizacion2 ] , whose proof is a generalization of the proof given by seshadri for lemma [ l : localizacion ] in this paper , plays a fundamental role in the most difficult part of the analysis , the study of ( semi)stable pure dimension one rank 1 sheaves on @xmath17 which are not line bundles .
subsection [ s : curvax ] completes then the description of the simpson schemes for the curves of type @xmath5 .
we finish the paper with a special study for the case of degree zero .
this case is very interesting because the polarization does not influence semistability conditions ( corollary [ c : nopolarizacion ] ) .
we also prove that if @xmath0 is a fiber of type @xmath4 , @xmath5 or @xmath6 , all strictly semistable pure dimension one rank 1 sheaves of degree 0 on @xmath0 are in the same @xmath14-equivalence class ( corollary [ c : equivalentes ] ) and that the moduli space of semistable pure dimension one sheaves of rank 1 and degree 0 on @xmath0 is never a fine moduli space ( corollary [ c : nofino ] ) .
these results can be found in @xcite for a curve @xmath0 of type @xmath23 where cldraru use them to prove that the connected component @xmath24 is isomorphic to a rational curve with one node .
if @xmath25 is an elliptic fibration with reduced fibers , the global structure of the compactified simpson jacobian @xmath26 is not totally known ( when the fibers of @xmath27 are geometrically integral some results can be found in @xcite ) .
we hope that the descriptions of the compactified simpson jacobians of degree zero given here ( propositions [ p : grado0 ] and [ p : propiasemi ] ) will be very useful to study the singular fibers of @xmath28 , to know the singularities of the variety @xmath26 and to find , using the results of @xcite , examples of non isomorphic elliptic fibrations having isomorphic derived categories .
two consequences of the given descriptions are the following : the first is that for reducible curves it is not true in general that the tensor product of semistable pure dimension one sheaves of rank 1 is semistable ( see example [ e : prodtens ] ) and the second is that the pullback of a stable pure dimension one rank 1 sheaf by a finite morphism of reducible curves is not stable either ( see example [ e : imageninversa ] ) .
these two statements constitute the principal problem to generalize to reducible curves the study of abel maps given by esteves , gagn and kleiman in @xcite for families of integral curves .
the results presented here are part of my ph .
d. thesis .
i am very grateful to my advisor d. hernndez ruiprez , who introduced me to the problem of compactifying the generalized jacobian of a curve , for his invaluable help and his constant encouragement .
i could not done this work without his support .
let @xmath0 be a projective curve over @xmath12 .
let @xmath29 be an ample invertible sheaf on @xmath0 , let @xmath3 be the associated polarization and let @xmath30 be the degree of @xmath3 .
let @xmath16 be a coherent sheaf on @xmath0 .
we say that @xmath16 is _ pure of dimension one _ or _ torsion free _ if for all nonzero subsheaves @xmath31 the dimension of @xmath32 is 1 .
the ( polarized ) _ rank _ and _ degree _ with respect to @xmath3 of @xmath16 are the rational numbers @xmath33 and @xmath34 determined by the hilbert polynomial @xmath35 the _ slope _ of @xmath16 is defined by @xmath36 the sheaf @xmath16 is _ stable _ ( resp . _
semistable _ ) with respect to @xmath3 if @xmath16 is pure of dimension one and for any proper subsheaf @xmath31 one has @xmath37 in @xcite simpson defined the multiplicity of @xmath16 as the integer number @xmath38 and the slope as the quotient @xmath39 stability and semistability considered in terms of simpson s slope and in terms of @xmath40 are equivalent .
we adopt these definitions of rank and degree of @xmath16 because they coincide with the classical ones when the curve @xmath0 is integral .
note however that if @xmath0 is not integral , the rank and the degree of a pure dimension one sheaf are not in general integer numbers .
according to the general theory , for every semistable sheaf @xmath16 with respect to @xmath3 there is a _ jordan - hlder filtration _
@xmath41 with stable quotients @xmath42 and @xmath43 for @xmath44 .
this filtration need not be unique , but _ the graded object _
@xmath45 does not depend on the choice of the jordan - hlder filtration .
two semistable sheaves @xmath16 and @xmath46 on @xmath0 are said to be _ @xmath14-equivalent _ if @xmath47 .
observe that two stable sheaves are @xmath14-equivalent only if they are isomorphic . in the relative case , given a scheme @xmath14 of finite type over @xmath12 , a projective morphism of schemes @xmath48 whose geometric fibers are curves and a relative polarization @xmath3
, we define the _ relative rank _ and _ degree _ of a coherent sheaf @xmath16 on @xmath0 , flat over @xmath14 , as its rank and degree on fibers , and we say that @xmath16 is _ relatively pure of dimension one _ ( resp . _ stable _ , resp .
_ semistable _ ) if it is flat over @xmath14 and if its restriction to every geometric fiber of @xmath49 is pure of dimension one ( resp .
stable , resp .
semistable ) .
let @xmath50 ( resp .
@xmath51 ) be the functor which to any @xmath14-scheme @xmath52 associates the set of equivalence classes of stable locally free ( resp .
relatively pure dimension one ) sheaves on @xmath53 with relative rank @xmath54 and degree @xmath2 .
two such sheaves @xmath16 and @xmath46 are said to be equivalent if @xmath55 , where @xmath56 is a line bundle on @xmath52 and @xmath57 is the natural projection .
similarly , we define the functor @xmath58 ( resp .
@xmath59 ) of semistable locally free ( resp .
relatively pure dimension one ) sheaves of relative rank @xmath54 and degree @xmath2 . as a particular case of the simpson s work @xcite
, there exists a projective scheme @xmath60 which coarsely represents the functor @xmath59 .
rational points of @xmath61 correspond to @xmath14-equivalence classes of semistable torsion free sheaves of rank @xmath54 and degree @xmath2 on @xmath62 ( @xmath63 ) .
moreover , @xmath58 is coarsely represented by a subscheme @xmath64 and there are open subschemes @xmath65 and @xmath66 which represent the other two functors .
[ d : jacobiana ] the simpson jacobian of degree @xmath2 of the curve @xmath0 is @xmath67 .
we denote @xmath68 .
the projective scheme @xmath1 is the compactification of the simpson jacobian of @xmath0 of degree @xmath2 . when @xmath0 is an integral curve every torsion free rank 1 sheaf is stable , and then @xmath10 is equal to the picard scheme @xmath69 and @xmath1 coincides with altman - kleiman s compactification @xcite .
in some papers about jacobians of non irreducible curves , for instance @xcite , @xcite , @xcite , @xcite , torsion free rank 1 sheaves are considered as those pure dimension one sheaves having rank 1 on every irreducible component of the curve .
if the notion of rank is given by the hilbert polynomial , there can be , depending on the degree of the fixed polarization @xmath3 on the curve @xmath0 , pure dimension one sheaves of rank 1 whose restrictions to some irreducible components of @xmath0 are not of rank 1 .
let @xmath0 be the nodal curve which is a union of two smooth curves
@xmath70 and @xmath71 meeting transversally at one point @xmath15 .
let @xmath3 be a polarization on @xmath0 such that @xmath72 and let @xmath16 be a locally free sheaf on @xmath70 of rank 2 and degree @xmath2 .
let us denote by @xmath73 the inclusion map .
the sheaf @xmath74 is pure of dimension one on @xmath0 and , since @xmath75 one has @xmath76 .
however , the restriction of @xmath74 to @xmath71 is a torsion sheaf supported at @xmath15 . in order to avoid the confusion
, we will say that a sheaf @xmath16 on @xmath0 is of _ polarized rank 1 _ if @xmath16 has rank 1 with the hilbert polynomial , that is , @xmath77 , whereas by _ rank 1 _ sheaves we mean those sheaves that have rank 1 on every irreducible component of @xmath0 .
semistable pure dimension one rank 1 sheaves of degree @xmath2 have polarized rank 1 with respect to any polarization and they are a connected component of the compactification @xmath1 of the simpson jacobian of @xmath0 .
let us denote by @xmath11 this connected component .
depending on the degree of the polarization @xmath3 on @xmath0 , @xmath11 need not be the unique connected component of the moduli space @xmath1 .
the following proposition shows all connected components that can appear in @xmath1 when @xmath0 is a union of two integral curves meeting transversally only at one point .
if @xmath0 is a projective and reduced curve with irreducible components @xmath78 and @xmath79 are the intersection points of @xmath78 , it is known ( see @xcite ) that for every pure dimension one sheaf @xmath16 on @xmath0 there is an exact sequence @xmath80 where we denote @xmath81 and @xmath52 is a torsion sheaf whose support is contained in the set @xmath82 .
[ p : otrascomponentes ] let @xmath83 be a projective curve with @xmath84 integral curves for @xmath85 and @xmath86 .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 and let @xmath87 be the degree of the induced polarization @xmath88 on @xmath84 for @xmath85 .
it holds that 1 . if @xmath30 is not a multiple of @xmath87 for @xmath85 , then the only connected component of @xmath1 is @xmath11 .
if @xmath89 only for one @xmath90 or @xmath91 , then @xmath92 where @xmath93 .
3 . if @xmath94 , then @xmath95 where @xmath96 and @xmath97 .
let @xmath16 be a semistable pure dimension one sheaf on @xmath0 of polarized rank 1 and degree @xmath2 with respect to @xmath3 . from the above exact sequence , we have @xmath98 thus , if @xmath99 for @xmath85 , the sheaf @xmath16 is of rank 1 and its @xmath14-equivalence class , denoted by @xmath100 $ ] , belongs to @xmath11 . since @xmath101 , if @xmath102 ( resp .
@xmath103 ) is zero , then @xmath104 ( resp .
@xmath105 ) must divide to @xmath30 , namely @xmath106 ( resp .
@xmath107 ) . in this case
, one has @xmath108 ( resp .
@xmath109 ) so that @xmath16 is a semistable pure dimension one sheaf on @xmath71 ( resp .
@xmath70 ) of rank @xmath110 ( resp .
@xmath54 ) and degree @xmath111 ( resp .
@xmath112 ) with respect to @xmath113 ( resp .
@xmath114 ) .
therefore , the @xmath14-equivalence class of @xmath16 belongs to @xmath115 ( resp .
@xmath116 ) and the result follows . the problem of describing the connected components of @xmath1 given by the polarized rank 1 sheaves which are not of rank 1 have to cover then the analysis of simpson s moduli spaces of sheaves of higher rank . furthermore ,
when the number of irreducible components of the curve @xmath0 is bigger than two , the compactification @xmath1 can even contain moduli spaces of rational rank sheaves on reducible curves as the following example shows .
[ e : fraccionario]let @xmath117 be a compact type curve .
let us consider a polarization @xmath3 on @xmath0 such that @xmath118 for @xmath119 .
let @xmath120 be an invertible sheaf on @xmath70 and let @xmath121 be a vector bundle of rank 2 on @xmath71 .
let us consider the sheaf @xmath122 where @xmath123 is the inclusion map .
the sheaf @xmath16 is pure of dimension one and it has polarized rank 1 with respect to @xmath3 .
moreover , the ( semi)stability of @xmath16 with respect to @xmath3 is equivalent to the ( semi)stability of @xmath124 with respect to the polarization @xmath125 ( see lemma [ l : estabilidadinducida ] ) .
however , @xmath124 is a sheaf of polarized rank 3/2
. nevertheless , when the curve @xmath0 is projective and reduced , the ( semi)stability notion of a sheaf given by the slope @xmath40 is equivalent to the ( semi)stability notion given by seshadri @xcite ( it is enough to consider as weights the rational numbers @xmath126 ) .
therefore the connected component @xmath11 of @xmath1 coincides with seshadri s compactification .
thus , in this case @xmath11 is a projective scheme that contains the simpson jacobian of @xmath0 so that it can be considered as a compactification of @xmath10 . on the other hand ,
if @xmath0 is a stable curve and @xmath127 denotes the compactification of the ( generalized ) jacobian constructed by caporaso in @xcite , by considering as polarization the canonical sheaf of @xmath0 and by using essentially the pandharipande s results of @xcite , one easily proves that there exists a bijective morphism @xmath128 from caporaso s compactification to the connected component @xmath11 .
finally , the relation between oda and seshadri s compactifications @xmath129 and the connected component @xmath11 , when @xmath0 is a nodal curve , can be found in @xcite .
let @xmath0 be a projective reduced and connected curve over @xmath12 .
let @xmath78 denote the irreducible components of @xmath0 and @xmath79 the intersection points of @xmath78 .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 .
henceforth we shall use the following notation .
* notation : * if @xmath16 is a pure dimension one sheaf on @xmath0 , for every proper subcurve @xmath130 of @xmath0 , we will denote by @xmath131 the restriction of @xmath16 to @xmath130 modulo torsion , that is , @xmath132 , @xmath133 will be the surjective morphism @xmath134 and @xmath135 .
we shall denote by @xmath136 the degree of the induced polarization @xmath137 on @xmath130 .
if @xmath138 then we shall write @xmath139
. the complementary subcurve of @xmath130 in @xmath0 , that is , the closure of @xmath140 , will be denoted by @xmath141 .
if @xmath142 denotes the arithmetic genus of @xmath0 , that is , the dimension of @xmath143 , for any pure dimension one sheaf @xmath16 on @xmath0 of polarized rank 1 and degree @xmath2 with respect to @xmath3 , let @xmath144 , @xmath145 , be the residue class of @xmath146 modulo @xmath30 so that @xmath147 for every proper subcurve @xmath130 of @xmath0 , we shall write @xmath148 if @xmath149 is a real number , we use [ @xmath149 ] to denote the greatest integer less than or equal to @xmath149 .
we collect here some general properties we will use later . the following lemma , due to seshadri , describes the stalk of a pure dimension one sheaf on @xmath0 at the intersection points @xmath151 that are ordinary double points .
[ l : localizacion ] let @xmath16 be a pure dimension one sheaf on @xmath0 . if @xmath151 is an ordinary double point lying in two irreducible components @xmath152 and @xmath153 , then @xmath154 where @xmath155 are the integer numbers determined by : @xmath156
see @xcite , huitime partie , prop .
[ l : estabilidadinducida ] let @xmath16 be a pure dimension one sheaf on @xmath0 supported on a subcurve @xmath130 of @xmath0
. then @xmath16 is stable ( resp .
semistable ) with respect to @xmath137 if and only if @xmath16 is stable ( resp .
semistable ) with respect to @xmath3 .
it follows from the equality @xmath157 where @xmath158 is the inclusion map .
[ l : buenossubhaces ] a torsion free rank 1 sheaf @xmath16 on @xmath0 is stable ( resp .
semistable ) if and only if @xmath159 ( resp .
@xmath160 ) for every proper subcurve @xmath130 of @xmath0 . given a subsheaf @xmath19 of @xmath16 such that @xmath161 , let us consider the complementary subcurve @xmath141 of @xmath130 in @xmath0 .
since @xmath162 is torsion free , we have @xmath163 with @xmath164 so that @xmath165 and the result follows .
[ l : desigualdadesgenerales ] let @xmath120 be an invertible sheaf on @xmath0 of degree @xmath2 . then @xmath120 is ( semi)stable with respect to @xmath3 if and only if for every proper connected subcurve @xmath130 of @xmath0 the following inequalities hold : @xmath166 where @xmath167 is the intersection multiplicity of @xmath130 and @xmath141 .
let us write @xmath168 .
if @xmath120 is ( semi)stable with respect to @xmath3 and @xmath130 is a proper subcurve of @xmath0 , the condition @xmath169 is @xmath170 which is equivalent to @xmath171 considering the subsheaf @xmath172 of @xmath120 , yields @xmath173 since @xmath174 and @xmath175 , we have @xmath176 , @xmath177 and @xmath178
. then and give the desired inequalities .
conversely , if @xmath130 is a connected subcurve of @xmath0 , by the left - hand side inequality of the statement , we have @xmath169 .
otherwise , this holds for every connected component of @xmath130 and it is easy to deduce it for @xmath130 .
the result follows then from the former lemma .
we collect here some results relating the picard groups of certain projective reduced curves .
they are consequences of the following proposition due to grothendieck ( @xcite , prop .
21.8.5 ) .
[ p : grothendieck ] let @xmath0 and @xmath17 be projective and reduced curves over @xmath12 .
let @xmath179 be a finite and birational of reducible curves we mean a morphism which is an isomorphism outside a discrete set of points of @xmath0 ] morphism .
let @xmath181 be the open subset of @xmath0 such that @xmath182 is an isomorphism and let @xmath183 .
let us denote @xmath184 .
then , there is an exact sequence @xmath185 if the canonical morphism @xmath186 is bijective , then the kernel of @xmath187 is isomorphic to @xmath188 .
[ c : grupoaditivo ] let @xmath0 and @xmath17 be two projective reduced and connected curves over @xmath12 .
let @xmath189 be a birational morphism which is an isomorphism outside @xmath190 .
if @xmath191 is one point @xmath192 and @xmath193 , then the sequence @xmath194 is exact .
[ c : odaseshadri ] let @xmath195 be a projective reduced and connected curve over @xmath12 .
suppose that the intersection points @xmath196 of its irreducible components are ordinary double points .
let @xmath197 be the partial normalization of @xmath0 at the nodes @xmath198 .
then , there is an exact sequence @xmath199 where @xmath200 .
in this section we describe the structure of the simpson jacobian @xmath10 and of the connected component @xmath11 of its compactification when @xmath0 is a tree - like curve .
we also give a recurrent algorithm to determine the @xmath14-equivalence class of a semistable sheaf on @xmath0 which we will use to analyze these simpson moduli spaces for reduced fibers of an elliptic fibration .
an ordinary double point @xmath15 of a projective curve @xmath0 is a disconnecting point if @xmath201 has two connected components .
[ d : tree]a tree - like curve is a projective , reduced and connected curve @xmath202 over @xmath12 such that the intersection points , @xmath79 , of its irreducible components are disconnecting ordinary double points .
note that the singularities lying only at one irreducible component of a tree - like curve can be arbitrary singularities . in this section
, we will assume that @xmath0 is a tree - like curve of arithmetic genus @xmath203 .
the number of intersection points of the irreducible components of @xmath0 is @xmath204 and then , by corollary [ c : odaseshadri ] , one has that @xmath205 when the irreducible components of @xmath0 are smooth , that is , @xmath0 is a curve of compact type , teixidor ( @xcite , lem.1 ) proves the following lemma .
her proof is also valid for any tree - like curve .
[ l : orden ] let @xmath206 a tree - like curve .
then , the following statements hold : 1 .
it is possible to order the irreducible components of @xmath0 , so that for every @xmath207 the subcurve @xmath208 is connected .
2 . for @xmath207
there are irreducible components , say @xmath209 , with all subindices smaller than @xmath210 , such that the subcurve @xmath211 is connected and intersects its complement @xmath212 in @xmath0 in just one point @xmath151 .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 and let us suppose from now on that an ordering of the irreducible components of @xmath0 as in the lemma has been fixed .
fixing the degree @xmath2 and using the notations of , this order allows us to define inductively integer numbers @xmath213 as follows : @xmath214 + 1-d_{i_1}^x-\hdots - d_{i_k}^x,\ \text { for } i=1,\hdots , n-1 \\ & d_n^x = d - d_1^x-\hdots - d_{n-1}^x.\notag\end{aligned}\ ] ] we are now going to modify the above numbers to obtain new numbers @xmath215 associated with @xmath0 .
this is accomplished by a recurrent algorithm . in order to describe it we start by saying that a connected subcurve @xmath216 , @xmath217 , of @xmath0 ordered according to lemma [ l : orden ] is _ final _ either when the numbers @xmath218 are not integers for @xmath219 or @xmath130 is irreducible .
if @xmath130 is a final curve , we define @xmath220 as follows : \1 . if @xmath221 , @xmath222 for @xmath223 .
\2 . if @xmath224 , @xmath225-\chi(\mathcal{o}_{c_{j_1}})$ ] .
[ algoritmo ] if the curve @xmath0 is final , @xmath226 for all @xmath210 .
otherwise , let @xmath210 be the first index for which @xmath227 .
we consider the two connected components , @xmath228 and @xmath229 , of @xmath230 and we reorder them according to lemma [ l : orden ] .
this induces a new ordering @xmath231 ( resp .
@xmath232 ) of the points @xmath233 ( @xmath234 ) in @xmath235 ( resp .
@xmath236 ) . then
, \a ) if @xmath235 ( resp .
@xmath236 ) is a final curve , the process finishes for @xmath235 ( resp .
@xmath236 ) .
\b ) if @xmath235 is not final , we take the first index @xmath54 of @xmath235 for which @xmath237 , consider the connected components , @xmath238 and @xmath239 , of @xmath240 and reorder them according to lemma [ l : orden ] .
if @xmath238 and @xmath239 are final , the process finishes for @xmath235 . otherwise , we iterate the above argument for those components that are not final and so on .
the process finishes for @xmath235 when all subcurves that we find are final .
\c ) if @xmath236 is not final , we take the first index @xmath241 of @xmath236 such that @xmath242 , consider the connected components , @xmath243 and @xmath244 , of @xmath245 and reorder them according to lemma [ l : orden ] .
if @xmath243 and @xmath244 are final , the process finishes for @xmath236 . otherwise , we repeat the above argument for those components that are not final and so on .
the process finishes for @xmath236 when all subcurves that we obtain are final .
the algorithm for @xmath0 finishes when it finishes for both @xmath235 and @xmath236 .
we can now state the theorem that determines the structure of the simpson jacobian of @xmath0 and of the schemes @xmath246 and @xmath11 .
[ t : treelike ] let @xmath202 , @xmath7 , be a tree - like curve .
\a ) if @xmath247 is not an integer for every @xmath207 , then @xmath248 @xmath249 where @xmath213 are the above integers .
\b ) if @xmath247 is an integer for some @xmath207 , then @xmath250 @xmath251 where @xmath215 are the integers constructed with the above algorithm .
we postpone the proof of this theorem to subsection [ ss : demostracion ] and we now give three examples to illustrate how algorithm [ algoritmo ] works .
[ e : ejemplo1]let @xmath202 , @xmath7 , be a tree - like curve with a polarization @xmath3 whose degree @xmath30 is a prime number .
suppose that the irreducible components of @xmath0 are ordered according to lemma [ l : orden ] .
then , since @xmath252 is not divisible by @xmath30 , @xmath253 is an integer if and only if @xmath254 .
therefore , by theorem [ t : treelike ] , we have that if @xmath255 , @xmath256 and @xmath257 whereas for @xmath254 , @xmath10 and @xmath246 are empty and @xmath258 moreover , in this case the number @xmath215 is given by @xmath259 actually , if @xmath120 is a strictly semistable line bundle of degree @xmath2 on @xmath0 , since @xmath260 is integer for all @xmath261 , the first index @xmath210 such that @xmath247 is integer is @xmath90 and the connected components of @xmath262 are @xmath263 and @xmath264 .
since @xmath70 is irreducible , it is a final subcurve so that @xmath265 . since @xmath266 is integer for @xmath267 ,
@xmath236 is not a final curve so we have to apply the procedure to the curve @xmath236 . here , the first index @xmath241 such that @xmath266 is integer is @xmath268 and the connected components of @xmath269 are @xmath270 and @xmath271 .
then @xmath272 .
since @xmath273 is integer for @xmath274 , we apply the procedure to the curve @xmath244 , and so on .
the iteration of this procedure will only finish when we obtain a sheaf supported on @xmath275 that belongs to @xmath276 .
thus , @xmath277 for @xmath278 .
we are now going to recover examples 1 and 2 of @xcite .
there , the irreducible components @xmath84 are taken to be smooth and @xmath279 .
then , the residue class of @xmath146 modulo @xmath30 is @xmath254 and @xmath280 .
it follows that @xmath247 is integer for @xmath278 and , arguing as in the former example , we have that @xmath281 and @xmath282 are empty and @xmath283 as asserted in @xcite .
let @xmath0 be the following tree - like curve fixing an ordering of the irreducible components of @xmath0 as in lemma [ l : orden ] , we obtain so that @xmath284 , @xmath285 , @xmath286 .
assume that the first index @xmath210 such that @xmath247 is integer is @xmath287 and let us compute the numbers @xmath215 , @xmath288 in this case .
the connected components of @xmath289 are @xmath290 and @xmath291 .
since @xmath292 is a final subcurve , we have that @xmath293 .
we now have to fix a new ordering for @xmath235 according to lemma [ l : orden ] .
we can take , for instance , @xmath294 with @xmath295 , @xmath296 and @xmath297 .
then , @xmath298 and @xmath299 . therefore ,
since @xmath300 and @xmath301 are not integers , @xmath235 is final and we conclude that @xmath302 + 1=d^x_1,\notag\\ d_3&=d_{\sigma(2)}^y=-\chi(\mathcal{o}_{y_{\sigma(2)}})+h_{y_{\sigma(2)}}t+ [ k_{y_{\sigma(2)}}]+1-d^y_{\sigma(1)}=\notag\\ & = -\chi(\mathcal{o}_y)+h_yt+k_y+1-d^x_1-d^x_2 - 1=d^x_3 - 1,\notag\\ d_2&=d_{\sigma(3)}^y = d_y - d_{\sigma(1)}^y - d_{\sigma(2)}^y = d^x_2.\notag\end{aligned}\ ] ] in order to prove that the simpson jacobian of @xmath0 is not empty only when @xmath303 for all @xmath207 , we need the next two lemmas that characterize stable invertible sheaves on @xmath0 .
these two lemmas are essentially steps 1 and 2 in @xcite .
[ l : lineatree ] let @xmath120 be a line bundle on @xmath0 of degree @xmath2 . if @xmath120 is stable , then @xmath247 is not an integer for every @xmath207 and @xmath120 is obtained by gluing invertible sheaves @xmath304 on @xmath84 of degrees @xmath213 , @xmath278 . considering the subcurves @xmath13 of @xmath0 , @xmath305 , given by lemma [ l : orden ] , from lemma [ l : desigualdadesgenerales ] , we get @xmath306 because @xmath307 .
we have @xmath308 so that @xmath309 is an integer .
then , if @xmath227 for some @xmath207 , becomes a contradiction .
thus @xmath303 for all @xmath207 and there is only one possibility for @xmath310 , namely @xmath311 + 1,\ \ \
\text { for } i=1,\hdots , n-1.\ ] ] from @xmath312 and the exact sequence @xmath313 we deduce that @xmath314 for all @xmath210 and the proof is complete .
the following lemma proves the converse of lemma [ l : lineatree ] .
[ l : reciproco ] let @xmath120 be the invertible sheaf on @xmath0 of degree @xmath168 obtained by gluing line bundles @xmath304 on @xmath84 of degrees @xmath213 , @xmath278 .
suppose that @xmath247 is not integer for every @xmath207 .
then @xmath120 is stable .
taking as weights @xmath315 and proceeding as in lemma 2 of @xcite , we obtain that for every proper connected subcurve @xmath130 of @xmath0 the following inequalities are satisfied : @xmath316 where @xmath167 .
the stability of @xmath120 follows then from lemma [ l : desigualdadesgenerales ] . by lemmas [ l : lineatree ] and [ l : reciproco ] , @xmath10 is not empty only if @xmath247 is not integer for every @xmath317 , and in this case it is equal to @xmath318 .
we prove now the remaining statements of the theorem . if @xmath120 is a strictly semistable line bundle on @xmath0 of degree @xmath2 then , @xmath310 is equal to one of the two extremal values of the inequality .
in particular , @xmath247 is an integer for some @xmath207 .
let @xmath210 be the first index such that @xmath247 is integer .
then , there are two possibilities for @xmath310 : \a ) @xmath319
\b ) @xmath320 let us construct a jordan - hlder filtration for @xmath120 in both cases .
since case a ) and case b ) are the same but with the roles of @xmath13 and @xmath321 intertwined , we give the construction in the case a ) .
we have that @xmath322 .
then , @xmath323 and @xmath324 are semistable with respect to @xmath3 and , by lemma [ l : estabilidadinducida ] , they are semistable with respect to @xmath325 and @xmath326 respectively . for simplicity , we shall write @xmath327 with @xmath328 and @xmath329 , which are again tree - like curves .
let us see when the sheaves @xmath330 and @xmath331 are stable .
we can fix an ordering for @xmath235 as in lemma [ l : orden ] , so that @xmath332 and we obtain subcurves @xmath333 of @xmath235 for @xmath334 . _ claim 1 . _
the sheaf @xmath330 is stable if and only if @xmath335 is not an integer for @xmath334 . since the residue class of @xmath336 modulo @xmath337 is @xmath338 , the numbers @xmath339 are not integers for @xmath334 .
then , from lemma [ l : reciproco ] , we have only to prove that @xmath330 is in @xmath340 , where @xmath341 are the integer numbers defined as @xmath213 but with the new ordering of @xmath235 and @xmath54 runs through the irreducible components of @xmath235 .
this is equivalent to proving that @xmath342 + 1\ \ \
\text { for } r=\sigma(i_0),\hdots,\sigma(i_{k-1}).\ ] ] actually , since @xmath120 is semistable and @xmath333 is a proper subcurve of @xmath0 , by lemma [ l : desigualdadesgenerales ] , we obtain @xmath343 where @xmath344 is the number of intersection points of @xmath333 and its complement in @xmath0 .
we have that @xmath345 and @xmath346 is not equal to the extremal values of because @xmath347 .
moreover , if it were @xmath342 + 2,\ ] ] since @xmath348 , @xmath349 and @xmath350 , @xmath351 being the complement of @xmath333 in @xmath235 , then @xmath352\ ] ] which contradicts the semistability of @xmath120 .
thus , @xmath342 + 1\ ] ] and the proof of the claim 1 is complete . on the other hand ,
the irreducible components of @xmath238 are ordered according the instructions in lemma [ l : orden ] and the subcurves @xmath353 , where @xmath241 runs through the irreducible components of @xmath238 and @xmath354 , are equal to either @xmath62 or @xmath355 .
_ claim 2 . _
the sheaf @xmath331 is stable if and only if @xmath356 is not an integer for every @xmath357 . since the residue class of @xmath358 modulo @xmath359 is @xmath360 and @xmath361 , by the hypothesis , the numbers @xmath362 are not integers for @xmath363 and , by the choice of @xmath210
, they are nt for @xmath364 either .
then , by lemma [ l : reciproco ] , it is enough to prove that @xmath365 + 1\ \ \
\text { for } s\leq n-1.\ ] ] since @xmath120 is semistable , we have that @xmath366 + 1.\ ] ] moreover , if @xmath367 then , @xmath368 and if @xmath369 then , @xmath370 . we obtain the desired result in both cases and the proof of the claim 2 is complete .
we return now to the proof of the theorem .
if @xmath335 and @xmath356 are not integers for @xmath334 and @xmath357 ( i.e. @xmath235 and @xmath238 are final curves ) , then @xmath371 is a jordan - hlder filtration for @xmath120 and the @xmath14-equivalence class of @xmath120 belongs to @xmath372 . on the other hand , if @xmath335 is integer for some @xmath334 , the sheaf @xmath330 is strictly semistable and we have to repeat the above procedure with @xmath330 in the place of @xmath120 and the curve @xmath235 in the place of @xmath0 .
similarly , if @xmath356 is integer for some @xmath357 , the sheaf @xmath331 is strictly semistable . then , we have to repeat the above procedure for @xmath331 . by iterating this procedure
, we get a jordan - hlder filtration for @xmath330 : @xmath373 and another for @xmath331 : @xmath374 therefore , a filtration for @xmath120 is given by @xmath375 thus , the @xmath14-equivalence class of @xmath120 belongs to @xmath376 , where @xmath215 are the integer numbers constructed with the algorithm .
finally , let us consider a pure dimension one sheaf @xmath16 on @xmath0 of rank 1 and degree @xmath2 which is not a line bundle .
when @xmath16 is locally free at the intersection points @xmath151 for all @xmath305 , calculations and results are analogous to the former ones . if @xmath16 is not locally free at @xmath151 for some @xmath305 , then there is a natural morphism @xmath377 where @xmath235 , @xmath238 are the connected components of @xmath378 , that is clearly an isomorphism outside @xmath151 . but
this is an isomorphism at @xmath151 as well because by lemma [
l : localizacion ] , @xmath379 and this is precisely the stalk of @xmath380 at @xmath151 .
we conclude that @xmath16 is not stable and if it is semistable , then @xmath381 and @xmath382 are integers , @xmath383 and @xmath384 are given by @xmath385 and @xmath386 and @xmath387 are semistable as well .
then , the construction of a jordan - hlder filtration for @xmath16 can be done as above and thus the @xmath14-equivalence class of @xmath16 belongs to @xmath388 .
we give here an example that shows that for reducible curves the tensor product of two semistable torsion free sheaves of rank 1 is not in general a semistable sheaf , even when the considered sheaves are line bundles .
[ e : prodtens ] let @xmath83 be a tree - like curve with @xmath86 .
suppose that @xmath84 are rational curves and fix on @xmath0 a polarization @xmath3 such that @xmath389 and @xmath118 for @xmath85 .
let @xmath390 and @xmath391 be two smooth points of @xmath70 . from lemma
[ l : desigualdadesgenerales ] , it is easy to prove that the sheaves @xmath392 and @xmath393 are semistable with respect to @xmath3 .
the same lemma proves that @xmath394 is not a semistable sheaf .
let @xmath395 be a projective smooth curve over @xmath12 and let @xmath396 be an elliptic fibration . by this we mean a proper flat morphism of schemes whose fibers are gorenstein curves of arithmetic genus 1 . by a kodaira s result ( thm .
6.2 @xcite ) the singular reduced fibers of @xmath27 can be classified as follows : @xmath397 a rational curve with one node .
@xmath398 , where @xmath70 and @xmath71 are rational smooth curves with @xmath399 .
@xmath400 , @xmath401 , where @xmath84 , @xmath402 , are rational smooth curves and @xmath403 .
@xmath404 a rational curve with one cusp .
@xmath405 where @xmath70 and @xmath71 are rational smooth curves with @xmath406 .
@xmath407 , where @xmath408 are rational smooth curves and @xmath409 . in this section
we give the description of the simpson jacobian and of the connected component @xmath11 of its compactification @xmath1 for all reduced singular fibers of an elliptic fibration . since when the fiber @xmath0 is an irreducible curve ( i.e. a rational curve with one node or one cusp ) every pure dimension one rank 1 sheaf on @xmath0 is stable , we mean the fibers of types @xmath4 , @xmath5 and @xmath6 for @xmath7 .
lemmas [ l : morfismo ] and [ l : gorenstein ] can be found in @xcite for integral curves and the proofs given are valid for finite morphisms of reducible curves .
[ l : morfismo ] let @xmath189 be a finite birational morphism of projective reduced and connected curves over @xmath12 .
let @xmath410 and @xmath411 be two pure dimension one rank 1 sheaves on @xmath17 and let @xmath412 be a morphism .
then there is a unique morphism @xmath413 such that @xmath414 .
[ l : gorenstein ] let @xmath0 be a projective reduced and connected curve over @xmath12 and @xmath15 a singular point of @xmath0 .
let @xmath415 denote the maximal ideal of @xmath15 .
set @xmath416 and @xmath417 .
let @xmath418 denote the natural map .
then the following assertions hold : 1 . @xmath419 .
2 . the curve @xmath0 is gorenstein at @xmath15 if and only if @xmath418 is a finite birational morphism and @xmath420 .
3 . if @xmath0 is gorenstein at @xmath15 and @xmath189 is a birational morphism nontrivial at @xmath15 , then @xmath9 factors trough @xmath421 .
moreover , we can adapt the proof of lemma 3.8 in @xcite to show the following : [ l : nodelinea ] let @xmath0 be a projective reduced and connected curve over @xmath12 and @xmath15 a singular point at which @xmath0 is gorenstein .
let @xmath415 denote the maximal ideal of @xmath15 .
set @xmath416 and @xmath417 .
let @xmath418 denote the natural map .
let @xmath16 be a pure dimension one rank 1 sheaf on @xmath0 .
then , @xmath16 is not locally free at @xmath15 if and only if there is a pure dimension one rank 1 sheaf @xmath19 on @xmath422 such that @xmath423 .
if @xmath19 exists , then it is unique . if @xmath19 exits , it is unique by lemma [ l : morfismo ] .
the sheaf @xmath19 exits if and only if @xmath16 is a @xmath424-module , so if and only if @xmath425 contains @xmath424 . to reduce the notation , we set @xmath426 , @xmath427 , @xmath428 and @xmath429
. hence @xmath19 does not exist if @xmath16 is invertible at @xmath15 because then @xmath430 , whereas the cokernel of @xmath431 has , by ( 2 ) of lemma [
l : gorenstein ] , length 1 .
suppose now that @xmath16 is not invertible .
we have to prove that @xmath432 .
set @xmath433 . since the curve @xmath0 is gorenstein at @xmath15 , then @xmath434 ( see , for example , @xcite ) .
let @xmath435 be the total normalization of @xmath0 , that is , if @xmath436 , then @xmath437 where @xmath438 is the normalization of the integral curve @xmath84 .
let us denote @xmath439 and @xmath440 .
since the @xmath441-module @xmath442 is free of rank 1 , there is an element @xmath443 such that @xmath444 .
then we have @xmath445 . by applying @xmath446
, we get @xmath447 since @xmath16 is not invertible , @xmath448 .
hence @xmath449 .
let @xmath450 be the jacobson radical of @xmath441 which is invertible as @xmath441-module .
set @xmath451 .
the natural morphism @xmath452 induces an inclusion @xmath453 so yields @xmath454 .
therefore , @xmath455 note that @xmath456 .
indeed , otherwise @xmath457 because @xmath458 is the largest @xmath441-submodule of @xmath459 .
however , if @xmath460 , there is a morphism @xmath461 and then @xmath462 hence , by nakayama s lemma , @xmath463 which is not true as lemma [ l : knonulo ] , that we will see later , proves .
the natural morphism @xmath464 gives an inclusion @xmath465 since @xmath466 , it is possible to consider @xmath467
. then @xmath468 , but @xmath469 . set @xmath470 . since @xmath0 is gorenstein , @xmath471 and we have @xmath472 and @xmath473 .
hence @xmath474 . therefore @xmath475 . since , by ( 1 ) of lemma [ l : gorenstein ] , it is @xmath476 , we have @xmath477 hence @xmath478 implies @xmath432 and the proof is complete .
[ l : knonulo ] using the above notations , one has @xmath479 .
since @xmath437 , we have @xmath480 with @xmath481 .
then @xmath482 and it is enough to prove that @xmath483 . using the local duality , one has @xmath484 since @xmath485 has depth one , @xmath486 and then @xmath487 .
the exact sequence @xmath488 yields @xmath489 because @xmath490 for @xmath491 . since @xmath52 is a torsion sheaf ,
we conclude that @xmath492 and the result follows .
[ l : estporimdi ] let @xmath189 be a finite birational and surjective morphism of projective reduced and connected curves with @xmath493 .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 such that @xmath494 has degree @xmath30 .
let @xmath19 be a pure dimension one sheaf on @xmath17 of rank 1 and degree @xmath2 with respect to @xmath495 .
then the following statements hold : 1 .
the sheaf @xmath496 is pure of dimension one of rank 1 and degree @xmath497 with respect to @xmath3 .
2 . @xmath19 is ( semi)stable with respect to @xmath495 if and only if @xmath496 es ( semi)stable with respect to @xmath3 .
the sheaf @xmath496 is pure of dimension one because it is the direct image of a torsion free sheaf by a finite morphism .
moreover , for any sheaf @xmath498 on @xmath17 , we have @xmath499 , so that @xmath500 then @xmath501 and the result is now straightforward . here
@xmath0 will denote a fiber of type @xmath4 of an elliptic fibration , that is , @xmath83 with @xmath70 and @xmath71 two rational smooth curves and @xmath406 ( figure 1 ) .
figure 1 .
if @xmath502 is the blow - up of @xmath0 at the point @xmath503 , then @xmath17 is a tree - like curve with @xmath86 and there is a finite birational morphism @xmath189 which is an isomorphism outside @xmath503 such that @xmath504 . since it is possible to write the completions of the local rings as @xmath505/x(x - y^2)\quad \text { and } \quad \widehat{{\mathcal{o}}}_{x',p}=\kappa[[y,\lambda]]/\lambda(\lambda - y)\ ] ] with @xmath506 , we have that @xmath507 .
hence , by corollary [ c : grupoaditivo ] and taking into account that @xmath17 is a tree - like curve , there is an exact sequence @xmath508 where the last morphism is given by @xmath509 .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 .
fixing the degree @xmath2 and using the notations of ( in this case @xmath510 ) , the proposition describing the structure of the simpson jacobian of @xmath0 and of the scheme @xmath511 of @xmath14-equivalence classes of semistable line bundles is the following : [ l : lineaiii ] let @xmath83 be a curve of type @xmath4 .
\a ) if @xmath512 , then there is an exact sequence @xmath513 @xmath514 \b ) if @xmath515 , then there is an exact sequence @xmath516}(c_i)\times { \operatorname{pic}}^{h_{c_j}t+[k_{c_j}]+1}(c_j)\to 0\ , .\ ] ] in this case , @xmath517 that is , there are not strictly semistable line bundles . since @xmath518 ,
if @xmath120 is a line bundle on @xmath0 of degree @xmath2 , one has the exact sequence @xmath519 where @xmath52 is a torsion sheaf with support at @xmath503 and @xmath520 . by lemma [ l : desigualdadesgenerales ] ,
the sheaf @xmath120 is ( semi)stable if and only if @xmath521 since @xmath522 and @xmath523 , we get the following : \a ) if @xmath512 , then @xmath120 is stable if and only if @xmath524 for @xmath85 .
\b ) if @xmath515 , then @xmath120 is stable if and only if @xmath525 $ ] and @xmath526 + 1 $ ] or the same but with the roles of @xmath70 and @xmath71 intertwined . from
, we obtain then the exact sequences in a ) and b ) .
suppose now that the sheaf @xmath120 is strictly semistable . in this case , @xmath527 is equal to one of the two extremal values of .
assume that @xmath528 ( the other case is the same but with the roles of @xmath70 and @xmath71 intertwined ) .
then @xmath512 , @xmath529 and @xmath530 and @xmath531 are stable sheaves with respect to @xmath114 and @xmath113 respectively .
thus , @xmath532 is a jordan - hlder filtration for @xmath120 and then its @xmath14-equivalence class belongs to @xmath533 .
hence , there is only one @xmath14-equivalence class of strictly semistable line bundles so that @xmath534 . as before ,
let @xmath502 be the tree - like curve obtained by gluing transversally at @xmath15 the irreducible components of the curve @xmath0 and @xmath189 the natural morphism .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 such that @xmath494 is of degree @xmath30 as well .
the following proposition proves that there is an isomorphism between the set of boundary points of the connected component @xmath11 and the component @xmath22 when we consider the polarization @xmath3 on @xmath0 and the polarization @xmath495 on @xmath17 . since @xmath17 is a tree - like curve , the structure of the border @xmath535 is determined by theorem [ t : treelike ] .
[ p : bordeiii ] let @xmath83 be a curve of type @xmath4 with @xmath406 . if @xmath502 with @xmath86 and we consider on @xmath0 ( resp .
@xmath17 a polarization @xmath3 ( resp .
@xmath495 ) as above , then there are isomorphisms @xmath536
@xmath537 let @xmath415 be the maximal ideal of @xmath503 in @xmath0 and let us denote @xmath538 and @xmath539 .
taking into account that @xmath0 is gorenstein and that the natural morphism @xmath189 is finite , birational and non trivial at @xmath503 , by ( 2 ) and ( 3 ) of lemma
[ l : gorenstein ] , we get a morphism @xmath540 .
since @xmath17 and @xmath422 are both of arithmetic genus zero , this morphism is an isomorphism .
let @xmath16 be a pure dimension one sheaf on @xmath0 of rank 1 and degree @xmath2 which is not a line bundle , that is , it is not locally free at the point @xmath503 . by lemma [ l : nodelinea ] , there is a unique pure dimension one rank 1 sheaf @xmath19 on @xmath17 such that @xmath541 .
since @xmath9 is finite , birational and surjective and @xmath542 , by lemma [ l : estporimdi ] , @xmath19 has degree @xmath543 and it is ( semi)stable with respect to @xmath495 if and only if @xmath16 is ( semi)stable with respect to @xmath3 .
therefore , the direct image @xmath544 produces the desired isomorphisms . in this part
, @xmath0 will denote a fiber of an elliptic fibration of type @xmath5 , that is , @xmath117 with @xmath84 rational smooth curves and @xmath545 ( figure 2 ) .
figure 2 .
let @xmath546 be the curve obtained by gluing the irreducible components of @xmath0 at the point @xmath15 in such a way that this curve @xmath17 can not be embedded , even locally , in a smooth surface ( see figure 3 ) .
figure 3 .
there exists a finite birational morphism @xmath189 which is an isomorphism outside @xmath15 and @xmath191 is the point @xmath15 of @xmath17 . since
the completions of the local rings of @xmath0 and @xmath17 at @xmath15 can be written as @xmath547/xy(y - x)\quad \text { and } \quad \widehat{\mathcal{o}}_{x',p}=\kappa[[\bar{x},\bar{y},\bar{z}]]/(\bar{x}\bar{y},\bar{x}\bar{z},\bar{y}\bar{z})\ ] ] with @xmath548 , @xmath549 by @xmath9 , it is easy to prove that @xmath550
. then , from corollary [ c : grupoaditivo ] , there is an exact sequence @xmath551 moreover , the picard group of @xmath17 is isomorphic to @xmath552 . indeed ,
if @xmath553 is the total normalization of @xmath17 at @xmath15 and @xmath554 is the projection map , by proposition [ p : grothendieck ] , we have the following exact sequence : @xmath555 considering the exact sequence @xmath556 where the morphism @xmath149 is defined by : @xmath557 we get @xmath558 since the morphism @xmath559 , which is given by @xmath560 is surjective , by , we obtain that @xmath561 from , we conclude that the picard group of the curve @xmath0 is given by the following exact sequence : @xmath562 we start now with the ( semi)stability analysis .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 .
as in , if @xmath2 is the degree of the jacobian we are considering , @xmath144 will be the residue class of @xmath543 modulo @xmath30 . when the numbers @xmath563 are not integers , we write @xmath564+a_i$ ] with @xmath565 . since @xmath566 , then @xmath567 is equal to 1 or 2 .
thus the proposition describing the schemes @xmath10 and @xmath511 is the following : [ p : lineaiv ] let @xmath117 be a curve of type @xmath5 .
we have the following three cases : \a ) if @xmath568 for @xmath119 , then there is an exact sequence @xmath569 @xmath570 where @xmath571 for all @xmath210 .
\b ) since it is not possible to have @xmath572 only for one index @xmath210 because the sum of these three numbers is an integer , the following case to considering is @xmath573 integer only for one @xmath210 .
set @xmath512 and @xmath572 for @xmath574 .
then , there is an exact sequence @xmath575 where @xmath576}(c_j)\times { \operatorname{pic}}^{h_{c_l}t+[k_{c_l}]+1}(c_l)$ ] . in this case
, we have @xmath570 where @xmath577 and @xmath111 , @xmath578 are the integer numbers obtained by applying algorithm [ algoritmo ] to the tree - like curve @xmath579 for a sheaf of degree @xmath580 .
\c ) suppose that @xmath572 for @xmath119 .
if @xmath581 , there is an exact sequence @xmath582 with @xmath583 + 1}(c_i)\times { \operatorname{pic}}^{h_{c_j}t+[k_{c_j}]}(c_j)\times { \operatorname{pic}}^{h_{c_l}t+[k_{c_l}]}(c_l)$ ] .
if @xmath584 , there is an exact sequence @xmath585 with @xmath583 + 1}(c_i)\times { \operatorname{pic}}^{h_{c_j}t+[k_{c_j}]+1}(c_j)\times { \operatorname{pic}}^{h_{c_l}t+[k_{c_l}]}(c_l)$ ] . in both cases , @xmath586 is empty , that is , every semistable line bundle is stable .
since @xmath587 , for every line bundle @xmath120 of degree @xmath2 on @xmath0 , one has an exact sequence @xmath588 where @xmath52 is a torsion sheaf supported at the point @xmath15 and @xmath589 .
the only connected subcurves of @xmath0 are @xmath84 , @xmath119 , and their complements @xmath590 and @xmath591 so that , by lemma [ l : desigualdadesgenerales ] , the sheaf @xmath120 is ( semi)stable if and only if @xmath592 taking into account that @xmath593 and @xmath594 , in case a ) @xmath120 is stable if and only if @xmath524 for @xmath119 .
since the following case is @xmath572 for two indices , reordering the irreducible components of @xmath0 , we can assume that we are in case b ) .
in this case , @xmath120 is stable if and only if @xmath595 , @xmath526 $ ] and @xmath596 + 1 $ ] or the same with the roles of @xmath71 and @xmath597 intertwined . finally , in case c ) , @xmath120 is stable if and only if @xmath598+\epsilon_i$ ] where @xmath599 .
therefore , reordering the irreducible components of @xmath0 if it were necessary , the possibilities are : \1 . if @xmath581 , then @xmath600 and @xmath601 .
\2 . if @xmath584 , then @xmath602 and @xmath603 .
this together with proves the exact sequences of the statement .
when @xmath120 is a strictly semistable line bundle on @xmath0 of degree @xmath2 , @xmath604 is equal to one of the two extremal values of , and then @xmath568 , for some @xmath119 .
reordering the irreducible components of @xmath0 , we can assume that @xmath512 and @xmath528 .
hence , @xmath605 so that , by lemma [ l : estabilidadinducida ] , @xmath530 and @xmath531 are semistable with respect to @xmath114 and @xmath606 respectively
. since @xmath70 is an integral curve @xmath607 is stable . on the other hand ,
since @xmath608 is a tree - like curve , by theorem [ t : treelike ] , we have that @xmath609\in \prod_{i=2}^3{\operatorname{pic}}^{d_i}(c_i)$ ] where @xmath111 and @xmath578 are the integer numbers obtained by applying algorithm [ algoritmo ] to @xmath579 for the sheaf @xmath531 which has degree @xmath610 .
bearing in mind that if @xmath568 for @xmath574 , the only final subcurves of @xmath579 are its irreducible components , we conclude that the @xmath14-equivalence class of @xmath120 belongs to @xmath611 with @xmath215 the integers of the statement .
we study now the set of boundary points of the connected component @xmath11 .
[ p : bordeiv ] let @xmath117 be a curve of type @xmath5 with @xmath612 and let @xmath546 be the curve of the figure 3 .
let @xmath189 denote the natural morphism .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 such that @xmath613 is also of degree @xmath30 . then , considering on @xmath17 the polarization @xmath495 , there are isomorphisms @xmath536
@xmath537 since the curve @xmath0 is gorenstein , @xmath542 and @xmath9 is a finite birational morphism non trivial at the point @xmath15 , arguing as in the proof of proposition [ p : bordeiii ] , the morphism @xmath544 produces the desired isomorphisms .
thus the description of the connected component @xmath11 when @xmath0 is a fiber of type @xmath5 implies the analysis of simpson schemes @xmath614 and @xmath615 for the curve @xmath17 of the figure 3 .
this analysis is given in the following subsection which completes then the description for the curves of type @xmath5 .
let @xmath546 be the curve of the figure 3 .
let @xmath495 be a polarization on @xmath17 of degree @xmath30 . with the notations we come using , we have the following if @xmath546 is the curve of the figure 3 , we have the following two cases : \a ) if @xmath568 for some @xmath119 , then @xmath616 @xmath617 where @xmath571 if @xmath568 and @xmath618 , @xmath619 , @xmath620 are the integer numbers obtained by applying algorithm [ algoritmo ] to the tree - like curve @xmath621 for a sheaf of degree @xmath622 .
\b ) suppose that @xmath572 for @xmath119 .
\1 . if @xmath581 , then @xmath623}(c_i)\ , .\ ] ] 2 .
if @xmath584 , then @xmath624 in these two cases , @xmath625 we have seen that the picard group of this curve @xmath17 is isomorphic to the direct product of the picard groups of its irreducible components . moreover , if @xmath120 is a line bundle on @xmath17 of degree @xmath2 , by lemma [ l : desigualdadesgenerales ] , @xmath120 is ( semi)stable if and only if @xmath626 then , the result follows arguing as in the proof of proposition [ p : lineaiv ] . in order to determine boundary points of @xmath615 , we need the following lemma that describes the stalk of a pure dimension one sheaf on the curve @xmath17 at its only singular point @xmath15 . [
l : localizacion2 ] if @xmath16 is a pure dimension one sheaf on the curve @xmath17 , then @xmath627 where @xmath628 and @xmath629 , @xmath630 and @xmath631 are integer numbers determined by the following equalities : @xmath632 since @xmath16 is a pure dimension one sheaf , the @xmath633-module @xmath634 has depth 1 . for @xmath119 ,
let @xmath635 be a local parameter of @xmath84 at @xmath15 .
if @xmath415 denotes the maximal ideal of @xmath15 in @xmath17 and @xmath636 is the maximal ideal of @xmath15 in @xmath84 , we have @xmath637 so that @xmath638 for @xmath491 . since @xmath639 is a torsion free @xmath640-module , it is free : @xmath641 the map @xmath642 is injective because @xmath643 has depth 1 . consider the inclusion map @xmath644 from , one has that @xmath645 and then @xmath646 hence , there is a map @xmath647 whose kernel is @xmath648 . for @xmath119 ,
let @xmath649 be the following submodule of @xmath643 : @xmath650 if @xmath651 , then @xmath652 .
indeed , let @xmath653 be a nonzero element of @xmath654 with @xmath655 for @xmath656 . since @xmath657 for @xmath658 , we can write @xmath659 for some @xmath660 . if @xmath661 , then @xmath662 and @xmath663 which proves the claim
. let @xmath664 be elements of @xmath665 such that @xmath666 are free and let us check that @xmath664 are free over @xmath667 .
suppose that @xmath668 with @xmath669 and @xmath670 for some @xmath210 .
let @xmath671 be the maximal power of @xmath672 dividing all @xmath673 and write @xmath674 for @xmath675 .
since @xmath638 for @xmath491 , we have that @xmath676 for @xmath119 , which implies that @xmath677 because @xmath643 has depth 1 .
recurrently , we get @xmath678 therefore , @xmath679 which is absurd because @xmath680 for some @xmath681 and @xmath682 are free .
thus @xmath665 is a free @xmath667-module .
the same argument proves that @xmath683 ( resp .
@xmath684 ) is a free @xmath685 ( resp .
@xmath686 ) module .
one has @xmath687 for @xmath491 as @xmath688 for @xmath689 and @xmath643 has depth 1 .
let @xmath690 be a vector subspace of @xmath691 which is supplementary of @xmath692 .
for @xmath491 with @xmath693 , let @xmath694 the following submodule of @xmath643 : @xmath695 arguing as above , if @xmath696 , then @xmath697 .
let @xmath698 be elements of @xmath699 such that @xmath666 are free and let us prove that @xmath664 are also free over @xmath700 .
suppose that @xmath701 where @xmath702 are not all nulls . since @xmath703 are free , @xmath704 for @xmath705 .
let @xmath671 and @xmath706 be the maximal powers of @xmath672 and @xmath707 dividing all @xmath673 and write @xmath708 , @xmath44 . since @xmath709 , we have that either @xmath710 are not all nulls or @xmath711 are not all nulls .
suppose that @xmath710 are not all nulls . by multiplying by @xmath672 the equality
, we get @xmath712 since @xmath713 for @xmath574 as well and @xmath643 has depth 1 , we have that @xmath714 recurrently , we get @xmath715 set @xmath716 . since @xmath717 , then @xmath718 so that @xmath719 is zero , but this is absurd because @xmath710 are not all nulls and @xmath720 are free . thus the @xmath721-module @xmath699 is free .
analogously , @xmath722 and @xmath723 are free modules and it is not difficult to prove that all intersections of @xmath694 and @xmath649 are zero . let @xmath724 a vector subspace of @xmath691 supplementary of @xmath725 .
let @xmath726 be elements of @xmath643 such that @xmath727 are a basis of @xmath724 .
let @xmath728 be the submodule of @xmath643 generated by @xmath729 .
let us prove that @xmath726 are free over @xmath633 .
suppose that @xmath730 where @xmath731 are not all nulls . since @xmath727 are free , @xmath704 for @xmath732 .
let @xmath671 , @xmath706 and @xmath733 be the maximal powers of @xmath672 , @xmath707 and @xmath734 dividing @xmath673 for @xmath732 and write @xmath735 , @xmath732 . since @xmath638 for @xmath491 , one of the following conditions holds : @xmath736 for some @xmath210 , @xmath737 for some @xmath210 or @xmath738 for some @xmath210 .
suppose that @xmath736 for some @xmath210 . by multiplying by @xmath672 equality and arguing as in the former case
, we get @xmath739 if @xmath740 , then @xmath741 is a nonzero element of @xmath724 . but
this is impossible because @xmath742 and @xmath743 for @xmath119 .
hence @xmath728 is a free @xmath633-module .
one has also @xmath744 for all @xmath693 . from nakayama s lemma
, we conclude then that @xmath745 which proves the result .
[ p : bordex ] if @xmath546 is the curve of the figure 3 , then the only stable pure dimension one rank 1 sheaves on @xmath17 are stable line bundles , that is , @xmath746 moreover , we have that \a ) if @xmath572 for all @xmath119 , then the set of boundary points is empty , that is , @xmath747
\b ) if @xmath568 for some @xmath119 , then @xmath748 where @xmath571 and @xmath618 @xmath658
are the integer numbers obtained by applying algorithm [ algoritmo ] to the tree - like curve @xmath621 for a sheaf of degree @xmath622 .
let @xmath16 be a pure dimension one sheaf on @xmath17 of rank 1 and degree @xmath2 with respect to the polarization @xmath495 .
if @xmath16 is not a line bundle , from lemma [ l : localizacion2 ] , we get that , reordering the irreducible components of @xmath17 if it were necessary , the sheaf @xmath16 is isomorphic either to @xmath749 or to @xmath750 .
so @xmath16 is not stable .
suppose that @xmath16 is semistable and let us determine its @xmath14-equivalence class .
if @xmath751 , the condition @xmath752 for @xmath119 implies that @xmath753 .
then @xmath568 for @xmath119 and the @xmath14-equivalence class of @xmath16 belongs to @xmath754 where @xmath215 are the integers of the statement because in this case the only final subcurves are the irreducible components . if @xmath755 , since @xmath756 , then @xmath757 so that @xmath758 and the sheaf @xmath759 is semistable of degree @xmath760 with respect to @xmath125 on the tree - like curve @xmath761 . by theorem [ t : treelike ] , we conclude that @xmath100\in \prod_{i=1}^3{\operatorname{pic}}^{d_i}(c_i)$ ] with @xmath215 the integers of the statement as well . in all this subsection @xmath0
will be a fiber of an elliptic fibration of type @xmath6 ( figure 4 ) , that is , if @xmath762 , then @xmath763 with @xmath764 , and if @xmath765 , then @xmath83 with @xmath399 . in both cases , the irreducible components of @xmath0 are rational smooth curves .
figure 4 .
since the number of intersection points of the irreducible components of @xmath0 is equal to the number of irreducible components , by corollary [ c : odaseshadri ] , there is an exact sequence @xmath766 let @xmath3 be a polarization on @xmath0 of degree @xmath30 . fixing the degree @xmath2 and using the above notations , that is , @xmath144 is the residue class of @xmath543 module @xmath30 and @xmath563 , let us write @xmath564+a_i$ ] with @xmath565 when these numbers are not integers and set us @xmath599 . then the proposition describing the structure of the simpson jacobian of @xmath0 of degree @xmath2 is the following : [ p : lineaestablein ] let @xmath767 be a curve of type @xmath6 , @xmath768 .
\a ) if @xmath568 for @xmath402 , there is an exact sequence @xmath769 \b ) if there are exactly @xmath770 indices , say @xmath771 , such that the numbers @xmath772 are not integers , then @xmath10 is not empty if and only if for any subset @xmath773 such that either @xmath774 is a connected subcurve or it is connected by adding irreducible components @xmath84 with @xmath775 the following inequalities hold : @xmath776 in this case , there is an exact sequence @xmath777+\epsilon_{i_j}}(c_{i_j})\to 0\ , .\ ] ] since @xmath778 , for every line bundle @xmath120 on @xmath0 of degree @xmath2 there is an exact sequence @xmath779 where @xmath52 is a torsion sheaf with @xmath780 . by the exact sequence ,
it is enough then to find the values of the degrees @xmath604 so that @xmath120 is stable .
note that every connected subcurve @xmath130 of @xmath0 is a tree - like curve of arithmetic genus 0 and @xmath781 .
hence , by lemma [ l : desigualdadesgenerales ] , @xmath120 is ( semi)stable if and only if @xmath782 for every connected subcurve @xmath130 of @xmath0 . since the degrees @xmath604 are integers , we have that \a ) if @xmath783 for @xmath278 , then @xmath120 is stable if and only if @xmath524 for @xmath278 , because @xmath784 if @xmath785 .
\b ) suppose that only @xmath772 , @xmath770 , are not integers . if @xmath120 is stable , by , @xmath786 for @xmath787 and @xmath788+\epsilon_{i_j}$ ] with @xmath789 for @xmath790 .
conversely , suppose that @xmath120 is a line bundle on @xmath0 obtained by gluing line bundles @xmath304 on @xmath84 of degrees @xmath786 for @xmath787 and @xmath788+\epsilon_{i_j}$ ] with @xmath789 for @xmath790 .
let @xmath130 be a connected subcurve of @xmath0 .
if @xmath130 contains no component @xmath791 , one easily check that @xmath792 holds .
otherwise , consider @xmath793 . since @xmath794 satisfies the second condition of the statement , by the hypothesis
, @xmath795 holds as well .
then @xmath120 is stable and the proof is complete . for the subscheme of strictly semistable line bundles
, we have : [ p : lineasemiin ] let @xmath767 be a curve of type @xmath6 , @xmath768 .
then , @xmath796 where @xmath210 ( resp .
@xmath797 ) runs through the irreducible components of a connected subcurve @xmath261 ( resp .
@xmath141 ) such that @xmath798 and @xmath215 ( resp .
@xmath618 ) are the integers obtained by applying algorithm [ algoritmo ] to @xmath130 ( resp .
@xmath141 ) for a sheaf of degree @xmath799 ( resp .
@xmath800 ) .
if there is not so connected subcurve @xmath130 of @xmath0 , then @xmath801 .
if @xmath120 is a strictly semistable line bundle on @xmath0 of degree @xmath2 , there is a connected subcurve @xmath130 of @xmath0 such that @xmath792 is equal to one of the two extremal values of . in particular ,
suppose that @xmath803 ( the other case is similar ) .
then @xmath804 and , by lemma [ l : estabilidadinducida ] , the sheaves @xmath805 and @xmath806 are semistable with respect to @xmath807 and @xmath137 respectively . since @xmath130 and @xmath141 are tree - like curves , we conclude thanks to theorem [ t : treelike ] . arguing as in the proof of proposition [ p : bordeiii ] , we get the following proposition which together with theorem [ t : treelike ] gives us the structure of the border of @xmath11 when @xmath0 is a curve of type @xmath6 . [
p : bordein ] let @xmath767 be a curve of type @xmath6 , @xmath768 and let @xmath808 be the tree - like curve obtained by blowing up @xmath0 at one of its singular points .
let @xmath189 denote the natural projection .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 such that @xmath809 is also of degree @xmath30 . then , considering on @xmath17 the polarization @xmath495 , there are isomorphisms @xmath810
@xmath811 it is known that the pull - black of a stable torsion free sheaf by a finite morphism of integral curves is stable .
using the above descriptions , the following example shows that this result is not longer true for reducible curves .
[ e : imageninversa ] let @xmath83 be a curve of type @xmath23 with @xmath399 .
let @xmath812 be the blow - up of @xmath0 at one of its singular points , say @xmath15 , and let @xmath189 be the natural morphism .
let @xmath3 be a polarization on @xmath0 of degree @xmath30 such that @xmath613 is also of degree @xmath30 .
let @xmath813 ( resp .
@xmath814 ) be a line bundle on @xmath70 ( resp .
@xmath71 ) of degree @xmath815 ( resp .
@xmath816 ) for some @xmath817 . let @xmath120 be a line bundle on @xmath0 such that @xmath818 for @xmath85 . with the above notations ,
since @xmath120 has degree @xmath819 , we have that the residue class of @xmath820 module @xmath30 is equal to @xmath821 , @xmath822 for @xmath85 and @xmath823 , @xmath824 , @xmath600 and @xmath825 .
since the conditions of the proposition [ p : lineaestablein ] hold for all @xmath826 , @xmath120 is stable with respect to @xmath3 .
the line bundle @xmath827 on @xmath17 has degree @xmath2 with respect to @xmath495 .
the curve @xmath17 is a tree - like curve of arithmetic genus zero , its irreducible components are ordered according to lemma [ l : orden ] and @xmath828 .
since now the residue class of @xmath829 is equal to @xmath830 and the corresponding number @xmath831 is integer , by theorem [ t : treelike ] , the sheaf @xmath827 is not stable .
in this section we want to make a special emphasis in the case of degree zero .
this case is particularly significative because , as corollary [ c : nopolarizacion ] proves , if @xmath0 is a fiber of type @xmath4 , @xmath5 or @xmath6 of an elliptic fibration , for degree zero the conditions of semistability are independent of the polarization on @xmath0 .
moreover , the results of the previous section now take a simpler and more explicit form .
suppose that the irreducible components of @xmath0 are ordered according to lemma [ l : orden ] .
since the arithmetic genus of @xmath0 is 0 and @xmath832 , using the notations of , we have that @xmath254 .
then , @xmath833 for all @xmath210 and the result follows from theorem [ t : treelike ] . 1 .
if @xmath120 is a line bundle on @xmath0 of degree 0 , then @xmath120 is stable if and only if @xmath835 for all @xmath210 .
every stable pure dimension one sheaf on @xmath0 of rank 1 and degree 0 is a line bundle .
\2 . if @xmath16 is a stable pure dimension one sheaf on @xmath0 of rank 1 and degree 0 which is not a line bundle , by propositions [ p : bordeiii ] , [ p : bordeiv ] and [ p : bordein ] , @xmath839 where @xmath19 is a stable pure dimension one sheaf of rank 1 and degree -1 on a curve @xmath17 we have determined . when @xmath0 is of type @xmath4 or @xmath6
, @xmath17 is a tree - like curve with rational components .
then this is impossible by lemma [ l : noest ] .
when @xmath0 is of type @xmath5 , @xmath17 is the curve of the figure 3 and this is also absurd by the description given in [ desx ] . 1 . if @xmath0 is of type @xmath4 , @xmath120 is strictly semistable if and only if @xmath840 and @xmath841 .
if @xmath0 is of type @xmath5 , @xmath120 is strictly semistable if and only if @xmath840 , @xmath842 and @xmath843 .
if @xmath0 is of type @xmath6 , @xmath120 is strictly semistable if and only if @xmath844 where @xmath845 or @xmath846 in such a way that when we remove the components @xmath84 for which @xmath847 there are neither two consecutive @xmath848 nor two consecutive @xmath849 .
if @xmath30 is the degree of the polarization on @xmath0 , then the residue class of @xmath850 module @xmath30 is @xmath254 and @xmath280 .
then , for every connected subcurve @xmath130 of @xmath0 , it is @xmath851 .
since the arithmetic genus of @xmath130 is 0 and @xmath781 , by lemma [ l : desigualdadesgenerales ] , @xmath120 is strictly ( semi)stable if and only if for every connected @xmath261 we have @xmath852 and @xmath792 is equal to one of the two extremal values for some @xmath130 .
the result is now straightforward . where the numbers denote the degrees of the line bundles we have to consider on each irreducible component .
the following examples in @xmath854 and @xmath855 are not possible for a strictly semistable line bundle of degree 0 : [ c : nopolarizacion]if @xmath0 is a polarized curve of type @xmath4 , @xmath5 or @xmath6 , the ( semi)stability of a pure dimension one sheaf of rank 1 and degree 0 on @xmath0 does not depend on the polarization . for line bundles , the result follows from propositions [ p : grado0 ] and [ p : propiasemi ]
. if @xmath16 is a pure dimension one sheaf on @xmath0 of rank 1 and degree 0 which is not a line bundle , using lemma [ l : buenossubhaces ] , it is semistable if and only if @xmath856 for any @xmath261 which does not depend on the polarization because @xmath857 . [ c : nofino ] let @xmath0 be a curve of type @xmath4 , @xmath5 or @xmath6 .
the moduli space of semistable pure dimension one sheaves of rank 1 and degree 0 on @xmath0 is not a fine moduli space , that is , there is no universal sheaf .
[ c : equivalentes ] let @xmath0 be a polarized curve of type @xmath4 , @xmath5 or @xmath6 .
if @xmath16 is a strictly semistable pure dimension one sheaf of rank 1 and degree 0 on @xmath0 , then its graded object is @xmath858 . | for any projective curve @xmath0 let @xmath1 be the simpson moduli space of pure dimension one rank 1 degree @xmath2 sheaves that are semistable with respect to a fixed polarization @xmath3 on @xmath0 .
when @xmath0 is a reduced curve the connected component of @xmath1 that contains semistable line bundles can be considered as the compactified jacobian of @xmath0 . in this paper
we give explicitly the structure of this compactified simpson jacobian for the following projective curves : tree - like curves and all reduced and reducible curves that can appear as kodaira singular fibers of an elliptic fibration , that is , the fibers of types @xmath4 , @xmath5 and @xmath6 with @xmath7 |
it has long been suspected that the most luminous agn are powered by accretion of matter onto massive black holes ( mbh ) .
this belief , combined with the observed evolution of the space - density of agn ( van der marel 1999 and references therein ) and the high incidence of low luminosity nuclear activity in nearby galaxies ( ho , filippenko , & sargent 1997 ) , implies that a significant fraction of luminous galaxies must host black holes of mass @xmath2m@xmath3 .
it is now clear that a large fraction of hot spheroids contain a mbh ( e.g. harms et al . 1994 ;
macchetto et al . 1997 ; van der marel et al . 1998 ; richstone et al . 1998 ) .
moreover , there is a suggestion in the data that the hole mass is proportional to the mass ( or luminosity ) of the host spheroid .
quantitatively , @xmath4 ( e.g. richstone et al .
this relation is still controversial , however , both because the sensitivity of published searches is correlated with bulge luminosity , and because there is substantial scatter in @xmath5 at fixed @xmath6 .
recently ferrarese & merrit ( 2000 ) and gebhardt et al .
( 2000 ) have shown that a tighter correlation holds between the bh mass and the velocity dispersion of the bulge .
clearly , any correlation of black hole and spheroid properties would have important implications for theories of galaxy formation in general , and bulge formation in particular . to date the majority of bh detections have taken place in luminous , early - type systems , and most of the data for spirals come either from studies of water masers , or are for nearby spirals in which a mbh can be detected through stellar dynamics . a more systematic study of spiral galaxies is clearly needed at this stage
prompted by the above motivations , we are undertaking a comprehensive survey for mbhs in spiral galaxies , both quiescent and active .
we identified a volume limited sample ( @xmath7km / s ) of 54 sb , sbb , sc , and sbc spiral galaxies from a ground - based study by axon et al . who obtained h@xmath0 and [ nii ] rotation curves at seeing - limited resolution of 1for 128 spiral galaxies from the rc3 catalogue .
the selected spiral galaxies are known to have nuclear gas disks and span wide ranges in bulge mass and concentration .
the low redshift cut - off was chosen to ensure a high spatial resolution on source in order to detect even low - mass mbhs .
the frequency of agn in our sample is typical of that found in other surveys of nearby spirals , with comparable numbers of weak nuclear radio sources and liners .
the main aim of our survey is to verify if the correlations of the bh mass with bulge mass and velocity dispersion ( richstone et al .
1998 , ferrarese & merrit 2000 , gebhardt et al . 2000 ) also hold for spiral galaxies .
recently , it has been shown that hierarchical galaxy formation models can account for these correlations ( haehnelt & kauffmann 2000 ) . if massive spheroids form by hierarchical merging , then the growth of mbhs may be also a hierarchical process .
small spheroids have undergone less mergers and are thus expected to have smaller bhs .
thus , the above correlations may hold the key to testing these ideas about galaxy evolution . furthermore
, one would like to compare the bh mass with other galaxy properties such as the shape of the nuclear light profiles ; a comparison with agn activity hosted in the galaxy would provide constraints on the efficiency of accretion .
for each galaxy , we obtain three medium resolution spectra with stis and the g750 m grating centered on h@xmath0 using the 01 or 02 slit . the resolution is thus @xmath8 or 3000 respectively .
the spectra are taken for three parallel slit positions , with the central one centered on the galaxy nucleus and the other two offset by @xmath9 .
the reason for this choice is explained in fig . 1 .
in order to remove cosmic rays and hot pixels , we obtain two exposures for each slit position with the object shifted by an integer number of detector pixels along the slit .
the use of hst guarantees a spatial resolution of about 01 which corresponds to 10pc on a galaxy at a distance of 20 mpc .
we currently have hst / stis data for 90% of the sample and have extracted rotation curves for these galaxies . generally , the rotation curves show a steep nuclear rise to a long plateau . in several cases
the nuclear rotation curve presents the characteristic unresolved s - shaped velocity structure expected for keplerian rotation around a mass concentration ( fig . 2 ) .
this dynamical signature is accompanied by an abrupt increase of the emission line brightness toward the center , indicative of the presence of a well defined morphological structure , very likely a rotating circumnuclear disk .
comparison with the stellar brightness profiles derived from the stis acquisition images yields stringent lower limits on the m / l ratios that strongly suggest the presence of a supermassive black hole . to interpret the observed rotation curves we have software in place to calculate and fit to the data the predictions of thin disk models with circular rotation ( e.g. , marconi et al .
2000 ; van der marel & van den bosch 1998 ) . as an example
, we now focus on ngc 4041 , which clearly shows the possibilities of this project .
ngc 4041 is at a distance of about 20 mpc .
the rotation curve shown in the right panel of fig .
2 shows the s - shaped signature of a rotating disk of @xmath10 pc radius .
note the small velocity amplitude of the rotation curve which is about 40 km / s peak to peak . assuming an edge - on disk with the line of nodes parallel to the slit one can infer that the minimum dynamical mass required to explain the data is @xmath11m@xmath12 .
this mass estimate takes into account the `` smearing '' of the rotation curve due to the finite slit size and instrumental point spread function .
this low mass value gives an indication of the lowest masses which can be detected with our project .
the inferred dynamical mass for ngc 4041 should be considered an upper limit to the mass of a possible mbh , since the nuclear star cluster in this galaxy may be responsible for all or part of the dynamically inferred mass . | we describe an on - going hst program aimed at determining the relationship between the nuclear black hole mass and bulge mass in spiral galaxies . we have selected a volume limited sample of 54 nearby spiral galaxies for which we already have ground based emission line rotation curves , ccd surface photometry and radio maps .
we are now obtaining hst / stis longslit observations of each of the galaxies in the sample in order to determine the nuclear h@xmath0 rotation curve at high ( @xmath1 ) spatial resolution .
we will use these data to measure the unresolved dark mass concentration at the nucleus of each object . here
we show the first results from observations of objects in the sample .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
we propose to apply some tools of multifractal analysis to the study of market crash situations ( nasdaq , dow jones industrial average ) .
we take the data of market price fluctuations on a daily basis and we process it in two ways : 1 ) the prior - to - crash market prices are taken as a discrete one dimensional signal this will be the cantordust , covered with adjacent boxes @xmath0 with equal length @xmath1 .
these boxes are intervals .
the weight @xmath2 of box @xmath0 will be the number of days in which the market prices are in box @xmath0 .
the weights @xmath2 are then normalized into probabilities @xmath3 , by dividing each @xmath2 by the total @xmath4 cardinality of the sample .
then , the concentration @xmath5 of a box @xmath0 is @xmath6 .
therefore , @xmath7 corresponds to the heaviest box(es ) , @xmath8 to the lightest one(s ) .
the segment @xmath9 $ ] is divided into adjacent intervals of equal length @xmath10 , all @xmath5 s in a @xmath10 are identified , and @xmath11 is their number .
we define @xmath12 .
this spectrum - by - definition is hereby called @xmath13 in order to distinguish it from the usual lagrangian @xmath14 given by the thermodynamic algorithm : @xmath15 , and @xmath16 , @xmath1 small , @xmath17 a parameter , @xmath18 . notice that , in order to obtain @xmath13 and/or @xmath14 , we _ process _
weights @xmath2 , @xmath4 , and even @xmath11 , which is for instancealways logarithmized and then divided by @xmath19 in order to produce @xmath13 .
2 ) in this paper , we alsouse the _ unprocessed _ data above : the _ unprocessed _ @xmath2 and the unlogarithmized @xmath11 , in order to define a market instability index as a quotient @xmath20 that characterizes the prior - to - crash market situation
. then we use the _ processeed _ @xmath13 and @xmath14 in order to strengthen the key properties of @xmath20 that reveal the market collapse situation . :
@xmath13 and @xmath14 do not always coincide .
we consider the simplest of cantordusts : the ternary @xmath21 of cantor in the unit interval @xmath22 $ ] .
we endow the subfractals or `` subcantors''in @xmath23 $ ] and @xmath24 $ ] with probability measures @xmath25 and @xmath26 respectively , @xmath27 , @xmath28 ; @xmath29 , @xmath30 , @xmath31 , and @xmath32 will be the weights of the @xmath33 subfractals of width @xmath34 in the second step of the construction of @xmath21 , and so on . this simplest of cases ascribes probability measure @xmath35 to all subfractals of width @xmath36 in the @xmath37th step , @xmath38 between 0 and @xmath37 .
then @xmath5 ( of such subfractal ) @xmath39 , where @xmath40 , and @xmath38 an integer , @xmath41 .
the corresponding @xmath14 , the same as @xmath13 in this simplest case , is @xmath42 , where @xmath43 reaches its maximum for @xmath44 and is entirely symmetric for @xmath38 at left and right of @xmath45 .
the concentration @xmath5 goes linearly with integer @xmath38 from an `` origin ''
@xmath46 ( where @xmath47 ) to @xmath48 , where @xmath14 reaches its maximum , to @xmath49 , where @xmath14 is again zero . in this simplest binomial case ,
the increasing and decreasing branches of the curve are a mirror image of each other , like an inverted parabola .
( in the case we work with below , such symmetry might be absent , still we will consider @xmath37 as a constant , @xmath5 going linearly with integer @xmath50 ) .
we will simplify this binomial case further , in order to be able to adapt it to our study of the market crash signal .
the binomial case comes from the unfolding of @xmath51 , with @xmath52 and @xmath53 , and the symmetric properties of @xmath43 .
we have a decreasing index for powers of @xmath54 s i.e. for @xmath25sto wit : @xmath55 , @xmath56 , coupled with an exponent increase for powers of @xmath57 or @xmath58 : @xmath59 , ; @xmath60 the corresponding number of them .
notice that the @xmath5 & @xmath14 structure above does not need the ternary of cantor as a starting point : if we divide the unit invertal @xmath61 in halfs , in quarters , where @xmath62 is the size of each interval - box in @xmath37th step , then we simply replace `` 3 '' by `` 2 '' in the denominators of @xmath5 and @xmath14 above .
in fact , if @xmath63 is the number of equal intervals dividing @xmath61 , we still have @xmath64 , with @xmath65 in the denominator of @xmath66 , @xmath63 in that of @xmath46 , and @xmath67 just as symmetric as above . next , consider a number @xmath68 instead of @xmath25 above , and the binomial expression @xmath51 , @xmath69 and @xmath70 .
we now have a decreasing sequence of weights @xmath2 : @xmath71 instead of @xmath72 has been replaced by 1 .
the total @xmath4 weight is now @xmath73 .
now , in order to make probabilities out of these weights @xmath2 we have to write @xmath74;@xmath75 the corresponding number of them . with @xmath63 as before , @xmath37 large and _ fixed _ , @xmath73 the size of the sample , or total number of such elements , we still have , as above , @xmath76 , @xmath77 : @xmath78 ; and @xmath67 is just as symmetric as above .
this will be our working structural framework .
@xmath20 is a function of @xmath5 or of @xmath38 indistinctly , and will be , by definition , @xmath79 , an obviously increasing function of @xmath38 , i.e. of @xmath5 , between @xmath80 or @xmath81and @xmath82 or @xmath83 in our binomial case .
notice that the instability quotient is made up of `` unprocessed '' weights not logarithms of probabilitiesand `` unlogarithmized '' @xmath11 , i.e. with the raw data which , once processed , yields @xmath13 and @xmath84
nq1000 refers to the 1000 daily fluctuations before the april 2000 nasdaq crash ; djia1600 , djia900 , etc .
refer to the 1600 , 900 , etc .
daily fluctuations before the october 1997 dow jones market crash , i.e. 1600 days prior to crash , 900 days ditto , etc .
( rotundo , 2006 ) .
we start with nq1000 , and we represent ( ditto for other cases ) the graph of the closing market price @xmath85 as a function of time @xmath86 one day at a time : one day a unit time .
time @xmath86 the horizontal axis , the corresponding price @xmath85 the vertical one .
we take the signal @xmath87 and project these discrete values onto the vertical axis @xmath85 . this set of @xmath88 points is a cantordust contained in a vertical segment @xmath89 .
we cover vertical @xmath89 with adjacent interval - boxes of equal size @xmath1 .
the number @xmath63 of boxes should not exceed , ideally , the square root of the total @xmath90 points in the signal .
each point in vertical @xmath89 is a certain market price : the higher the point in @xmath89 i.e. the higher the box @xmath0 that contains itthe higher the price @xmath85 .
the weight @xmath2 of box @xmath0 is the number of days in which market prices @xmath85 fell between the vertical bounds or extremes of @xmath0 .
thus , each such box represents an increment @xmath91 in prices @xmath85 .
let @xmath92 and @xmath93 be the extremes of segment @xmath89 : the lower and upper prices . for nq1000 as well as for the djia s , we always notice a rather small set @xmath94 of very heavy boxes quite near @xmath92 .
these @xmath94 heavy boxes represent many days in which market prices stood firm , with constant values , i.e. prices rather stable , not very expensive .
the short lower subsegment in @xmath89 between @xmath94 and @xmath92 represents prices going down hence its smallness : we will do without it , as generally , prices tend to go up .
hence , we will work with heavy boxes as the lowest @xmath0 boxes , and study the nature of price increase .
the remaining vertical upper segment in @xmath89 , from heaviest boxes at the bottom to the lightest ones on top , will be normalized to be the unit segment @xmath95=i$ ] , the cantordust thus reduced will be _ the _ fractal cantordust @xmath21 . since the signal has a total @xmath4 of 1000 points , then @xmath63 the number of @xmath0 boxesshould not exceed @xmath96 .
the sum of all weights in all boxes is 1000 .
the situation at the top , near @xmath93 , presents a problem : there are boxes with just 1 point , on top of which we find a box with 0 points , then , on top , another box with 1 point these isolated and extremely light boxescorrespond to prices quite high and most unstable , for they came to pass during 1 day ( the weight @xmath2 of that @xmath0 box ) only , a situation made worse by the upper _ empty _ box , followed by an even _ higher _ box again with 1 point even higher market price , during one day , with an increase of @xmath97 : skipping the empty box
these abrupt jumps in prices not only reek of market crash , they pose a geometrical problem , for isolated and far away points do not belong to a fractal a problem we will deal with below .
the boxes at the bottom of vertical @xmath61 , being the heaviest , correspond to @xmath80 , the highest ones , being the lightest , to @xmath8 .
we have a number of things superimposed , illustrated in diagram d ( fig .
1 ) vertical cantordust @xmath21 in @xmath61 , the sequence of price increases @xmath91 ( one such increase for each box @xmath0 with weight @xmath2 ) , vertical axis @xmath85 with stable price for lower box and higher price for upper one , and the corresponding vertical interval @xmath98 $ ] , with @xmath82 somewhere in the middle .
we will concentrate our study of market crash situation in two separate parts : the analysis of what happens between @xmath80 and @xmath82 , and the study of the descending branch @xmath82 and @xmath8of the spectrum .
the number of adjacent intervals @xmath99 , of equal size , covering vertical interval @xmath98 $ ] should ideallynot exceed the square root of the number @xmath63 of boxes @xmath0 covering @xmath61 . that leaves us with six such intervals @xmath10 , four of which cover the interval @xmath100 $ ] .
the value @xmath82 is calculated in two ways that agree : it is the value of @xmath5 for which @xmath101 in the lagrangian coordinates . on the other hand , going to the spectrum - by - definition , it is in the fourth interval @xmath10 that all values of @xmath5 in that interval cram together around the value for which @xmath101 , as observed in a histogram that contains _ all _ @xmath5sone for each box @xmath0 .
@xmath11 for that particular @xmath10 is no smaller than for any other @xmath10 , also : its @xmath5 values are the only ones quite near one another , crowded around the value of @xmath5 for which @xmath102 . in the symmetric case above we had a decreasing sequence of weights @xmath103 with @xmath68 , each weight corresponded to a certain @xmath10 , each @xmath10 was reduced to a point in that ideal case , with @xmath104 boxes @xmath0 . here we have the lowest @xmath10 , corresponding to @xmath80 , endowed with @xmath105 such boxes with weights @xmath106 and @xmath107 days , respectively , with an average of @xmath108 days , or @xmath109 months . in the same way in which we identify @xmath5 s in a certain @xmath10 @xmath11 the number of such @xmath5swe will average weights @xmath110 s of boxes @xmath0 in that @xmath10 .
we recall that each @xmath0 box pertaining to an interval @xmath10represents a price increase @xmath91 inside pre - determined bounds @xmath111 $ ] given precisely by the extremes of interval @xmath10 .
the second @xmath10 , immediately above the lower @xmath112 , has @xmath113 boxes @xmath0 with weights @xmath114 and @xmath115 days , respectively , with an average of @xmath116 days or approx .
1.5 months .
the third @xmath10 above has an average of 20 days or three weeks , and the fourth one , reaching @xmath82 , an average of 10 days .
so we have a sequence of @xmath117 months , slightly more than @xmath118 months or @xmath119 weeks , @xmath120 weeks or 20 days , 10 days : it does look like @xmath121 with @xmath122 and @xmath123 slightly larger than @xmath124 indeed , the average of these `` @xmath123 s '' is @xmath125 ; each @xmath126 is taken as the ratio of two consecutive average number of days above .
( in fact , the two remaining @xmath10 s between @xmath82 and @xmath8 do continue with this division by @xmath127 : the average of 5th interval @xmath10 is slightly less than @xmath128 days , and the last one corresponds to the 1 or 2 days per box . )
we can now proceed in _ three different ways _ that indicate an early warning of the market crash situation .
we call it `` early '' because @xmath82 is situated far below @xmath8 , and @xmath8 is at the very top of the market price list : @xmath93 .
so @xmath82 is midway between @xmath92 and @xmath93 , far below @xmath93 .
the logic is that before prices reached @xmath93 values , on their way up , they passed through these critical values corresponding to @xmath82 : hence , what happens at this value is an _ early _ warning of future market collapse . in sec .
2.1 we defined the instability index @xmath129 we recall that @xmath68 , that @xmath37 was fixed , that @xmath130 , so we can write @xmath131 or @xmath132 ; that for the binomial case @xmath132 is steadily increasing with @xmath38 ( or @xmath5 ) , and that we are between @xmath133 and @xmath82 , which means @xmath83 . in this last ideal case , @xmath132 grows with no change of curvature between @xmath81 and @xmath83 ( fig .
2 , out of scale ) . in the case of our nasdaq signal ,
the weights @xmath134 are @xmath135 and @xmath136 days ( sec . 4 , with a value of @xmath137 until we reach @xmath82 ; the four @xmath11 being @xmath138 , and @xmath139 boxes in the four intervals @xmath10 ( sec .
3 ) between @xmath80 and @xmath82 . the graph @xmath132 , in this case , plotted against four integer values of @xmath38 , shown in fig .
3 ( out of scale ) , has an inflection point , between @xmath80 and @xmath82 , near the latter .
the values @xmath140 , do not imply that the second and fourth @xmath10 have maxima , for in the fourth @xmath10 the @xmath5 s are concentrated around @xmath141 , whereas in the second @xmath10 they are dispersed .
let us interpet @xmath20 .
the first @xmath142 is much smaller than the second one : @xmath143 : the instability increased . in the first instance , prices increased every 3.5 months , a rather stable situation , and they did so for 4 consecutive @xmath91 price increases .
next , the prices began to increase @xmath91 each month and a half , a situation that repeated itself seven times , that is why @xmath20 increased .
the inflection point above is due to an anomaly in the evolution of the market : 5 consecutive price increases @xmath91 each 3 weeks are immediately followed by price increases each 10 days only ! and this abnormal jump takes place 7 consecutive times the logic underneath is the following : we had a price increase each month and a half , 7 times .
next , we find a @xmath91 increase each 3 weeks , which is a much worse situation .
but , if this situation is repeated a rather - large - than-7 times _ instead of 5 times _ , there is room for the system to adapt , there is valuable time to adjust to the new situation _ before _ time intervals between consecutive @xmath91 shorten again .
this numerical anomaly in the values of @xmath11 causes the inflection point in the graph of @xmath131 .
so , real stability of the system does depend on @xmath20 remaining low , but if it grows , the occurrence of a crash depends on the geometry of its graph ( inflection point ) , as described above .
we continue with nasdaq , we take boxes @xmath0 with weight @xmath2 , divide each @xmath2 by @xmath90 and obtain probabilities @xmath3 .
then we can apply the thermodynamical algorithm and obtain @xmath14 , as shown in fig 4 .
two observations : a ) the curve @xmath144 is _ not _ symmetric as is the binomial one ; b ) there is a noticeable gap between @xmath145 and @xmath141 entirely in the interval @xmath100 $ ] . since @xmath141 means @xmath146 , we observe that the inflection point of @xmath20 , very near @xmath82 , is located precisely in that gap . if we consider a time span where there is no important crash , e.g. the daily fluctuations of dow jones from 1st jan .
1990 to 1st jan .
1994 , the corresponding curve @xmath144 _ is _ symmetric , as shown in fig .
5 . it should be noticed that symmetric spectra are quite rare .
this case is , as far as we are aware , unique in empirical context the binomial case is a purely mathematical one .
moreover , if we take the data from the djia1600 sample , still the corresponding @xmath144 curve mantains both the gap _ and _ the asymmetry ( fig .
6 ) : enlarging the prior - to - crash sample does not `` dilute '' these crash features . further
, if we do the opposite , that is , take smaller and smaller samples prior to crash : djia900 , 800 , 700 , 600 , always maintaining @xmath147 ( @xmath63 = number of @xmath0 boxes , @xmath4 = size of the sample , all as above ) , we obtain _ the same _ @xmath14 as in fig . 4 with minute variations ( fig .
7 ) : the process is _ strictly _ self similar ( i.e. independent of size sample , or of scale ) around market crash point . : in order to plot the thermodynamical lagrangian spectra curve @xmath144 in each case throughout this section , we have used only integer values for @xmath148 , since they are enough to determine the key properties of @xmath14 : the graph dots for @xmath149 , would be deceiving , since the corresponding graph points at the extremes of the curve would simply pile up on top of each other , hence the upper gaps would fill up artificially , without adding any real meaning to the geometry of the lagrangian process .
we continue with nasdaq data .
recall that @xmath90 , and that we have between 30 and 40 @xmath5 s , from boxes @xmath0 grouped in 6 intervals @xmath10 , the first four span @xmath100 $ ] , the @xmath5-interval of our study .
we also recall that @xmath13 is the spectrum - by - definition and @xmath14 the lagrangian one .
the first @xmath10 interval , the one corresponding to @xmath80 , has 4 @xmath5 s , more or less distributed evenly in the interval .
the second adjacent @xmath10 ( seven @xmath5 s ) has its @xmath5 s not so dispersed , some are near the @xmath5 for which @xmath150 .
the third @xmath10 has very few and dispersed @xmath5 s .
so @xmath13 and @xmath14 share the same @xmath80 , then @xmath13 grows to a local maximum at an @xmath5 for which @xmath145,
then it goes _ down _ , and it does it in the gap between @xmath145 and @xmath141 .
the fourth interval @xmath10 includes the particular @xmath5 for which @xmath141 .
all its @xmath5 s are clustered around this particular @xmath5 , which implies that a spectrum ( another one ) @xmath151 should have an _ absolute maximum _ at that value of @xmath5 , just as @xmath14 does in the 5th and 6th remaining @xmath10 intervals the number of @xmath5 s steadily decreases , and so does this last @xmath13 .
the picture corresponds to _ two _ @xmath13 : @xmath152 starts in @xmath80 , reaches a maximum @xmath153 for @xmath150 , and descends in the gap between @xmath154 and @xmath102 .
the other , @xmath155 , goes from its own maximum at @xmath156 ( for which @xmath101 ) and decreases as @xmath5 nears @xmath8 ( shared by @xmath155 and @xmath14 ) . by considering a larger sample prior to market crash , djia1600 , we have one more @xmath10 , which clarifies the situation : @xmath155 starts with a _ low _ value in the gap between @xmath153 and @xmath156 ( i.e. @xmath154 and @xmath102 ) , then grows to an absolute maximum at @xmath156 , where @xmath157 , and then decreases as @xmath5 nears @xmath8 : clearly , two different @xmath158 ( with negative second derivative ) , which seem to collide in `` the gap '' ; fig
. 8 shows a schematic diagram representing this situation .
the two @xmath158 multifractal spectra bi - multifractalityare two curves for which @xmath14 acts as an envelope , as a smooth upper `` roof '' of both . according to radon and stoop ( 1996 ) , this situation bi - multifractality with an upper envelopeoccurs when _ two _ different fractal measures are processed with _ one and the same _ multifractal algorithm . and @xmath159,width=377 ]
as @xmath5 approaches @xmath8 , the market prices increase more and more , to the point in which they can be sustained during one day only . this situation , as we noticed before , is reflected in boxes with 1 or 0 points ( days ) near the @xmath93 top .
we have several such boxes at the top prices , separated from each other by zero point empty boxes .
this situation , which we call `` dispersion '' , depicts a number of isolated points far away from the main body of the fractal cantordust , and , from the point of view of fractal geometry , they should be discarded no isolated point contributes dimensionally : that is why we used several times the expression `` between 30 and 40 boxes '' , for quite some such boxes have to be discarded when doing a spectrum - by - definition @xmath13 . instead , the lagrangian @xmath14 process does not seem to care _ where _ the boxes are : isolated and far away or in the very midst of the cantordust , but _
how many _ they are , for each value of @xmath5 , regardless of their location .
of these two spectra , we choose @xmath14 in order to interpret and quantify the dispersion , since it takes into consideration _ all _ the dispersed boxes near @xmath8 . near crash
, prices jump so high , and so high , that we have many dispersed boxes : @xmath160 is large , and so is @xmath161 , as we observe for nasdaq crash in fig .
4 , where the large dispersion in @xmath8 substantially contributes to the non - symmetry of the curve . carefully notice that , when we enlarge the prior - to - crash sample , e.g. djia 1600 , fig .
6 , we slightly `` dilute '' the crash effect on the descending right branch of the corresponding @xmath14 , for @xmath161 is much smaller and the dots closing the right branch are much more separated than in the nasdaq case , though they are drawn with the same parametric @xmath17 values as for nasdaq .
also , fig .
7 shows that as we take the sample nearer and nearer crashpoint , djia900 , 800 , 700 , 600, the higher the value of @xmath161 : the highest one corresponds to djia600 ( the numerically smallest sample which still yields a cantordust ) . the economic interpretation : those extraordinarily high jumps in market prices the dispersionincrease as the system destabilizes , as we get dangerously near its collapse . :
an observation on the inflection point in @xmath20 ( fig .
3 ) : @xmath20 grows between @xmath80 and @xmath82 , since the numerator @xmath162 grows , and the denominator is a decreasing power .
when @xmath163 grows ( up to @xmath154 ) then @xmath20 grows , ditto when @xmath164 grows but , in the middle of these two growing spectra , exactly between @xmath154 and @xmath102 , @xmath20 does not grow but sits down , hence the inflection .
a prior - to - crash market signal is studied as a cantordust , its fractal properties , responsible for the crash situation , are studied with the tools of multifractal analysis .
two different multifractal spectra @xmath144 , and an instability quotient @xmath20 characterize and quantify an `` early warning '' of market collapse , with market prices ( not yet very high ) corresponding to values of @xmath5 between @xmath80 and @xmath82 ; the study of the highest market prices , corresponding to values of @xmath5 near @xmath8 , is done by analysing the fractal dispersion of the cantordust prior - to - crash market signal .
radon , g. and stoop , r. ( 1996 ) .
`` superpositons of multifractals : generator of phase transition in the generalized thermodynamic formalism ''
. j. stat .
3/4 , 163 - 180 .
rotundo g. ( 2006 ) .
`` logistic function in large financial crashes '' in logistic map and the route to chaos .
springer ny . | we take prior - to - crash market prices ( nasdaq , dow jones industrial average ) as a signal , a function of time , we project these discrete values onto a vertical axis , thus obtaining a cantordust .
we study said cantordust with the tools of multifractal analysis , obtaining spectra by definition and by lagrangian coordinates .
these spectra have properties that typify the prior - to - crash market situation .
any of these spectra entail elaborate processing of the raw signal data . with the unprocessed raw data we obtain an instability index , also with properties that typify the prior - to - crisis market situation . both spectra and the instability index
agree in characterizing such crashes , and in giving an early warning of them .
* m. piacquadio * secretara de investigacin y doctorado , facultad de ingeniera , universidad de buenos aires paseo coln 850 , ciudad autnoma de buenos aires , argentina * f. o. redelico * laboratorio de sistemas complejos , facultad de ingeniera , universidad de buenos aires paseo coln 850 , ciudad autnoma de buenos aires , argentina facultad de ciencias fisicomatemticas e ingeniera , universidad catlica argentina alicia moreau de justo 1500 , ciudad autnoma de buenos aires , argentina |
cold dark matter ( cdm ) theory has serious issues on galactic and sub - galactic mass scales , even though it has been successful in interpreting a multitude of observational phenomena at cosmological scales .
these include the so - called missing satellite , core / cusp discrepancy , and too - big - to - fail ( tbtf ) problems .
one of the possible mechanisms to resolve these problems within the cdm paradigm is to rely on the effects of baryonic process such as the photoionization of neutral gas with uv background and the impact of gas outflow mainly due to mainly supernova feedback on dark halo structures .
more radically , a possible solution that is well motivated from particle physics is to replace cdm with warm dark matter ( wdm ) .
the main difference between cdm and wdm is that while cdm particles have negligible thermal velocities , wdm ones have significant thermal velocities during the era of structure formation and thus these particles have a free - streaming length , giving rise to a cutoff matter power spectrum at larger wave numbers corresponding to small spatial scales .
the possible wdm candidates are the sterile neutrino ( see , e.g. , * ? ? ?
* for a review ) or the gravitino ( e.g. , * ? ? ?
these candidates would have a mass in a few kev , leading to the suppression of structures on dwarf - galaxy mass scales .
therefore , since the cutoff wavelength depends on the particle mass , investigating the nature of dark halo structures at low - mass scales leads to useful limits on the wdm particle mass , @xmath2 .
one of the powerful tools for constraining the wdm particle mass is ly@xmath3 forest observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
these observations allow us to study a matter power spectrum down to small scales and over a large redshift range . @xcite
have observed the flux power spectrum of ly@xmath3 using the high - resolution spectra and have compared it to predicted ones from high - resolution @xmath4-body and hydrodynamical simulations , assuming that wdm masses are 1 , 2 , and 4 kev .
they concluded that in order to reproduce the observed power spectrum , wdm particles are required to have a lower limit of @xmath5 kev ( @xmath6 confidence level ) .
dwarf spheroidal ( dsph ) galaxies also provide useful tests to distinguish between cdm and wdm or set constraints on particle mass of wdm .
first , based on the difference of dark matter concentration between cdm and wdm subhalos , reproducing the maximum circular velocity of the milky way ( mw ) satellites can distinguish dark matter scenarios .
this test is linked to the tbtf problem in cdm .
@xcite performed @xmath4-body simulations with wdm and presented that the tbtf problem does not exist within the frameworks of wdm .
also , @xcite showed that a wdm particle mass less than 4 kev can mitigate the tbtf problem in high - resolution simulations .
the second test is the method to resolve the missing satellite problem .
based on the difference of the abundance of subhalos within a host cdm and wdm dark halo , @xcite found a conservative lower limit of @xmath7 kev in each @xmath4-body simulation .
however , these tests largely rely on the mass of the mw halo , where the estimates of the mw halo mass is still very uncertain with mass ranging from @xmath8 to @xmath9 ( e.g. , * ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
for the case of the tbtf problem , cdm models can resolve the issue if the mass of the mw halo is less than @xmath10 @xcite .
for the abundance of subhalos , if the mw halo mass is heavier than @xmath11 , only masses of wdm particles greater than 2 kev are accepted @xcite . because of the dependence on the assumed mw halo mass and the uncertainty in it , the two tests above using the dwarf satellites are not yet sufficient to confine cdm or wdm scenarios . in this letter , as yet another test on this issue from dwarf - galaxy scales , we revisit and consider the dark matter surface density , which appears nearly constant and independent of galaxy luminosity discovered by previous studies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
we propose that the re - definition and adoption of this surface density estimated within a radius of maximum circular velocity directly allow us to obtain the limits on particle masses of wdm .
this work is in part based on the kinematical analysis of the mw and andromeda ( m31 ) dsphs , for which more detailed descriptions will be presented elsewhere ( * ? ? ?
* in preparation ) .
recently , analyzing the data of velocity dispersions , rotation curves of hi gas , and weak lensing for samples of dsph , dwarf irregular , spiral , and elliptical galaxies , @xcite and @xcite fitted these data based on cored dark matter density profiles , such as the burkert and pseudo - isothermal profiles , and found that the product of the central core density and core radius , @xmath12 , which is proportional to the central dark matter surface density , is constant for all these galaxies .
while these works offer us important suggestions on the universality of the dark halo structure , we note that this definition of @xmath12 can not be applied directly to a system with a cusped density profile .
both @xmath13 and @xmath14 in such a case may be defined in a specific way depending on an assumed density profile , so there exists some ambiguities when we compare surface densities between different systems and also compare with predictions from cdm or wdm . to adequately compare dark matter distributions being independent of their density profiles
, we introduce a mean surface density of a dark halo within a radius of maximum circular velocity , @xmath15 where @xmath16 @xmath17 indicates any dark matter density profiles , and @xmath18 is a radius at maximum circular velocity , @xmath19 , of assumed dark halo profiles , @xmath20 . the surface density defined in equation ( [ svmax ] )
is exactly proportional to the central dark matter surface density , @xmath21 , where @xmath22 and @xmath23 are the central density and scale length of arbitrary dark halo profiles , respectively .
we note that @xmath18 is a better - defined scale than a core radius , @xmath14 , because the latter is uncertain for a cuspy density profile : @xmath14 may be defined as a half - mass radius or some fraction of it in such a case , but it then suffers from the ambiguity of at what outer boundary the halo is tidally limited , e.g. , the navarro frenk white ( nfw ) profile having a diverging total mass . , within a radius at maximum circular velocity , @xmath19 , as a function of @xmath19 for different galaxies and hubble types . while the inset shows all of the samples as well as cdm ( _ wmap7 _ ) prediction , the main panel highlights the plots only at @xmath24 km s@xmath1 .
diamonds denote the mw and m31 dsphs derived in ( * ? ? ?
* in preparation ) , where magenta and cyan ones correspond to dsphs having a cored and cusped dark halo profile , respectively .
blue and red triangles are the original @xcite sample of spiral galaxy data and nearby spirals in things @xcite , respectively .
the data for dwarf irregular galaxies are indicated by green circles , and spirals and ellipticals investigated by weak lensing are labeled by black squares .
solid lines show the results of cdm models with different cosmological parameters of _ wmap7 _ ( dark green ) and _ planck _ ( orange ) measurements .
red and green dashed lines denote wdm models with particle masses of 2 and 3 kev , respectively.,width=302 ] using this definition ( [ svmax ] ) , we evaluate @xmath25 within @xmath18 taken from the literature based on the hi gas rotation curve of late- and early - type spirals with pseudo - isothermal dark halos @xcite , and dwarf irregulars with burkert dark halos @xcite , and the galaxy galaxy weak lensing from spiral and elliptical galaxies with burkert profiles ( analyzed by @xcite ; data from @xcite ) . for smaller , dwarf - galaxy scales ,
we analyze the stellar kinematics in the seven mw ( carina , fornax , sculptor , sextans , draco , leo i , and leo ii ) and five m31 ( and i , and ii , and iii , and v , and and vii ) dsphs and construct axisymmetric mass models for these dark - matter - dominated dwarf galaxies ( * ? ? ?
* in preparation , which considers more generalized models than our previous work , @xcite ) .
for all samples , except for dsphs , we use the dark halo parameters ( central density and core radius ) estimated by the above original papers . for dsphs
, we employ the axisymmetric jeans equations as in @xcite , but we consider the effect of an additional velocity anisotropy of tracer stars , @xmath26 , to determine the global shape of a dark halo , where @xmath27 and @xmath28 are velocity dispersions of stars in cylindrical coordinates ( see ( * ? ? ? * in preparation ) , for more details ) . together with @xmath29
, we adopt an inner slope of the dark halo density profile as a free parameter as well as the parameterization of central density and scale length of a dark halo , apply our models to the two - dimensional line - of - sight velocity dispersion map for the above dsphs , and obtain the best - fit dark halo parameters .
we have found that whilst most of the dsphs have cored or shallower cusped dark halos , the draco , leo i , and iii , and and v dsphs show a steep inner density slope ; not all of the dsphs have cored density profiles in their dark halos , as already suggested in @xcite . the symbols with error bars in figure [ svmax2map ]
show the estimated @xmath25 of the above data sample as a function of @xmath19 .
it is found that even though dark halos in each galaxy sample are assumed to have different mass profiles ( cusp or core ) and are estimated by independent methods , the dark matter surface density within @xmath18 is sufficiently _ constant _ across a wide range of galaxy masses .
we discuss probable systematic uncertainties for the disk and dsph galaxies , respectively . for the disk galaxies ,
we investigate the impact of the uncertainties in the adopted stellar mass - to - light ratio , @xmath30 , on @xmath25 .
@xcite assumed the three @xmath30s and fitted dark halo parameters for each assumption . adopting these parameters
, we confirm that these uncertainties in @xmath30 do not largely affect the estimation of @xmath25 , varying it by a factor of three at maximum . for dsphs , we consider the influences of baryonic components and the velocity anisotropy of stars on the estimation of @xmath25 .
since the fraction of the stellar mass in the total mass of the dsphs is so small , ranging from @xmath31 to @xmath32 , that the estimated values of @xmath25 may not at least change beyond a factor of 10 , even if these small effects of baryonic components are taken into account . in our models , we assume a _ constant _ velocity anisotropy of their member stars and find that this anisotropy is almost independent of other model parameters , such as a central density and scale length as well as an inner slope of a dark halo ( * ? ? ?
* in preparation ) .
therefore , the observational evidence for the constancy of @xmath25 is robust .
the use of the dark matter surface density defined by using @xmath19 and @xmath18 allows us to compare the results from the data sample with those from cosmological @xmath4-body simulations within the @xmath33-dominated cdm and wdm models .
we calculate @xmath25 from the outcomes of the @xmath4-body simulations as follows .
first , we adopt the nfw dark matter density profile : @xmath34 which can reproduce cdm ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and wdm ( e.g. , * ? ? ?
* ; * ? ? ?
* ) dark halos within a wide halo mass range .
second , we calculate the scale density , @xmath35 , and virial radius , @xmath36 , for a nfw dark halo at the present time ( @xmath37 ) from cosmological context : @xmath38 where @xmath39 is a so - called virial overdensity at @xmath37 @xcite , @xmath40 is a concentration parameter defined as @xmath41 , @xmath42 is a critical density of the current universe , and @xmath43 denotes the hubble constant at @xmath37 .
third , using the mass concentration relation ( mcr ) for a dark halo , we obtain the scale radius , @xmath44 . for cdm models , we adopt two different mcrs at @xmath37 based on @xmath4-body simulations with slightly different cosmological parameters : @xmath45 & & \nonumber \\ { \rm ( for\hspace{2mm}wmap7 ) } & & , \nonumber \\
c_{\rm p } = 10.7\bigr(\frac{m_{\rm vir}}{10^{12}h^{-1}m_{\odot}}\bigl)^{-0.11 } \bigr[1 + \bigr(\frac{m_{\rm vir}}{2.4\times10^{16}h^{-1}m_{\odot}}\bigl)^{0.4}\bigl]&&\nonumber \\ { \rm ( for\hspace{2mm}planck ) } & & \nonumber,\end{aligned}\ ] ] where @xmath46 is a dimensionless hubble constant .
the mcr with cosmological parameters of _ wmap7 _ is derived by @xcite , whilst another one is found by @xcite . for wdm models ,
we adopt the method to derive mcr using the power spectra developed by ( * ? ? ?
* hereafter s14 ) .
in order to evaluate the power spectrum of wdm models , the transfer function formula of the wdm universe by @xcite is adopted : @xmath47^{-5/\mu},\ ] ] with @xmath48 and @xmath49^{-1.11}\bigr[\frac{\omega_{\rm wdm}}{0.25}\bigl]^{0.11 } \bigr[\frac{h}{0.7}\bigl]^{1.22 } \hspace{3 mm } h^{-1}{\rm mpc},\ ] ] where @xmath50 is a wavenumber and @xmath51 denotes a cosmological density parameter for wdm .
then , the _ linear _ power spectrum in the case of wdm is derived by @xmath52 using the above power spectrum and mcr s fitting function described by equation ( 24 ) in @xcite , mcrs in the case of 2 and 3 kev particle masss of wdm are derived by comparing with the @xmath4-body simulations ( see s14 for more details ) .
figure [ cmvir ] displays the mcrs derived from cdm models with _ planck _
data and wdm models with 2 and 3 kev particle masses at redshift 0 .
while cdm models predict that a concentration parameter monotonically increases with decreasing halo mass , wdm halos have a clear turnover at a low - mass halo . finally , using the dark matter density profiles obtained by the above procedure
, we compute @xmath19 and @xmath18 , and then calculate @xmath25 for cdm ( _ wmap _ and _ planck _ ) and wdm ( particle masses of 2 and 3 kev ) models .
the solid and dashed lines in figure [ svmax2map ] are predicted @xmath25 versus @xmath19 for four cases of the dark matter models .
we emphasize that , surprisingly , these theoretical lines can well reproduce the data at high - mass scales , even though we do not perform any fitting to the data ; actually , the data points ( including low - mass scales ) appear slightly above even the theoretical line giving the highest @xmath25 ( cdm with _ planck
_ parameters ) , which may suggest there are still some uncertainties in structural parameters for dark halos or even cosmology , but the following results for the limit of wdm remain unaffected . for all mass ranges ,
cdm scenarios give a reasonable match to the data , where the results based on _ wmap _ and _ planck _ data are not so different at low - mass scales .
this is only due to the slight effect of the difference of cosmological parameters on the concentration of dark halos . in particular , @xmath19 or @xmath18
can depend on both the rms amplitude of linear mass fluctuation in a sphere of @xmath53 mpc , @xmath54 , and the spectral index of the initial power spectrum , @xmath55 @xcite . on the other hand ,
wdm ones are inconsistent with the universality of @xmath25 at dwarf - galaxy mass scales .
this different feature between cdm and wdm models at the low - mass end stems from the properties of the halo concentrations at low halo - mass scales . from figure [ cmvir ] ,
the concentrations of wdm halos differ substantially from those of cdm halos at low halo - mass scales , thereby the value of @xmath25 decreases prominently at small @xmath19 corresponding to dwarf galaxies .
we also investigate how far wdm models deviate from the data .
figure [ sver ] is the same as figure [ svmax2map ] , but zooms in the dwarf - galaxy scale and plots @xmath56 and @xmath57 error areas of each sample .
focusing on @xmath58
km s@xmath1 where the difference in @xmath25 emerges significantly , even a 3 kev particle mass of wdm appears to deviate beyond a @xmath56 error in our sample .
following this result , in order to be in agreement with the universality of @xmath25 from the data , wdm particles need to be heavier than 3 kev .
, but for dsphs only .
black and gray shaded regions are @xmath56 and @xmath57 error areas , respectively , for each sample.,width=302 ]
in this letter , we define and estimate a mean surface density of a dark halo within a radius of maximum circular velocity , @xmath25 and apply it to dark halos in various galaxies .
this physical value reflects the central dark matter surface density , thus not strongly affected by the influence of external disturbances .
we find that even though dark halo parameters in each galaxy sample are derived based on the assumption of different dark halo density profiles and determined by several independent methods , this surface density is universal across a wide luminosity range of galaxies , irrespective of dark matter distribution in each of the galaxies .
furthermore , this constancy is found to be useful for constraining dark matter models , especially the particle mass of wdm , because @xmath25 is directly derivable from the theoretical predictions based on @xmath4-body simulations .
our results show that at larger values of @xmath19 corresponding to high halo - mass scales , this constancy for real galaxies can be successfully reproduced in both cdm and wdm models , even though we do not perform any fitting to the data .
however , @xmath25 derived from wdm largely deviates from the observational constancy at dwarf - galaxy mass scales , whilst that of cdm is in good agreement with it .
this is because there is a prominent difference between mcr for cdm and wdm at low halo - mass scales . in order to be in agreement with this constancy
, we have found that wdm particles need to be heavier than 3 kev .
this letter proposes an astrophysical test on the nature of dark matter .
the key dark halo mass range to distinguish dark matter models in this test is the lowest halo - mass scales , such as those of classical and ultra - faint dwarf galaxies .
wide - field spectroscopic studies of these galaxies over large areas with the planned prime focus spectrograph to be mounted on the subaru telescope and high - precision spectroscopy for a plenty of much fainter stars with the thirty meter telescope will provide a better determination of dark halo properties of the mw and m31 dsphs , and thus allow us to increase our knowledge about the basic properties of dark matter .
the authors thank the referee for her / his constructive comments that have helped us to improve our paper .
this work is supported in part by a grants - in - aid for scientific research from the japan society for the promotion of science ( jsps ; no .
26 - 3302 for kh ) .
this work has been supported in part by a grant - in - aid for scientific research ( 25287062 ) of the ministry of education , culture , sports , science and technology in japan and by the jsps core - to - core program `` international research network for dark energy . '' boylan - kolchin , m. , bullock , j. s. , sohn , s. t. , besla , g. , & van der marel , r. p. 2013
, , 768 , 140 burkert , a. 2015 , arxiv:1501.06604 de blok , w. j. g. , walter , f. , brinks , e. , et al . 2008 , , 136 , 2648 deason , a. j. , belokurov , v. , evans , n. w. , et al .
2012 , , 425 , 2840 donato , f. , gentile , g. , salucci , p. , et al . 2009 , , 397 , 1169 gentile , g. , burkert , a. , salucci , p. , klein , u. , & walter , f. 2005 , apjl , 634 , l145 gentile , g. , famaey , b. , zhao , h. , & salucci , p. 2009
, , 461 , 627 gentile , g. , salucci , p. , klein , u. , & granato , g. l. 2007 , , 375 , 199 hayashi , k. , & chiba , m. 2012 , , 755 , 145 hayashi , k. , & chiba , m. 2015 , in preparation hoekstra , h. , hsieh , b. c. , yee , h. k. c. , lin , h. , & gladders , m. d. 2005 , , 635 , 73 kafle , p. r. , sharma , s. , lewis , g. f. , & bland - hawthorn , j. 2014 , , 794 , 59 kennedy , r. , frenk , c. , cole , s. , & benson , a. 2014 , , 442 , 2487 klypin , a. , yepes , g. , gottlober , s. , prada , f. , & hess , s. 2014 , arxiv:1411.4001 kormendy , j. , & freeman , k. c. 2014 , arxiv:1411.2170 kusenko , a. 2009 , phr , 481 , 1 lovell , m. r. , frenk , c. s. , eke , v. r. , et al .
2014 , , 439 , 300 lovell , m. r. , eke , v. , frenk , c. s. , et al . 2012 , , 420 , 2318 moroi , t. , murayama , h. , & yamaguchi , m. 1993 , phlb , 303 , 289 narayanan , v. k. , spergel , d. n. , dav , r. , & ma , c .- p .
2000 , apjl , 543 , l103 navarro , j. f. , frenk , c. s. , & white , s. d. m. 1996 , , 462 , 563 navarro , j. f. , frenk , c. s. , & white , s. d. m. 1997 , , 490 , 493 piffl , t. , scannapieco , c. , binney , j. , et al .
2014 , , 562 , aa91 polisensky , e. , & ricotti , m. 2014 , , 437 , 2922 prada , f. , klypin , a. a. , cuesta , a. j. , betancort - rijo , j. e. , & primack , j. 2012 , , 423 , 3018 sakamoto , t. , chiba , m. , & beers , t. c. 2003 , , 397 , 899 salucci , p. , wilkinson , m. i. , walker , m. g. , et al . 2012 , , 420 , 2034 schneider , a. , smith , r. e. , macci , a. v. , & moore , b. 2012 , , 424 , 684 schneider , a. 2014 , arxiv:1412.2133 spano , m. , marcelin , m. , amram , p. , et al .
2008 , , 383 , 297 viel , m. , becker , g. d. , bolton , j. s. , & haehnelt , m. g. 2013 , phrvd , 88 , 043502 viel , m. , becker , g. d. , bolton , j. s. , et al .
2008 , phrvl , 100 , 041304 viel , m. , lesgourgues , j. , haehnelt , m. g. , matarrese , s. , & riotto , a. 2005 , phrvd , 71 , 063534 wang , j. , frenk , c. s. , navarro , j. f. , gao , l. , & sawala , t. 2012 , , 424 , 2715 | in an attempt to place an explicit constraint on dark matter models , we define and estimate the mean surface density of a dark halo within the radius of the maximum circular velocity , which is derivable for various galaxies with any dark matter density profiles . we find that this surface density is generally constant across a wide range of maximum circular velocities of @xmath0 10400 km s@xmath1 , irrespective of the different density distributions in each of the galaxies .
this common surface density at high halo - mass scales is found to be naturally reproduced by both cold and warm dark matter ( cdm and wdm , respectively ) models , even without employing any fitting procedures .
however , the common surface density at dwarf - galaxy scales , for which we have derived from the milky way and andromeda dwarf satellites , is reproduced only in a massive range of wdm particle masses , whereas cdm provides a reasonable agreement with the observed constancy .
this is due to the striking difference between mass concentration relations for cdm and wdm halos at low halo - mass scales . in order to explain the universal surface density of dwarf - galaxy scales in wdm models
, we suggest that wdm particles need to be heavier than 3 kev . |
there has been a long history for the research of epidemic spreading @xcite . and
in the general case , the epidemic system can be represented as a network where nodes stand for individuals and an edge connecting two nodes denotes the interaction between individuals . in the past
, researchers mainly focused the disease transmission study on the conventional networks @xcite such as lattices , regular tree , and er random graph . since late 1990s
, scientists have presented a series of statistical complex topological characteristics @xcite such as the small - world ( sm ) phenomenon @xcite and scale - free ( sf ) property @xcite by investigating many real networks including the internet @xcite , the www @xcite , the scientific web @xcite , the protein networks @xcite and so on .
subsequently , the studies of dynamical processes on complex networks also have attracted lots of interests with various subjects @xcite , and as one of the typical dynamical processes built on complex networks , epidemic spreading has been investigating intensively once more . the basic conceptual tools in understanding the epidemic spreading and the related effective strategies for epidemic controlling should be the epidemiological models @xcite . among the numerous possible models ,
the most investigated and classical models are si model @xcite , sis model @xcite and sir model @xcite , which can approximately describe the spreading of real viruses such as hiv , encephalitis , influenza virus in biological networks ; computer virus , trash mail in technological networks ; and even gossip in social networks .
the most valuable result in standard sir model ( or sis model ) is that the critical threshold ( of transmission rate ) vanishes for the scale - free networks in the limit of infinite network size @xcite . in consideration of epidemic spreading in real cases , there
yet has been some inappropriate assumption in the details of the standard sir model .
we know that in the classical sir model the transmission rate @xmath3 is a constant , but in the real world , @xmath3 should be different among individuals . based on this assumption , in reference @xcite , jaewook joo et al . proposed the effective transmission rate that they introduced a effective coefficient @xmath4 based on standard transmission rate @xmath3 for the edge @xmath5 ; similarly , in reference @xcite , ronen olinky et al .
also studied the effectiveness of the transmission rate , and in their work , the transmission rate is @xmath6 where @xmath7 means the probability that a susceptible node actually acquires the epidemic through an edge connected an infected node with degree @xmath8 . in these previous studies ,
the transmission rate on a given edge is just treated as a function with degrees of the two connecting nodes , which will induce the transmission rates of two opposite directions on the same edge are symmetrical . and
, their analytical methods and results which are based on the stationary state of the sis model may be not valid in the sir model .
thus , in order to make the transmission rate accord with the realistic cases much more , we take into account the effects of the weights of edges and the strengths of nodes which are of great importance measures in the weighted networks @xcite . and
indeed , the weight ( or the strength ) is one of the most important indications in lots of real networks , for example , in social networks it can represent the intimacy between individuals ; in the internet the weight can imply the knowledge of its traffic flow or the bandwidths of routers @xcite ; in the world - wide airport networks it can evaluate the importance of a airport @xcite , and so on @xcite . particularly , for epidemic
spreading , the weight can indicate the extent of frequency of the contacting of two nodes in scale - free networks , the larger the weight is , the more intensively the two nodes communicate , at the same time , the more possible a susceptible individual will be infected through the edge where the transmission rate is larger . on the other hand , in the classical sir model , each infected individual can establish contacts with all his / her acquaintances ( neighbors ) within one time step , that is to say
, each infected node s infectivity equals its degree .
but in the real case , a individual ca nt contact all his intimate friends , particularly when he is a patient . in reference
@xcite , the infectivity is assumed as a constant @xmath9 , which means each infected individual will generate @xmath9 contacts at each time step .
recently , fu et al . proposed a piecewise linear infectivity @xcite , which means : if the degree @xmath8 of a node is small , its infectivity is @xmath10 ; otherwise its infectivity is @xmath9 as a saturated value when @xmath8 is beyond a constant @xmath11 .
both the constant or the piecewise linear method , the heterogeneous infectivity of nodes with different degrees is not considered as adequately as possible in scale - free networks , that is to say there may be some nodes with different degrees which have the same infectivity , and there will be a large number of such nodes if the constant @xmath9 is assigned irrelevantly or the size of underlying networks is infinite .
so , in order to solve these problems , we introduce the nonlinear infectivity , namely , a infectivity exponent @xmath0 will be introduced to take control of the number of contacts that a infected node generates within one time step , and the @xmath0 is between 0 and 1 which is convenient to adjust for different scale - free networks . in this paper
, we present the modified sir model where the infectivity exponent @xmath0 and the weight exponent @xmath1 are added ; based on the modified model , the dynamical differential equations for epidemic spreading is proposed .
we parse the equations to investigate the threshold behavior and propagation behavior for epidemic spreading ; and the analytical results we obtain is verified by the necessary numerical stimulations .
we show that one can adjust the values of @xmath0 and @xmath1 to rebuild the epidemic threshold to a nonzero finite value for different networks , which can prohibit or delay the epidemic outbreaks to some extent .
and we find @xmath0 is more sensitive than @xmath1 in the transformation of the epidemic threshold and epidemic prevalence , which indicates the intrinsic factor ( the infectivity exponent @xmath0 ) take more responsibility than the extrinsic factor ( the weight exponent @xmath1 ) for the epidemic outbreaks in large scale - free networks .
epidemic modeling has a history of researching , and mathematicians also put forward many epidemic models @xcite . in the domain of complex networks ,
sir model @xcite is one of the most investigated and classical epidemic models . in the standard sir model
, individuals can be divided into three classes depending on their states : susceptible ( healthy ) , infected and removed ( immunized or dead ) . in order to take into account the heterogeneity induced by the presence of nodes with different degrees , we use @xmath12 , @xmath13 , @xmath14 to denote the densities of susceptible , infected and removed individuals with degree @xmath8 at time @xmath15 , respectively . and
these variables are connected by means of the normalization : @xmath16 .
the global quantities such as the ( average ) epidemic prevalence are therefore expressed by an average over the various degree classes , i.e. , @xmath17 .
for the standard sir model , the epidemic evolves by the following rules : at each time step , a susceptible individual acquires the infection at the transmission rate @xmath3 in one contact with any neighboring infected individual , which means if a susceptible individual has a edge connecting a infected individual , the disease will transmitted to the susceptible one through the edge with a specific probability @xmath3 . on the other hand , the infected ones will recover and become immune ( ca nt be infected any more ) with rate @xmath18 , one can set @xmath19 without loss of generality . for a comparison
, firstly we review some classical results from reference @xcite , where moreno et al . used the mean field theory to describe the dynamical differential equations of sir model as follows : @xmath20
@xmath21 @xmath22 where @xmath23 denotes the conditional probability for a node with degree @xmath8 to connect a node with degree @xmath24 . in the uncorrelated case , moreno et al . obtained the epidemic threshold : @xmath25 , which implies the absence of the epidemic threshold in a wide range of scale - free networks @xmath26 , @xmath27 .
this result is a bad message for epidemic controlling and preventing , since the epidemic will prevail in many real networks with any nonzero value of transmission rate @xmath3 .
in this section , we will give a detailed investigation about the modified sir model into which we introduce the weighted transmission rate and nonlinear infectivity .
the results we obtain might deliver some useful information for the epidemiology . and for a better analysis , we firstly describe the general differential equations for sir model based on the mean field theory , as follows : @xmath28 @xmath29 @xmath30 where @xmath31 , @xmath32 , @xmath33 have the same meaning with the standard sir model ( see section 2 ) ; and @xmath34 , @xmath35 denote the infectivity of nodes with degree @xmath24 and the transmission rate from nodes with degree @xmath24 to nodes with degree @xmath8 , respectively .
different from the previous studies , in this paper , we mainly focus the sir model on the weighted networks . among varieties of weighted patterns in complex networks
, making use of nodes s degrees to express the weights of edges is very important , namely , the weight between two nodes with degree @xmath8 and @xmath24 may represent as a function of their degrees @xcite , i.e. , @xmath36 where the basic parameter @xmath37 and the exponent @xmath1 depend on the particular complex networks ( e.g. , in the _ e.coli _ matabolic network @xmath1 = 0.5 ; in the us airport network ( usan ) @xmath1 = 0.8 @xcite ; in the scientist collaboration networks ( scn ) @xmath1 = 0 @xcite ) .
noteworthily , the weight @xmath38 belongs to an edge , similarly , a node ( with degree @xmath8 ) also can be measured by weights , i.e. , the strength of a node ( with degree @xmath8 ) , which can be obtained by summing the weights of the links that connected to it , i.e. , @xmath39 where @xmath40 is the strength of a node with degree @xmath8 . in this paper , for simplicity , we focus on uncorrelated ( also called non - assortative mixing ) networks where the conditional probability satisfies @xmath41 @xcite .
thus , one can obtain @xmath42 . here
, for each node with degree @xmath8 we fixed a total transmission rate which is given by @xmath43 , and the transmission rate on the edge from the @xmath8-degree node to @xmath24-degree node , will be redistributed by the proportion of the k - degree node s strength that the edge s weight accounts for , that s to say the @xmath44 can be defined as follows : @xmath45 from which we know the more proportion of @xmath40 that the weight @xmath38 of an edge accounts for , the more possible the disease will transmit through the edge . in the uncorrelated case , one can obtain @xmath46 . moreover ,
the reasonable total probability that a susceptible node with degree @xmath8 will be infected at time step @xmath15 is given by @xmath47 , where @xmath48 denotes the degree sequence of neighboring infected nodes that connect to the susceptible node with degree @xmath8 at time step @xmath15 . on the other hand , from the general differential equations of sir model ( equations ( [ eq4 ] ) - ( [ eq6 ] ) )
, we know @xmath49 denotes the infectivity of nodes with degree @xmath8 , and here , in the present model we define it as follows : @xmath50 which is to say , each infected individual can establish contacts with its @xmath51 neighbors within one time step .
the exponent @xmath0 will dominate the infectivity among nodes with different degrees . since @xmath52
, it can be adjusted to make the contacts fall on a more realistic range . and
the node s infectivity will grow nonlinearly with the increasing degree @xmath8 .
we take the simplified expressions of @xmath49 and @xmath44 into the equations ( [ eq4 ] ) - ( [ eq6 ] ) with @xmath19 ( without lack of generality ) , we obtain as follows : @xmath53 @xmath54 @xmath55 where @xmath56 .
the above equations combined with the initial conditions @xmath57 , and @xmath58 . and in the general case @xmath59 is very small , then we can obtain @xmath60 . under this approximation ,
equation ( [ eq9 ] ) can be directly integrated , as follows : @xmath61 where @xmath62 , and in the last equality we have made use of equation ( [ eq11 ] ) . in order to obtain some material results for the epidemic threshold and the average epidemic prevalence , firstly we compute the time derivative of the magnitude @xmath63 @xcite : @xmath64\nonumber\\ & = & \langle k^{\alpha}\rangle-\phi(t)-\sum_{k}k^{\alpha}p(k)s_{k}\nonumber\\ & = & \langle k^{\alpha}\rangle-\phi(t)-\sum_{k}k^{\alpha}p(k)e^{-\frac{\lambda k^{1+\beta}}{\langle k^{1+\beta}\rangle}\phi(t)}.\end{aligned}\ ] ] for the general @xmath65 distribution , equation ( [ eq13 ] ) can not be solved in a closed form .
however , we can still obtain some useful results in the steady state of the epidemics . since in the steady stage with sufficiently large @xmath15 , we have that @xmath66 and consequently @xmath67 d@xmath68/dt=0 , then one can get the self - consistent equation for @xmath63 from equation ( [ eq13 ] ) as follows : @xmath69 the value @xmath63=0 is always a ( trivial ) solution
. then we compute the second order derivative of the rhs of equation ( [ eq14 ] ) for @xmath63 , and note that @xmath70 we can see the rhs of equation ( [ eq14 ] ) is a convex function , therefore , a nontrivial solution of equation ( [ eq14 ] ) exists only if the condition @xmath71 can be satisfied .
this relation implies @xmath72 and the above inequation defines the epidemic threshold : @xmath73 below which the average epidemic prevalence ( @xmath74 ) will finally be approximatively null , and above which it will attain a finite value .
one can see if @xmath75 , then @xmath76 , which induces the absence of the epidemic threshold in a wide range of scale - free networks @xcite . and
if @xmath77 , the threshold will be a finite value given by @xmath78 ; similarly , if @xmath79 , one can also get a finite threshold which is @xmath80 .
furthermore , we consider the epidemic threshold in the case of general scale - free networks of which the degree distribution is @xmath81 , where @xmath82 is the normalization constant .
then , we obtain @xmath83 and @xmath84 , where @xmath85 @xmath86 denotes the largest ( smallest ) degree in the underlying networks . substituting into equation ( [ eq18 ] )
, one can rehandle the epidemic threshold as follows : @xmath87 from equation ( [ eq19 ] ) , one can see that the infinite of the largest degree ( @xmath88 or equally @xmath89 , since @xmath90 @xcite ) will make the epidemic threshold @xmath91 tends towards zero if @xmath92 ; on the other hand , if @xmath93 , the epidemic threshold @xmath91 is approximate to be a finite value , given by @xmath94 .
thus , the critical border is @xmath95 .
although for most real networks including the internet @xcite , the www @xcite , the world - wide airport networks @xcite and the scientific collaborations networks @xcite , the topology exponent @xmath96 exists between 2 and 3 , which is incidental to induce the absence of the epidemic threshold , one can adjust the infectivity exponent @xmath0 and the weight exponent @xmath1 to restore a nonzero threshold for a given networks ( a fixed value of @xmath96 ) . (
a ) : the epidemic threshold @xmath91 versus @xmath0 with the exponent @xmath97 in ba networks .
( b ) : the epidemic threshold @xmath91 versus @xmath1 with the exponent @xmath98 in ba networks.,title="fig:",scaledwidth=37.6% ] ( a ) : the epidemic threshold @xmath91 versus @xmath0 with the exponent @xmath97 in ba networks .
( b ) : the epidemic threshold @xmath91 versus @xmath1 with the exponent @xmath98 in ba networks.,title="fig:",scaledwidth=38.0% ] the epidemic threshold @xmath91 in a 3d - graph , which is made up with various @xmath0 and @xmath1 in ba networks.,scaledwidth=61.0% ] in order to get a intuitionistic relation with @xmath91 , @xmath0 and @xmath1 , we have performed numerical simulations for the epidemic threshold .
firstly , on the basis of the simulated stochastic realizations of sir model on ba networks @xcite of which the theoretical scale - free exponent is 3 ( @xmath99 ) , we fix @xmath97 and adjust @xmath0 between 0 and 1 to show the transformation of epidemic threshold @xmath91 . in this case
the critical value of @xmath0 is : @xmath100 , below which @xmath91 is a nonzero finite value . as figure [ threshold12](a )
displays , the value of @xmath91 is greater than 0.01 with the relation @xmath101 , vice versa .
secondly , we fix @xmath98 and adjust @xmath1 between -2 and 2 to show the transformation of @xmath91 . in this case
the critical value of @xmath1 is : @xmath102 , below which @xmath91 is a nonzero finite value . as figure [ threshold12](b )
displays , at the point of @xmath103 , @xmath104 , and @xmath91 will take on a much faster change in @xmath105 than the one in @xmath106 , and the finiteness of @xmath91 is apparent when @xmath107 . from figure
[ threshold12 ] , one can see the simulations are consistent with the analytic results about critical threshold when we consider the effect of finite scale of the substrate work we use .
moreover , it is observed that the decreasing trend of @xmath91 with increasing @xmath0 is much quicker than the one with increasing @xmath1 , that is to say , @xmath0 is more sensitive than @xmath1 in the transformation of epidemic threshold @xmath91 , which means @xmath0 is the leading factor for the transformation of @xmath91 in the present model .
figure [ threshold3 ] displays the epidemic threshold in a 3d - graph , which is made up with various @xmath0 and @xmath1 , and the critical condition for a nonzero finite threshold is @xmath108 .
one can see that @xmath91 is small in the blue area where the great mass of data meet the condition that @xmath109 , which is the condition of threshold vanishing .
the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath97 , and @xmath0=1.0 , 0.9 , 0.8 , @xmath112 , 0.2 , 0.1 ( from top to bottom).,scaledwidth=57.0% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath98 , and ( a ) : @xmath1=0 , 0.5 , 1.0 , 1.5 , 2.0 ( from top to bottom ) ; ( b ) : @xmath1=-0.5 , -1.0 , -1.5 , -2.0 ( from top to bottom).,title="fig:",scaledwidth=57.0% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath98 , and ( a ) : @xmath1=0 , 0.5 , 1.0 , 1.5 , 2.0 ( from top to bottom ) ; ( b ) : @xmath1=-0.5 , -1.0 , -1.5 , -2.0 ( from top to bottom).,title="fig:",scaledwidth=57.0% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath79 , and from top to bottom : @xmath0=0.9 , 0.7 , 0.5 , 0.3 , 0.1 ; accordingly @xmath1=-1.9 , -1.7 , -1.5 , -1.3 , -1.1 .
the inset shows the susceptible density @xmath113 versus @xmath3 with the same combination of @xmath0 and @xmath1 .
, scaledwidth=57.0% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath77 , and from top to bottom : @xmath0=0.8 , 0.6 , 0.4 , 0.2 ; accordingly @xmath1=-0.8 , -0.6 , -0.4 , -0.2 .
the inset shows the susceptible density @xmath113 versus @xmath3 with the same combination of @xmath0 and @xmath1 . ,
scaledwidth=57.0% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath114 , and from top to bottom : @xmath0=0.7 , 0.6 , 0.5 , 0.4 , 0.3 ; accordingly @xmath1=0.3 , 0.4 , 0.5 , 0.6 , 0.7 .
the inset shows the susceptible density @xmath113 versus @xmath3 with the same combination of @xmath0 and @xmath1 .
, scaledwidth=57.5% ] the steady epidemic prevalence @xmath2 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath115 , and from top to bottom : @xmath0=0.9 , 0.8 , 0.7 , 0.5 , 0.3 , 0.1 ; accordingly @xmath1=1.1 , 1.2 , 1.3 , 1.5 , 1.7 , 1.9 .
the inset shows the susceptible density @xmath113 versus @xmath3 with the same combination of @xmath0 and @xmath1 .
, scaledwidth=58.0% ] for further investigation of the epidemic dynamics of the present model , we study the propagation behavior of the epidemic spreading .
firstly , we investigate the average epidemic prevalence in the steady stage of epidemic evolution with different combinations of @xmath0 and @xmath1 . from the analysis in section 3 ,
it is easily to conclude that @xmath116 at the epidemic critical point of @xmath3 , which will induce a quite small @xmath33 , and we can approximately get @xmath117 according to the relationship @xmath118 . then , expanding the rhs of equation ( [ eq14 ] ) for the small @xmath63 , and ignoring the higher - order terms , we obtain @xmath119 next , we compute the derivative of equation ( [ eq20 ] ) for @xmath3 at the critical point , as follows : @xmath120 as referred above , one can obtain that @xmath121 consequently , @xmath122 similarly , in the considering of general scale - free networks as referred above , equation ( [ eq23 ] ) can be written as follows : @xmath123 the obtained results shows , the exponent @xmath0 will make a primary contribution to the velocity of increasing of the steady epidemic prevalence ( @xmath2 ) by given the topology of a underlying network . combining the analysis in section 3 , for a fixed sum of @xmath0 and @xmath1
, one can conclude that the more ratio of @xmath0 to @xmath1 is , the larger the @xmath91 is , and the slowly the @xmath2 grows as @xmath3 increases ( since the most large - scale real networks have the relationship @xmath124 ) . for a better understanding of the epidemic propagation behavior , we take numerical simulations with various combination of @xmath0 and @xmath1 on ba networks ( @xmath99 ) .
firstly , we investigate the impact of @xmath0 and @xmath1 separately , which are the two particular cases : @xmath98 with different @xmath1 , and @xmath97 with different @xmath0 .
figure [ rl1 ] displays the effects of @xmath0 on the steady epidemic prevalence @xmath2 with @xmath97 .
as the figure shows , for a same @xmath3 , one can observe the slope of @xmath2 grows as @xmath0 increases ( see @xmath0=0.9 , 0.8 , @xmath112 , 0.2 , 0.1 ) , which is consistent with the analytical results from equation ( [ eq24 ] ) .
figure [ rl23 ] displays @xmath2 versus @xmath3 in the case of @xmath98 , and figure [ rl23](a ) : @xmath1=0 , 0.5 , 1.0 , 1.5 , 2.5 ( from top to bottom ) ; figure [ rl23](b ) : @xmath1=-0.5 , -1.0 , -1.5 , -2.5 ( from top to bottom ) . as shown in figure [ rl23
] , the larger the absolute value of @xmath1 is , the more slowly the @xmath2 grows . on the other hand , for a general case ( @xmath125 and @xmath126 ) , according to the critical equation @xmath127 and the algebraic sign of the sum @xmath128 , we consider four representative combinations of @xmath0 and @xmath1 , which are @xmath79 , @xmath77 , @xmath114 and @xmath115 . in each combination ,
there also has been divided into several different configurations by the ratio of @xmath0 to @xmath1 . as shown in figures ( [ rl4 ] - [ rl7 ] ) , for a fixed sum of @xmath0 and @xmath1
, one can see the @xmath2 grows more quickly as the ratio increasing , which is consistent with our analytical results that @xmath0 is a more sensitive factor to @xmath2 . and
further more , since @xmath109 , one can see the epidemic threshold @xmath91 is quite small ( tends to zero , considering the effects of finite size ) in figure [ rl23](a ) and figure [ rl7 ] , which is also consistent with the critical condition of @xmath91 ; otherwise , for @xmath108 , @xmath91 becomes to be a nonzero finite value as shown in figure [ rl1 ] , figure [ rl23](b ) , figure [ rl4 ] and figure [ rl5 ] . for the early stage of @xmath129 in figures ( [ rl1 ] - [ rl7 ] ) , one can see the @xmath2 grows in an exponential form with @xmath3 increases , then the growth rate will take off slowly for @xmath130 , at last , it will tend to be zero , which means the value of @xmath2 have attained a steady value . the temporal epidemic prevalence @xmath74 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath79 , and from top to bottom : @xmath0=0.9 , 0.8 , 0.7 , @xmath112 , 0.2 , 0.1 ; accordingly @xmath1=-1.9 , -1.8 , -1.7 , @xmath112 , -1.2 , -1.1.,scaledwidth=56.0% ] the temporal epidemic prevalence @xmath74 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath77 , and from top to bottom : @xmath0=0.8 , 0.7 , @xmath112 , 0.2 , 0.1 ; accordingly @xmath1=-0.8 , -0.7 , @xmath112 , -0.2 , -0.1.,scaledwidth=56.0% ] the temporal epidemic prevalence @xmath74 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath114 , and from top to bottom : @xmath0=0.9 , 0.8 , 0.7 , @xmath112 , 0.2 , 0.1 ; accordingly @xmath1=0.1 , 0.2 , 0.3 , @xmath112 , 0.8 , 0.9.,scaledwidth=56.0% ] the temporal epidemic prevalence @xmath74 versus @xmath3 for sir model in ba networks with @xmath110 , @xmath111 , @xmath115 , and from top to bottom : @xmath0=0.9 , 0.8 , 0.7 , @xmath112 , 0.2 , 0.1 ; accordingly @xmath1=1.1 , 1.2 , 1.3 , @xmath112 , 1.8 , 1.9.,scaledwidth=56.0% ] for investigating the temporal propagation behavior , we simulate the time behavior of @xmath74 for sir model on ba networks with @xmath131 .
as displayed in figure [ rt1 ] ( @xmath79 ) , figure [ rt2 ] ( @xmath77 ) , figure [ rt3 ] ( @xmath114 ) and figure [ rt4 ] ( @xmath115 ) , one can see that , the prevalence @xmath74 grows in an exponential form in the early stage , and then stabilizes in a nonzero value as time goes on .
since @xmath3 may be smaller than @xmath91 when the value of @xmath0 is much small such as @xmath132 , thus if @xmath133 , the steady value of @xmath74 ( i.e. , @xmath2 ) will be quite small , which approximatively equals to the initial density of infected nodes ; when @xmath134 , the prevalence @xmath74 is higher as the ratio of @xmath0 to @xmath1 gets larger at the same time of @xmath15 , as the figures display . moreover , it is observed that the steady value of @xmath74 is smaller in figure [ rt4 ] compared with the other figures ( [ rt1 ] - [ rt3 ] ) .
although the exact solution of @xmath74 about @xmath0 and @xmath1 which can demonstrate the difference well is difficult to be managed here , from a qualitative perspective , we believe that s because of the trait of sir model . in figure
[ rt4 ] , @xmath115 which will induce a small threshold @xmath91 , thus at the early stage of evolution many susceptible nodes will be infected in view of the relation of @xmath130 , and due to the immune rate we set is unity ( @xmath19 ) , the old infected nodes will become removed ones that cantt be infected any more at the same time .
consequently the optional objects for a infected node will decrease as time evolutes , moreover , there will be many infected nodes surrounded by the removed ones . under this situation , the epidemic spreading will arrive at a equilibrium much more quickly , and thus the epidemic prevalence in the steady stage also will be a small value as figure [ rt4 ] displays .
to sum up , in this paper , we have investigated the dynamical behavior of sir model with weighted transmission rate and nonlinear infectivity , we present that one can adjust the exponent @xmath0 and @xmath1 to control the epidemic threshold which is absent for the standard sir model in scale - free networks .
the critical value just depends on the exponent @xmath0 and @xmath1 for a given topology of networks ( a fixed value of @xmath96 ) , and @xmath0 is more sensitive than @xmath1 for the transformation of the epidemic threshold and epidemic prevalence , which agrees with the numerical simulations very well . and the numerical results of the time behavior of @xmath74 also have been presented , where the remarkable result is that , for a fixed @xmath3 , the smaller threshold @xmath91 will induce a smaller epidemic prevalence at the equilibrium . in a way , epidemic spreading can be managed as a reaction - diffusion process @xcite , which also has a very close relation with information retrieval , peer - trust and influence spreading .
the efficient diffusion or inefficient diffusion maybe has its merits in various natural and artificial networks .
our work might deliver some useful information or new insights in designing data layout , city layout and network layout for performing their best advantages .
we benefited from useful discussions with yichao zhang , ming tang .
this research was supported by the national basic research program of china under grant no .
2007cb310806 , the national natural science foundation of china under grant nos .
60704044 , 60873040 and 60873070 , shanghai leading academic discipline project no .
b114 , and the program for new century excellent talents in university of china ( ncet-06 - 0376 ) .
ref1 kermack w o and mckendrick a g 1927 _ proc .
r. soc . london .
ser . _ a * 115 * 700 | in this paper , we investigate the epidemic spreading for sir model in weighted scale - free networks with nonlinear infectivity , where the transmission rate in our analytical model is weighted .
concretely , we introduce the infectivity exponent @xmath0 and the weight exponent @xmath1 into the analytical sir model , then examine the combination effects of @xmath0 and @xmath1 on the epidemic threshold and phase transition .
we show that one can adjust the values of @xmath0 and @xmath1 to rebuild the epidemic threshold to a finite value , and it is observed that the steady epidemic prevalence @xmath2 grows in an exponential form in the early stage , then follows hierarchical dynamics .
furthermore , we find @xmath0 is more sensitive than @xmath1 in the transformation of the epidemic threshold and epidemic prevalence , which might deliver some useful information or new insights in the epidemic spreading and the correlative immunization schemes . |
the sloan digital sky survey @xcite galaxy redshift survey has an unprecedented combination of large area , depth , and photometric quality , thanks to the combination of a large format camera @xcite , high throughput multi - object spectrographs ( a. uomoto et al .
, in preparation ) , careful calibration procedures @xcite , and an efficient series of data reduction and targeting pipelines @xcite .
recently @xcite ( @xcite , hereafter b04 ) have searched the sdss second data release ( dr2 ; @xcite ) to identify a population of extremely low luminosity field galaxies . with absolute magnitudes @xmath12 to @xmath13 ( for the hubble parameter @xmath4 adopted throughout this paper ) , this population represents a range of luminosities that has previously been accessible to systematic study only in the local group and in nearby groups and clusters .
we have obtained h@xmath0 rotation curves of two of these objects , using the same methods that we are using for a study of the tully - fisher ( @xcite , hereafter tf ) relation of more luminous sdss galaxies ( j. pizagno et al .
, in preparation , hereafter p04 ) .
both objects appear to be `` field '' galaxies rather than satellites of brighter systems .
although the sample is small and the data quality limited , these measurements provide one of the first insights into the dynamical properties of this new population .
the galaxies in this paper were selected from the b04 sample according to their availability during an observing run for the p04 project . as detailed in b04 and @xcite , this low luminosity galaxy sample has been checked for contamination by double stars , errors in the photometric pipeline s automated deblending , and other complications that are unimportant for the great majority of sdss galaxies but have a significant impact for low luminosity and low redshift systems .
the two galaxies selected for follow - up observations were j123654.9 + 013654.2 and j091858.60 + 581407.7 . for brevity , we will hereafter refer to these as galaxy 1 and galaxy 2 , respectively .
distances for galaxies in the b04 sample are estimated from their redshifts assuming hubble flow and a model for the local peculiar field based on the iras 1.2-jy redshift survey @xcite .
details of this procedure are described by @xcite . because these galaxies have low redshifts ( heliocentric values of @xmath14 and @xmath15 , respectively ) , peculiar velocity uncertainties dominate the distance uncertainties , which in turn dominate the uncertainties in luminosity and dynamically inferred mass - to - light ratios in our discussions below .
the @xcite procedure yields distances of @xmath16 and @xmath17 to galaxies 1 and 2 , for @xmath18 .
the @xmath19 error bars are derived from the peculiar velocity probability function , which includes a @xmath20 local dispersion around the smooth velocity field predicted from the iras 1.2-jy galaxies .
the corresponding petrosian @xmath1-band absolute magnitudes are @xmath2 and @xmath3 ( again for @xmath18 ) .
neither galaxy has a bright neighbor ( @xmath21 ) within 0.7 mpc projected separation and @xmath22 redshift separation .
galaxy 1 is close to ngc 4536 in projection , but the redshift difference , @xmath14 vs. @xmath23 @xcite , makes a physical association unlikely . we conclude that both of these objects are `` field '' galaxies rather than satellites of brighter parents
. table 1 lists the magnitudes , redshifts , and distances for galaxies 1 and 2 along with the @xmath24 colors and axis ratios , exponential disk scale lengths , and rotation speed parameters described below .
the top panels of figure 1 show @xmath25-band images from the sdss .
a search through the nasa extragalactic database ( ned ) shows an hi observation ( from @xcite ) for the galaxy fgc 1475 , which can be identified as our galaxy 1 on the basis of angular proximity ( @xmath26 difference between the ned and sdss positions ) , similar axis ratios , and similarity between the @xcite and sdss redshifts .
spectroscopic observations were made with the ccds long - slit spectrograph , set up to observe redshifted h@xmath0 , at the 2.4 meter mdm telescope during the night of 13 april 2004 .
we used a @xmath27 slit width with a 600 lines / mm grating in the second order producing a 0.41 / pixel dispersion , and a @xmath28 spatial scale .
the seeing during the night was @xmath29 .
the h@xmath0 emission lines typically had a total signal - to - noise ratio of @xmath30 , with an intensity weighted velocity centroid uncertainty of @xmath31 8 km / sec .
the dispersion axis was aligned to be perpendicular to the columns of the ccd to an accuracy of 0.1 , as judged from the telluric lines .
this alignment ensures that the h@xmath0 recession velocity is measured to the accuracy of the intensity weighted velocity centroids .
the details of the data reduction and rotation curve extraction can be found in p04 .
the bottom panels of figure 1 show the linearized and flat - fielded spectra of both galaxies .
figure 2 presents the extracted rotation curve measurements as observed velocity relative to the continuum center vs. position along the slit in arc - seconds . following @xcite and p04
, we fit the rotation curve measurements with an arc - tangent function , which has a minimal number of free parameters while still describing the global shape of typical galaxy rotation curves quite well .
specifically , we use a levenburg - marquardt @xmath32 minimization routine @xcite to fit the data with the functional form @xmath33 where @xmath34 is the central velocity , @xmath35 is the asymptotic circular velocity , @xmath1 is the position along the slit , @xmath36 is the center of the rotation curve ( where @xmath37 ) , and @xmath38 is the turnover radius at which the rotation curve begins to flatten .
the parameter uncertainties are derived from the covariance matrix returned by the levenburg - marquardt routine . for galaxy 1 , we allow all four parameters ( @xmath34 , @xmath35 , @xmath36 , @xmath38 ) to vary . for galaxy 2 ,
the best - fit arc - tangent parameters have very large @xmath35 and @xmath38 , so that they effectively describe a straight line over the region covered by the data points .
we therefore fit a straight line to the rotation curve instead of an arc - tangent function , simplifying the determination of best - fit parameters and uncertainties .
since galaxy 2 s rotation curve is more extended on one side , we fix the line s intercept such that the line goes through the location of the conintuum center at the center of mass velocity .
figure 2 shows the arc - tangent and linear fits to the data points as smooth solid curves . for galaxy 1 ,
the outermost data points probe the turnover region of the rotation curve , while for galaxy 2 they are still linearly rising .
clearly these data do not yield good constraints on the asymptotic circular velocity of a flat rotation curve .
however , we find in p04 that the velocity @xmath39 defined by the value of the arc - tangent fit at the radius of the outermost data point often provides a useful measurement of rotation speed even when the observed rotation curve is still rising at this radius .
in particular , we find that if we use @xmath39 as the measure of rotation speed in the tf relation , then the observed tf scatter does not increase significantly when we include galaxies that have rising rotation curves in the sample . while the value of @xmath39 is a lower limit to the asymptotic circular velocity
, it can be much more robustly measured than @xmath35 from truncated rotation curves like those in figure 2 .
we correct @xmath39 for inclination by using the galfit program @xcite to fit inclined exponential disks to the @xmath25-band galaxy images shown in figure 1 .
the disk axis ratios and exponential scale lengths are listed in table 1 ; formal uncertainties are @xmath40 in @xmath41 and @xmath42 in @xmath43 , though the systematic errors associated with assuming an exponential disk model are probably larger .
the inclination corrected velocity is @xmath44 where 0.19 is the assumed intrinsic axis ratio for an edge - on disk and @xmath41 is the measured @xmath25-band axis ratio .
different ranges for the intrinsic axis ratio vary by 0.10 to 0.20 depending on galaxy type @xcite .
we chose 0.19 , typical for spiral galaxies , and note that the range in intrinsic axis ratios causes a small variation ( typically @xmath311@xmath45 ) in @xmath46 .
galfit yields a position angle for galaxy 2 that differs from the sdss dr2 value by ten degrees , consistent with the visual impression of slight misalignment in figure 1 .
we therefore apply a small ( @xmath47 ) correction to the measured @xmath39 assuming a tilted - ring model as described by @xcite ( @xcite ; see their eq .
table 1 lists inclination - corrected values @xmath46 including this slit misalignment correction for galaxy 2 .
@xcite report an hi velocity width for galaxy 1 of @xmath48 , where the subscript @xmath49 indicates a correction for instrumental resolution .
they do not give an error bar on @xmath50 , though the total signal - to - noise ratio of the hi line is 4.8 , so the uncertainty in the width is probably not negligible .
@xcite compare hi linewidths to optically measured , maximum rotation speeds and report a correlation @xmath51 in @xmath45 units .
equation ( 3 ) predicts @xmath52 , compared to our measured value of @xmath53 .
( note that the inclination correction is negligible for this galaxy . )
the two measurements thus appear to be consistent within the observational errors ( allowing for a few percent uncertainty in the measured @xmath50 ) , though the somewhat higher value inferred from the hi data could reflect a continuing rise of the rotation curve beyond the radius probed by our h@xmath0 measurements .
how do our measured rotation speeds compare to expectations based on the tf relation defined by more luminous galaxies ? in p04 , we use a sample of 170 galaxies with @xmath54 to measure the forward ( @xmath55 vs. @xmath56 ) and inverse ( @xmath56 vs. @xmath55 ) tf relations in the sdss bands . for the @xmath1-band inverse relation ,
we fit @xmath57 with @xmath58 and the constant @xmath59 chosen to yield minimal correlation between the errors in @xmath60 and @xmath61 .
we find @xmath62 and @xmath63 , yielding predicted values of @xmath46 for galaxies 1 and 2 of @xmath64 and @xmath65 , respectively .
the uncertainties in @xmath66 and in the tf slope both contribute significantly to the errors on @xmath67 , and both contributions are non - linear .
we have computed upper @xmath19 error bars by separately varying @xmath66 by @xmath68 and @xmath60 by @xmath69 and adding the corresponding values of @xmath70 in quadrature ; we follow an analogous procedure for the lower @xmath19 error bar .
the predicted rotation speeds are well below the measured values of @xmath71 and @xmath72 , though marginally compatible in the latter case because of the large observational errors .
dotted curves in figure 2 show rotation curves with the same parameters as the solid curves but scaled by @xmath73 . despite the uncertainties in the measurements , it is clear that the galaxies are rotating substantially faster than predicted by an extrapolation of the inverse tf relation to these faint magnitudes . to match the observed rotation speeds while maintaining the p04 normalization at @xmath74 would require an inverse tf slope of @xmath75 , compared to the slope of @xmath76 measured in the bright galaxy regime .
we can use our measurements to estimate dynamical mass - to - light ratios within the radius of the outermost h@xmath0 data point in each galaxy .
we use the simple mass estimator @xmath77 ignoring possible corrections for asymmetric drift or non - circular motions , which would be small compared to our observational error bars .
the values of @xmath78 are 19 and 11 arc - seconds for galaxies 1 and 2 , respectively , corresponding to 1.7 and 1.8 times the exponential scale lengths measured by galfit . at our estimated distances of @xmath79 mpc and @xmath80 mpc ,
the physical values of @xmath78 are @xmath81 kpc ( galaxy 1 ) and @xmath82 kpc ( galaxy 2 ) .
we measure the fraction of @xmath1-band light within @xmath78 for each galaxy and correct the values of @xmath66 in table 1 accordingly , by 0.24 and 0.14 magnitudes .
we do not apply any internal extinction corrections . adopting @xmath83
, we find @xmath84 for galaxy 1 and @xmath10 for galaxy 2 . because @xmath85 and the fractional uncertainties in these quantities are substantial , we compute the upper @xmath19 error bar by separately varying @xmath86 by @xmath68 and @xmath39 by @xmath69 and adding the two changes @xmath87 in quadrature , and we follow an analogous procedure for the lower @xmath19 error bar .
the @xmath88 value for galaxy 1 is much higher than the value @xmath89 expected for a stellar population with the observed @xmath24 color of the galaxy @xcite , so this system is strongly dominated by dark matter within @xmath78 .
the case for dark matter domination in galaxy 2 is less clear because of the large uncertainty in @xmath88 , though the central value again corresponds to a large ratio of dark matter to stellar mass within @xmath78 .
( the @xcite [ @xcite ] models imply a stellar mass - to - light ratio of @xmath90 for galaxy 2 s @xmath24 color . )
while all disk galaxies become dark matter dominated at sufficiently large radii , these low luminosity systems appear to have high dark matter fractions even within two disk scale lengths .
to put our measurements in the context of previous results , figure 3 plots inclination corrected circular velocities against @xmath91-band absolute magnitude for our galaxies ( filled circles with error bars ) and for galaxies with hi line widths and @xmath92 from @xcite ( @xcite ; open squares ) , @xcite ( @xcite ; open circles ) , and hi line widths from @xcite ( @xcite ; open triangle ) . for our galaxies , we use @xmath46 as the indicator of circular velocity , and we convert the ( ab ) @xmath93-band luminosity to the ( vega ) @xmath91-band luminosity using the galaxy @xmath24 color and @xmath94 $ ] , obtained using the @xmath95-correction code of @xcite . for @xcite
we use the value of @xmath96 listed in their tables , and for @xcite and @xcite we estimate @xmath96 visually from the plotted rotation curves . we have not attempted to put observational error bars on the literature data points because this would require a detailed assessment of the uncertainties in the distance measurements , which use a somewhat different method in each case .
@xcite observe galaxies in the sculptor and centaurus a groups and assign group distances based on a variety of indicators ( see @xcite and references therein ) .
@xcite assign both of their galaxies to the ngc 4696 group on the basis of proximity and adopt a distance based on the brightest stars @xcite .
@xcite use a combination of group assignment and brightest stars to infer the distance of ddo 154 .
the solid line in figure 3 shows the @xmath91-band inverse tf relation extrapolated to the low luminosity regime .
we convert the @xmath93-band relation of p04 to @xmath91-band using @xmath97 , appropriate for a galaxy with @xmath98 .
dotted lines show relations with the slope varied by @xmath99 . for bright galaxies , the intrinsic scatter about
the mean relation is @xmath100dex in @xmath101 ( p04 ) , smaller than the @xmath102dex observational error bar on galaxy 1 .
as noted earlier ( on the basis of @xmath1-band data ) , galaxy 1 is rotating substantially faster than the extrapolated tf relation predicts , while the large error bar on @xmath39 leaves galaxy 2 marginally consistent with the tf extrapolation .
the rotation speed of galaxy 1 appears reasonably in line with that of the other dwarf galaxies , while the rotation speed of galaxy 2 is noticeably low , perhaps because the optical rotation curve is still rising steadily at our outermost data point .
all of these systems are rotating faster than the tf extrapolation predicts ; equivalently , they are fainter than predicted given their rotation speeds .
@xcite show that low luminosity galaxies are fainter than predicted by the ( forward ) tf relation defined by bright galaxies , consistent with the result shown in figure 3 .
however , faint galaxies are often gas rich , and they show that adding the gas masses inferred from hi observations to the stellar masses leads to a linear relation between @xmath103 and @xmath104 over four decades in baryonic mass @xmath105 , extending down to a few @xmath106 . for our galaxy 1 ,
@xcite report an hi flux integral of 1.24 jy@xmath45 , which implies an hi mass to optical luminosity ratio of @xmath107 in @xmath1-band , or roughly equal mass in neutral gas and stars .
this ratio is below the median of @xmath108 found ( in @xmath109-band ) by @xcite , but within the range spanned by their data .
simply doubling the luminosity of galaxy 1 and thus shifting it rightwards by 0.75 mag in figure 3 would not move it onto the p04 inverse tf relation .
however , @xcite find a steep slope for the ( forward ) baryonic mass tf relation , and the baryonic mass of @xmath110 inferred for galaxy 1 at a distance of 9.9 mpc agrees respectably with the value @xmath111 predicted by their fitted relation for @xmath112 , given the substantial error bars on both the predicted and measured values .
the luminosity - velocity relations and mass - to - light ratios of low luminosity galaxies offer clues to the role of supernovae and photoionization in regulating star formation , since these feedback effects are generally expected to be stronger in lower mass halos ( see , e.g. , @xcite ) .
the influence of feedback may be different in galaxies that are central objects of their parent dark matter halos and in galaxies that are satellites in more massive halos semi - analytic models and hydrodynamic simulations predict systematically different properties for these two populations @xcite .
most of the very low luminosity galaxies studied to date have been discovered or investigated because they are members of groups or clusters defined by brighter galaxies .
the two galaxies studied in this paper are not satellites of bright neighbors , but their properties are roughly in line with those of satellite dwarfs . the b04 sample drawn from the sdss
provides the opportunity to study such faint galaxies in detail over the full range of environments in which they appear .
our present investigation represents a start on this broader program .
jp and dw acknowledge support from nsf grants ast-0098584 and ast-0407125 .
we thank eric f. bell for useful comments .
funding for the creation and distribution of the sdss archive has been provided by the alfred p. sloan foundation , the participating institutions , the national aeronautics and space administration , the national science foundation , the u.s .
department of energy , the japanese monbukagakusho , and the max planck society .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium ( arc ) for the participating institutions .
the participating institutions are the university of chicago , fermilab , the institute for advanced study , the japan participation group , the johns hopkins university , korean scientist group , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , university of pittsburgh , princeton university , the united states naval observatory , and the university of washington .
benson , a. j. , lacey , c. g. , baugh , c. m. , cole , s. , & frenk , c. s. 2002 , , 333 , 156 berlind , a. a. , weinberg , d. h. , benson , a. j. , baugh , c. m. , cole , s. , dav , r. , frenk , c. s. , jenkins , a. , katz , n. , & lacey , c. g. 2003 , , 593 , 1 blanton , m. r. , lupton , r. h. , maley , f. m. , young , n. , zehavi , i. , & loveday , j. 2003a , , 125 , 2276 blanton , m. r. , et al .
2003b , , 125 , 2348 blanton , m. r. , et al .
2004a , aj , submitted , astro - ph/0410164 ( b04 ) blanton , m. r. , et al .
2004b , aj , submitted , astro - ph/0410166 carignan , c. & freeman , k. c. 1988 , , 332 , l33 dekel , a. , & silk , j. 1986 , , 303 , 39 fukugita , m. , ichikawa , t. , gunn , j. e. , doi , m. , shimasaku , k. , & schneider , d. p. 1996
, , 111 , 1748 grogin , n. a. , geller , m. j. , & huchra , j. 1998 apjs , 199 , 277 hogg , d. w. , schlegel , d. j. , finkbeiner , d. p. , & gunn , j. e. 2001 , aj , 122 , 2129 huchtmeier , w. k. , karachentsev , i. d. , karachentseva , v. e. , & ehle , m. 2000 , , 141 , 469 kannappan , s. j. , fabricant , d. g. , & franx , m. 2002 , aj , 123 , 2358 kauffmann , g. , white , s. d. m. , & guiderdoni , b. 1993 , , 264 , 201 lupton , r. , gunn , j. e. , ivezi , . ,
knapp , g. r. , kent , s. , & yasuda , n. 2001 , in asp conf .
238 , astronomical data analysis software and systems x , ed .
f. r. harnden , jr .
, f. a. primini , and h. e. payne ( san francisco : astr .
soc . pac . ) , p. 269 , astro - ph/0101420 marchesini , d. , donghia , chincarini , g. , firmani , c. , conconi , p. , molinari , e. , & zacchei , a. 2002 , apj , 575 , 801 pier , j. r. , munn , j. a. , hindsley , r. b. , hennessy , g. s. , kent , s. m. , lupton , r. h. , & ivezi , .
2003 , , 125 , 1559 pildis , r. a. , schombert , j. m. , & eder , j. 1997 , , 481 , 157 press , w. h. , teukolsky , s. a. , vettering , w. t. , & flannery , b. p. 1992
, numerical recipes ( cambridge : cambridge univ . press ) quinn , t. , katz , n. , & efstathiou , g. 1996 , , 278 , l49 schombert , j. m. , pildis , r. a. , & eder , j. 1997 , , 111 , 233 smith , j. a. , et al .
2002 , , 123 , 2121 somerville , r. s. 2002 , , 572 , l23 somerville , r. s. , & primack , j. r. 1999 , , 310 , 1087 stoughton , c. , et al .
2002 , , 123 , 485 strauss , m.a . , et al .
2002 , , 124 , 1810 thoul , a. a. & weinberg , d. h. 1996 , , 465 , 608 tully , r. b. , & fisher , j. r. 1977 , a&a , 54 , 661 willick , j. a. , strauss , m. a. , dekel , a. , & kolatt , t. 1997 , , 486 , 629 york , d. g. et al .
2000 , , 120 , 1579 zheng , z. , berlind , a. a. , weinberg , d. h. , benson , a. j. , baugh , c. m. , cole , s. , dav , r. , frenk , c. s. , katz , n. , & lacey , c. g. 2004 , , submitted , astro - ph/0408564 cccccccccc j123654.9 + 013654.2 & 16.11 & 0.41 & 594 & @xmath79 & @xmath113 & 11.2 & 0.20 & @xmath114 & @xmath115 + j091858.60 + 581407.7 & 16.84 & 0.31 & 1157 & @xmath116 & @xmath117 & 6.2 & 0.40 & @xmath118 & @xmath119 | we present h@xmath0 rotation curves of two low luminosity field galaxies with @xmath1-band absolute magnitudes @xmath2 and @xmath3 ( for @xmath4 ; the large error bars reflect distance uncertainties ) .
most previously studied galaxies in this luminosity range are members of groups defined by brighter galaxies , but these two systems , selected from blanton et al.s ( 2004 ) sample of low luminosity galaxies in the sloan digital sky survey ( sdss ) , appear to have no bright companions .
the measured rotation speeds at the outer extent of the h@xmath0 rotation curves , @xmath5 and @xmath6 , are larger than the values of @xmath7 and @xmath8 predicted by extrapolating the inverse tully - fisher relation of luminous sdss galaxies to these faint luminosities .
however , a previous hi measurement of the first galaxy shows that it has a gas mass similar to its stellar mass , and the total baryonic mass is consistent with that predicted by mcgaugh et al.s ( 2000 ) `` baryonic tully - fisher relation . ''
we find @xmath1-band dynamical mass - to - light ratios within the radii of the last h@xmath0 data points ( about 1.8 disk scale lengths in each case ) of @xmath9 and @xmath10 , much higher than the values @xmath11 expected for the stellar populations .
the dynamical properties of these galaxies , including the rotation speeds and evidence for high gas fractions and dark matter domination within the luminous extent of the galaxy , are consistent with those of previously studied faint galaxies in nearby groups .
further studies of the sdss sample will allow characterization of low luminosity galaxies over the full range of environments in which they reside . |
the superconducting state in the most of the low temperature superconductors is being induced by the electron - phonon interaction @xcite , @xcite .
the coupling between the electron and phonon system can be modeled with a use of the frhlich hamiltonian @xcite .
let us notice that using a canonical transformation , during the elimination of the phonon degrees of freedom in frhlich operator , it is possible to obtain the hamiltonian of the bcs theory @xcite , @xcite .
it should be clearly marked that the bcs model is able to accurately describe the thermodynamic properties of the low temperature superconductors in the limit of the weak coupling , i.e. for @xmath7 , where @xmath8 denotes the electron - phonon coupling constant . in order to precisely estimate the thermodynamic parameters of the superconductors that have a greater value of @xmath8
, the approach proposed by eliashberg should be used @xcite . in the eliashberg scheme
the analysis of the superconductivity issue is being started directly from the frhlich hamiltonian , which is initially written in the nambu notation @xcite .
next , with a use of the matrix matsubara functions the dyson equations are determined ; the self - energy of the system is being calculated with an accuracy to the second order in the equations of motion @xcite , @xcite , @xcite . in the last step , the eliashberg set is determined in a self - consistent way .
the set of the eliashberg equations generalizes the fundamental equation of the bcs model . in particular
, one can take into consideration the complicated form of the electron - phonon interaction with a use of the eliashberg function .
moreover , the application of the full version of the self - consistent method enables to estimate the electron band effective mass in the presence of the electron - phonon interaction and the energy shift function that renormalizes the electron band energy @xcite . from the mathematical point of view
the solution of the eliashberg set is a truly complicated matter .
for that reason , the effect of the electron band energy renormalization is usually omitted what results in the reduction of the equation s number by one third .
the approximation given above , in the most cases , does not affect the final results in a significant manner .
this is evidenced by the good agreement between theoretical predictions and the experimental data @xcite .
it need to be marked that the strict solution of the simplified eliashberg set is not easy and can be performed only with a use of a powerful computer and highly advanced numerical methods @xcite .
the electron - phonon coupling constant for @xmath0 superconductor is equal to @xmath9 @xcite . in this case , the exact estimation of the thermodynamic parameters is possible only in the framework of the eliashberg approach . in the literature related to the topic
there are some reports that the simplest version of the eliashberg equations might not be able to cope with the description of all relevant physical quantities .
in particular , the dependence of the upper critical field ( @xmath10 ) on the temperature @xcite , @xcite or the results obtained with a use of the directional point - contact spectroscopy @xcite , @xcite suggest the necessity of the two - band model being used . on the other hand , some researchers basing on the thermal and spectroscopic experiments are moving toward the direction of the one - band models with a non - trivial wave symmetry ( @xmath11 or even @xmath12-wave symmetry ) @xcite-@xcite .
the issue is clouding by the fact that the calculations conducted till the present day for the one - band eliashberg model were not strict ; the study were based on very simplified form of the eliashberg function @xcite or approximated analytical formulas @xcite .
for this reson , in the presented paper , the most important thermodynamic parameters for @xmath0 superconductor were analyzed exactly in the framework of the one - band eliashberg model .
the calculations were based on the effective eliashberg function : @xmath13_{\rm{eff}}=1.283 \left[\alpha^{2}f\left(\omega\right)\right]_{\rm{tr}}$ ] , where the transport function ( @xmath13_{\rm{tr}}$ ] ) was determined by gonnelli _
_ in @xcite ( for details see also @xcite ) .
the eliashberg equations on the imaginary axis can be written in the following form @xcite : @xmath14 @xmath15 where @xmath16 represents the order parameter and @xmath17 is the wave function renormalization factor ; the @xmath18-th matsubara frequency is denoted by : @xmath19 , where @xmath20 ( @xmath21 is the boltzmann constant ) . in the eliashberg formalism the coupling between the electron and phonon system
is considered with using of the formula : @xmath22_{\rm{eff}}\omega } { \left(\omega_n-\omega_m\right)^2+\omega ^2}.\ ] ] the effective eliashberg function @xmath13_{\rm{eff}}$ ] for @xmath0 superconductor is shown in fig .. on the basis of the presented data , it is easy to see an exceptionally strong coupling that appears between the electron gas and the crystal lattice vibrations of frequency of about @xmath23 and @xmath24 mev . the mentioned effect was confirmed with a use of the point - contact spectroscopy measurements in @xcite .
the value of the maximum phonon frequency ( @xmath25 ) is equal to @xmath26 mev .
[ ht ] the form of the effective eliashberg function determined by using the transport function obtained by gonnelli _
et al . _ @xcite.,title="fig : " ] from the mathematical point of view , the strict analysis of the eliashberg equations on the imaginary axis , however complicated , is much simpler than on the real axis . in the imaginary domain
the arguments of functions @xmath27 and @xmath28 take discreet values , thus all problems connected with the numerical analysis of the non - linear integral equations are eliminated @xcite .
on the imaginary axis the eliashberg equations can be solved in the following way @xcite : in the first step one need to define functions @xmath29 and @xmath30 for quantities @xmath31 and @xmath28 respectively , where : @xmath32 .
next , the fact that functions @xmath33 and @xmath28 are symmetric is used .
the eliashberg equations take the form : @xmath34 @xmath35 where : @xmath36 and @xmath37 the symbol @xmath38 appearing in eq .
denotes the kronecker delta .
let us notice , that the parameter @xmath39 must be chosen in a such way , that the solutions of the eliashberg equations for the large values of @xmath18 and @xmath40_{\rm{min}}$ ] would take their asymptotic form : @xmath41 and @xmath42 . in the case of @xmath0 it
is enough to assume @xmath43 .
finally , the eliashberg set is solved in an iterative way .
the eliashberg equations were solved for the temperature range from @xmath44_{\rm{min}}=0.2 $ ] mev to @xmath45 mev .
the form of the order parameter for the selected values of the temperature is presented in fig .. it is easy to notice that together with the decreasing of the temperature , the function @xmath27 takes the higher maximum values ( always for @xmath46 ) and it becomes wider .
the dependence of the order parameter on the temperature can be traced , in the most convenient way , by plotting the function @xmath47 ; see inset in fig .. in the considered case we have taken @xmath48 accurate numerical values of the order parameter for @xmath46 .
we notice that the function @xmath47 can be fitted by the simple formula : @xmath49 where @xmath50 mev and @xmath51 .
[ ht ] the order parameter function on the imaginary axis for the selected values of the temperature ; the first @xmath52 values of @xmath27 are presented .
the dependence of the order parameter @xmath53 on the temperature is plotted in the inset.,title="fig : " ] the familiarity with the form of the function @xmath27 for @xmath54 mev enables the calculation of the order parameter value near the temperature of zero kelvin ( @xmath55 ) . in particular it was assumed that @xmath56 . in order to achieve that , the order parameter on the imaginary axis needs
to be analytically continued on the real axis ( @xmath57 ) , deriving the coefficients @xmath58 and @xmath59 in the expression @xcite , @xcite : @xmath60 where @xmath61 .
the plot of the real ( re ) and imaginary ( i m ) part of the function @xmath57 is shown in fig .. in the last step , the parameter @xmath55 should be calculated on the basis of the equation @xcite , @xcite , @xcite : @xmath62 $ ] .
the value @xmath63 mev was obtained .
the determination of the parameter @xmath55 allows to estimate the dimensionless ratio : @xmath64 which , in the bcs theory , is the universal constant of the model and @xmath65_{\rm{bcs}}=3.53 $ ] @xcite . in the case of @xmath0 ,
the greater value of @xmath4 , equal to @xmath66 , was obtained . with reference to the result predicted by the two - band eliashberg model
, the value of @xmath4 estimated in the framework of the one - band model should be interpreted as the resultant quantity .
in particular the two - band eliashberg model , which very well reconstructs the experimental value of @xmath4 , predicts : @xmath67_{l}+0.29\left[r_{1}\right]_{s}=3.78 $ ] , where @xmath65_{l}$ ] and @xmath65_{s}$ ] denote the ratios for the large and the small value of the order parameter respectively @xcite . when comparing the obtained results it is easy to notice , that the one - band model insignificantly overestimates @xmath4 .
[ ht ] the real and imaginary part of the order parameter on the real axis ; @xmath54 mev was assumed.,title="fig : " ] in fig .
the dependence of the wave function renormalization factor on the matsubara frequency for the selected values of the temperature is presented .
on the basis of the achieved results it has been stated , that function @xmath28 takes its maximum always for @xmath46 . in the framework of the eliashberg formalism the value @xmath68 plays very important role , because it determines the ratio : @xmath69 , where @xmath70 denotes the electron band effective mass in the presence of the electron - phonon coupling and @xmath71 is the electron band effective mass in absence of the electron - phonon interaction . basing on the results presented in figure s inset , it has been concluded , that @xmath70 takes its maximum value equal to @xmath72 for @xmath73 .
let us mark the fact , that in the case @xmath73 , the electron band effective mass in the presence of the electron - phonon coupling can be calculated with an use of the formula : @xmath74 on the basis of expression it has been stated , that the calculated value of @xmath70 is identical with the value obtained using the eliashberg equations .
the above result partially confirms the accurateness of the advanced numerical calculations .
[ ht ] the wave function renormalization factor for the selected values of the temperature ; the first @xmath75 values of @xmath28 are presented .
the dependence of the wave function renormalization factor on the temperature for the first matsubara frequency is plotted in the inset.,title="fig : " ] next , we have calculated the following ratios : @xmath76 and @xmath77 where @xmath78 stands for the difference between specific heat of the superconducting and normal state .
the dependence of @xmath79 on the temperature is being determined with a use of an expression : @xmath80 } { d\left(k_{b}t\right)^{2}}.\ ] ] in eq .
the symbol @xmath81 denotes the difference between the free energy of the superconducting and normal state ; @xmath82 is the value of the electron density of states at the fermi level . on the other hand
, the specific heat in the normal state can be calculated using the formula : @xmath83 , where @xmath84 .
finally , the thermodynamic critical field ( @xmath85 ) should be estimated in accordance with the expression : @xmath86}$ ] .
having an open solutions of the eliashberg equations , @xmath81 is determined directly from @xcite : @xmath87 where symbols @xmath88 and @xmath89 denote the wave function renormalization factor for the superconducting state and the normal state respectively .
[ ht ] ( a ) the dependence of the free energy difference between the superconducting and normal state on the temperature .
( b ) the specific heat of the superconducting and normal state as a function of the temperature.,title="fig : " ] the determined form of @xmath90 is presented in fig .
( a ) , while in fig .
( b ) we are shown the dependence of the specific heat on the temperature for the superconducting and normal state .
the characteristic jump of the specific heat that appears at the critical temperature was marked with a vertical line .
we notice that the specific heat of the superconducting state was obtained on the basis of the 350 accurate values of @xmath91 . on the basis of the determined thermodynamic functions ,
the parameters @xmath5 and @xmath6 were calculated . in the case of @xmath0
these coefficients take following values : @xmath92 and @xmath93 .
let us notice , that @xmath5 and @xmath6 are the universal constants of the bcs model and @xmath94_{\rm{bcs}}=1.43 $ ] and @xmath95_{\rm{bcs}}=0.168 $ ] @xcite . when basing on the above results one can clearly see , that for @xmath0 superconductor the value of ratio @xmath5 significantly deviates from the bcs prediction .
the experimental ratios @xmath5 and @xmath6 were determined in @xcite .
the following results were obtained : @xmath94_{\rm{exp}}=1.77 $ ] and @xmath95_{\rm{exp}}=0.160 $ ] .
when comparing our theoretical values of @xmath5 and @xmath6 with the experimental ones , it can be easily noticed that the one - band eliashberg model properly describes the experimental data .
in the framework of the one - band eliashberg model the selected thermodynamic properties of @xmath0 superconductor was calculated . in particular , the fundamental ratios @xmath96 were determined .
it has been stated that , in the case of @xmath4 the one - band eliashberg model predicts a value non - significantly higher than the value determined by the two - band model , which determines @xmath4 with a very high accuracy .
the result above is connected with the simplified description of the superconducting phase in the one - band model in comparison with the two - band model ; we have taken the one effective eliashberg function instead of the four eliashberg functions and the four elements of the coulomb pseudopotential matrix . in the cases of
the two remain ratios - the agreement between predictions of the one - band eliashberg model and the experimental data is very good .
it is worth to notice that , in the opposition to @xmath4 and @xmath6 , the value of @xmath5 significantly deviates from the prediction of the bcs model . in the last part of the summary
let us turn a reader s attention toward the problem of the correct determination of the upper critical field .
when basing on the papers cited in the introduction one can suppose with a large probability , that the exact form of the function @xmath97 can be reproduced only in the framework of the two - band theory .
let us remind that the two - band approach was successfully used by us for description of the thermodynamic properties of the superconducting state inducing in @xmath98 @xcite .
however , in the case of @xmath0 discussed problem is far more complicated because the appropriate eliashberg functions have not been calculated in the branch press ( only electron - phonon coupling constants are being used in the existing two - band approach ) . at present
this issue is being intensively studied by us with a use of the _ ab initio _
approach @xcite .
the authors wish to thank prof .
k. dziliski for the creation of the excellent working conditions and providing the financial support .
some computational resources have been provided by the rsc computing center .
carbotte , rev .
phys . * 62 * 1027 ( 1990 ) .
carbotte , f. marsiglio , in : _ the physics of superconductors _ , edited by k.h .
bennemann , j.b .
ketterson , ( springer , berlin , 2003 ) , vol 1 , p. 223 .
( a ) h. frhlich , phys . rev . * 79 * , 845 ( 1950 ) ; + ( b ) h. frhlich , proc .
a * 223 * , 296 ( 1954 ) .
taylor , o. heinonen , _ quantum approach to condensed matter physics _ , cambridge university press , 2002 .
( a ) j. bardeen , l.n .
cooper , j.r .
schrieffer , phys . rev . *
106 * , 162 ( 1957 ) ; + ( b ) j. bardeen , l.n .
cooper , j.r .
schrieffer , phys . rev . * 108 * , 1175 ( 1957 ) . for discussion of the eliashberg equations [ originally formulated by g.m .
eliashberg , soviet .
jetp * 11 * , 696 ( 1960 ) ] we refer to : p.b .
allen , b. mitrovi , in : _ solid state physics : advances in research and applications _ , edited by h. ehrenreich , f. seitz , d. turnbull , ( academic , new york , 1982 ) , vol 37 , p. 1 .
y. nambu , phys . rev . * 117 * , 648 ( 1960 ) .
( a ) r. szczniak , physica status solidi ( b ) * 244 * , 2538 ( 2007 ) ; + ( b ) r. szczniak , solid state commun . * 144 * , 27 ( 2007 ) ; + ( c ) r. szczniak , m.w .
jarosik , solid state commun . * 149 * , 2053 ( 2009 ) ; + ( d ) r. szczniak , a. barasiski , acta phys . pol .
a * 116 * , 1053 ( 2009 ) . ( a )
r. szczniak , acta phys . pol .
a * 109 * , 179 ( 2006 ) ; + ( b ) r. szczniak , solid state commun . *
138 * , 347 ( 2006 ) .
jarosik , r. szczniak , d. szczniak , acta phys . pol .
a * 118 * , 1031 ( 2010 ) .
shulga , s .- l .
drechsler , g. fuchs , k .- h .
muller , k. winzer , m. heinecke , k. krug , phys .
80 * , 1730 ( 1998 ) .
h. doh , m. sigrist , b.k .
lee , phys . rev .
lett . * 83 * , 5350 ( 1999 ) .
s. muhhopadhyay , g. sheet , p. raychaudhuri , h. takeya , phys .
b * 72 * , 014545 ( 2005 ) .
bashlakov , y.g .
naidyuk , i.k .
yanson , s.g .
wimbush , b. holzapfel , g. fuch , s.l .
drechsler , supercond .
technol . * 18 * , 1094 ( 2005 ) .
k. maki , p. thalmeier , h. won , phys .
rev b * 65 * , 140502(r ) ( 2002 ) .
q. yuan , p. thalmeier , phys .
b * 68 * , 174501 ( 2003 ) .
m. nohara , m. isshiki , h. takagi , r.j .
cava , j. phys .
66 * , 1888 ( 1997 ) .
m. nohara , m. isshiki , f. sakai , h. takagi , j. phys .
. jpn . * 68 * , 1078 ( 1999 ) .
gonnelli , v.a .
stepanov , a. morello , g.a .
ummarino , g. behr , g. graw , s.v .
shulga , s ,-
drechsler , physica c * 341 * , 1957 ( 2000 ) .
h. michor , t. holubar , c. dusek , g. hilscher , phys .
b * 52 * , 16165 ( 1995 ) .
naidyuk , d.l .
bashlakov , i.k .
yanson , g. fuchs , g. behr , d. souptel , s .- l .
drechsler , physica c * 460 * , 103 ( 2007 ) .
k.s.d . beach , r.j .
gooding , f. marsiglio , phys .
b * 61 * , 5147 ( 2000 ) .
vidberg , j.w .
serene , j. low temp
* 29 * , 179 ( 1977 ) .
huang , j .- y .
lin , c.p .
sun , t.k .
lee , j.d .
kim , e.m .
choi , s.i .
lee , h.d .
yang , phys .
b * 73 * , 012502 ( 2006 ) .
j. bardeen , m. stephen , phys . rev . * 136 * , a1485 ( 1964 ) .
( a ) j. nagamatsu , n. nakagawa , t. muranaka , y. zenitani , a. akimitsu , nature * 410 * , 63 ( 2001 ) ; + ( b ) g.a .
ummarino , r.s .
gonnelli , s. massidda , a. bianconi , physica c * 407 * , 121 ( 2004 ) ; + ( c ) r. szczsniak , solid state commun . * 145 * , 137 ( 2008 ) . | the basic thermodynamic parameters of the superconducting state in @xmath0 were calculated in the framework of the one - band eliashberg model . the effective eliashberg function , determined on the basis of the transport function ( r.s .
gonnelli , _ et al .
_ , physica c * 341 * , 1957 ( 2000 ) ) , was used during calculations .
it was shown that the dimensionless ratios are equal to : @xmath1 , @xmath2 and @xmath3 .
the value @xmath4 fairly agrees with the experimental data whereas @xmath5 and @xmath6 agree very well .
# 1([#1 ] ) # 1[#1 ] # 1table[#1 ] |
the daya bay reactor neutrino experiment is designed to measure the neutrino - mixing angle @xmath1 with high precision @xcite . to reach this goal , the experiment utilizes detectors located in one far and two near underground experimental sites at different distances from three pairs of reactors . at each site
multiple detectors filled with 0.1@xmath0 gd - loaded liquid scintillator are used to detect electron anti - neutrinos from the reactors .
each site is shielded by significant overburden to reduce the flux of cosmic - ray muons , and each anti - neutrino detector ( ad ) is submerged in a water pool to suppress ambient gamma - ray and neutron backgrounds . except at the top
, each water pool is segmented into two optically isolated zones that are instrumented with 20-cm - diameter photomultiplier tubes ( pmts ) as independent water cherenkov detectors for detecting muons @xcite . + in each of the two near sites ( eh1 and eh2 ) , the water pool is approximately @xmath2 , and is instrumented with 288 20-cm waterproof pmts . in the far site ( eh3 ) , the water pool is approximately @xmath3 , and is populated with 384 20-cm pmts . of the 960 pmts ,
619 are new hamamatsu r5912 pmts @xcite custom - built as waterproof assemblies for daya bay .
since the pmts in the water pools are used only for tagging incoming muons , we decided to recycle a number of pmts that had been used in the macro ( monopole , astrophysics and cosmic ray observatory ) experiment in italy @xcite .
we succeeded in waterproofing about 400 macro pmts that passed a pressure test in water , and selected 341 of them to instrument the water pools of the daya bay experiment .
+ in section 2 , we discuss the design of the waterproof pmt assemblies , which includes the macro pmt , voltage divider , potting shell , and studies of potting materials . in section 3
we describe the potting procedure in detail , including the pressure test used to verify that the potted assemblies were waterproof .
based on the initial pressure test results , the fabrication procedure was improved and the mass production of all @xmath4400 waterproofed pmts was completed successfully .
the waterproofed assemblies that passed the pressure test and a subsequent electrical test were installed into the water pools . their performance in the water pools during data taking is presented .
section 4 is the summary .
the macro pmts we used are emi ( now electron tubes ) models 9350ka , and d642 kb @xcite .
they have a 20-cm - diameter hemispherical glass window with a blue - green sensitive bialkali photocathode .
each has 11 high - gain sbcs dynodes and 3 becu dynodes of linear - focused design for good linearity and timing . since their structure ,
geometry and light detection features are similar to those of the newly purchased hamamatsu r5912 , the 9350ka and d642 kb are viable candidates for instrumenting the cherenkov detectors of the daya bay experiment . however , in the macro experiment , these pmts were immersed in mineral oil and their bases were not sealed , whereas , at daya bay , they are submerged in ultra - pure water and subject to greater pressures .
thus , it was necessary to encapsulate the electrical components in the bases with epoxy to make them waterproof , and ensure they could withstand sufficient pressure before installing them in the water pools at daya bay . in the macro experiment ,
two cables were utilized to operate a pmt , one carrying negative high - voltage ( hv ) and the other one carrying the signal . however , in the daya bay experiment , a single cable carries both positive hv and signal from the base to a decoupler , which isolates the signals from the applied hv and directs them to the front - end electronics .
thus , we designed and fabricated new bases for the macro pmts .
several operational characteristics of the pmts are sensitive to the base design , such as the working hv at a particular gain , rise and fall times , linearity and collection efficiency .
generally , the specifications of the macro pmts are similar to those of the hamamatsu r5912s ; for example , a maximum gain of 3@xmath5 at a hv of less than 2 @xmath6 , and linearity better than @xmath7 at @xmath8 peak anode - current . + in our design of the circuit as shown in figure [ fig : basediagram ] , the values of the resistors follow the recommendation of the manufacturer of the emi 9350ka . in particular , the resistor between the photocathode and first dynode provides a potential difference of around @xmath9 at our nominal operating voltage .
our tests have shown that similar performance is achieved for the emi d642 kb pmts with the same circuit design . based on the evaluation of a few prototypes
, we found that the combination of a 50-@xmath10 terminating resistor ( r30 ) and no damping resistor ( r31 ) at the output of the anode provided an optimal waveform , i.e. , it minimized both the recovery time after the overshoot and the ratio of the overshoot amplitude to the signal amplitude .
the circuit was carefully laid out to reduce potential electromagnetic interference between components . to ensure that the refurbished macro pmts work in water , their bases must be well sealed . to achieve this goal
, the potting shell has to have minimal contact between the epoxy and water .
our design shown in the left picture of figure [ fig : shelldesign ] also needs to satisfy the constraint that the potted tube geometry and pmt mounts be compatible with those for the purchased hamamatsu waterproof assemblies . to achieve this ,
the profiles of a few macro pmts were scanned and modeled by a coordinate measuring machine ( cmm ) .
the tube length , socket diameter and dome diameter of all 400 macro pmts were then measured .
the variations of the parameters were within tolerance for mounting the potted pmts on support frames for installation .
the drawing on the right of figure [ fig : shelldesign ] is a vertical cross - section of the two - piece potting shell .
one piece is the acrylic base that sits on the cone - shaped glass envelope of the pmt and holds the plastic socket with the connecting pins .
the voltage - divider board plugs onto the socket .
the other piece is the acrylic cap , which joins with the acrylic base .
the top of the acrylic cap has a crossing rib design that strengthens the potting shell .
the side wall of the acrylic cap has an opening that allows the coaxial cable to exit the base .
another opening in the top of the acrylic cap ( labeled as `` filling hole '' in figure [ fig : shelldesign ] ) is for pouring epoxy into the volume between the two acrylic pieces to seal the pmt base . at the narrow end of the acrylic base , there are four layers of protection from inside to outside : epoxy , sealant , mastic tape and sealant . at the acrylic cap , the coaxial cable also has four layers of protection : epoxy , sealing plug , sealing acrylic cap and mastic tape .
the acrylic base and acrylic cap are sealed together with acrylic cement .
the filling hole in the top of the acrylic cap is closed by a smaller acrylic cap and sealed with the same acrylic cement .
thus , whichever side of the potting shell is seen by water , there are multiple layers of protection to stop water from entering the potting shell and reaching the electrical connections .
epoxy is the most important ingredient of the potting materials .
it plays the key role in sealing quality and must be effective
. it must be compatible with the electrical circuitry of the base , the pmt glass and the cable jacket .
the following criteria were considered when choosing the epoxy : + 1 .
suitable for casting , and with low viscosity ; 2 . curing at room temperature or below
good bonding with acrylic , polyethylene ( pe ) and glass ; 4 .
exothermal temperature rise during curing not to exceed 105@xmath11c ( pe limit ) ; 5 .
suitable for protecting electronics against long term submersion in ultra - pure water at 23@xmath12c .
acceptable amount of natural radioactivity . based on the above criteria ,
three candidates were selected for testing : master bond ep42 - 2lv , master bond ep30m3lv and 3 m scotch cast sc2130 .
3 m scotch cast sc2130 was chosen for its low viscosity , lower surface and center curing temperatures , slower hardening and softer `` feel '' ( which is believed to indicate greater elasticity ) , and the absence of temperature spikes in the casting tests . the procedure of the exothermal test consisted of pouring epoxy into a mold and placing a temperature sensor just below the surface at the center ( figure [ fig : epoxyandmastictape ] ) .
the maximum temperature of the 3 m scotch cast sc2130 within 30 minutes was @xmath13c .
the amount of natural radioactivity of the 3 m scotch cast sc2130 was determined to be @xmath14 from the @xmath15u series , @xmath16 for the @xmath17th sequence , and ( @xmath18 for natural potassium .
it is low enough that this source of radioactive background does not impact the performance of the water cherenkov detectors .
mastic tape is another important sealing material .
it bonds the acrylic base to the cone - shaped part of the glass envelope of the pmt .
it is also used to seal the space between the coaxial cable and the opening of the acrylic cap .
3 m scotch seal 2229 was selected for its following merits : 1 .
malleable , durable , tacky mastic ; 2 .
excellent adhesion and sealing to synthetic cable insulations ; 3 . very low cold - flow ; 4 .
used in sealing high - voltage cable splices ; 5 . seals against dust , soil and water .
radioactivity of the scotch seal 2229 mastic tape was determined to be @xmath19 ( @xmath15u series ) , @xmath20 ( @xmath17th sequence ) and 0.142% ( natural k ) . since only a small amount of this tape
is used in each assembly , the radioactive contaminants are again not a concern .
additional sealing materials were also used to mitigate leakage .
one such material is the 3 m marine adhesive sealant fast cure 5200 , which is normally used in marine underwater applications .
the marine sealant was applied inside and outside the acrylic base around the mastic tape , as shown in the right drawing of figure [ fig : shelldesign ] , where it is labeled as `` rtv '' .
+ the cmm results of 14 macro pmts gave a variation in the cylindricality of the blue socket from @xmath21 to @xmath22 . referring to the left drawing of figure [ fig : foamdonut ] , the distance between the centerlines of the socket and the cone - shaped bottom of the pmt varied with an average of @xmath23 ( with the largest difference being @xmath24 ) , which means there is a slight angle between the two centerlines .
given this , we first applied the marine sealant to the socket connector .
then , because of the angle , we added a pe foam donut in the clearance between the acrylic base and socket ( see the right image of figure [ fig : foamdonut ] ) , and sealed around the mastic tape inside the acrylic shell with the marine sealant .
the cable carrying the signal and hv is a 52-m - long 50-@xmath10 coaxial cable ( model number c07947e ) made by judd wire inc .
, with the outer jacket made of high - density polyethylene @xcite .
as displayed in figure [ fig : shelldesign ] there is an opening in the side wall of the acrylic cap for the coaxial cable to penetrate . a steel / plastic sealing plug that is custom - made by the university of wisconsin - madison physical sciences lab ( uw - psl )
was chosen to create the seal between the cable and the opening in the acrylic cap , as shown in figure [ fig : uwplug ] .
the diameter of the sealing plug matches the size of the opening of the acrylic shell .
two o - rings seal the plug - cap interface and two smaller o - rings seal the cable inside the sealing plug . to further protect the base from water entering along the coaxial cable
, the cable jacket was etched to improve its adhesion to the epoxy .
chemical etching and plasma etching were tested , with the latter approach yielding better results . after etching about @xmath25 of the jacket ,
the cable was dipped into the scotch cast sc2130 potting compound , and cured overnight before trimming and soldering it onto the voltage - divider board .
the acrylic shells , epoxy , mastic tape , marine sealant , and cable sealing plug were all compatible with ultra - pure water .
to complete the base sealing of about 400 pmts , a potting laboratory was set up for mass production .
the laboratory was partitioned into specialized areas for cleaning , assembling , potting , and mechanical and electrical acceptance testing .
the potting of all 400 pmts lasted for about six months and was completed in march 2010 . before cleaning
, each macro pmt was tested with a removable base in a dark box .
its noise signals were viewed with an oscilloscope and were required to be consistent with those due to single photo - electrons .
pmts that failed the check were rejected .
in addition , the integrity of the coaxial cable was checked .
components that passed visual inspection were cleaned with detergent ( alconox ) dissolved in 60@xmath11c water in an ultrasonic cleaner for 10 minutes .
the potting components were thoroughly brushed and rinsed in ultra - pure water .
they were then cleaned again with 60@xmath11c ultra - pure water in another ultrasonic cleaner for 10 minutes .
all cleaned acrylic potting components were baked at 60@xmath11c in an oven for about six hours .
once baking was done , visual inspection of the parts was performed again before assembling them onto a cleaned pmt .
the voltage - divider boards were cleaned with ethanol at 40@xmath11c in an ultrasonic cleaner for 3 minutes .
the macro pmts were also carefully cleaned .
the glass surface was gently wiped with fiberless cloths .
the plastic socket usually contained some mineral oil left behind from the macro experiment . to remove the mineral oil , the socket
was submerged in ethanol for 5 to 30 minutes , depending on how much oil was present .
+ prior to the assembly , the acrylic shells were annealed in an oven to release any mechanical stress incurred in the molding process .
assembly of the components was performed inside a laminar - flow fume hood to keep everything clean .
the assembly consisted of several steps .
the first step was to attach the acrylic base to the pmt .
mastic tape cut with a water jet was then applied to the cone - shaped region of the glass envelope at the proper height .
a heat gun was used to soften the mastic tape beforehand to improve the sealing quality .
after the mastic tape was applied , the acrylic base was pressed onto the mastic tape so that the glass , mastic tape and acrylic base adhered to each other tightly .
the marine sealant was then used to fill the space between the acrylic base and the pmt socket .
it was also applied around the outer perimeter , between the acrylic base and pmt glass .
the cure time of the marine sealant was about 24 hours .
a macro pmt with the acrylic base assembled is shown on the left side of figure [ fig : preassembly ] .
the next step was to assemble the pmt base with its hv cable and an acrylic cap .
the hv cable was fed into the acrylic cap through the opening on the side wall .
the etched end of the cable without an shv connector was then soldered onto the voltage - divider board .
the right picture of figure [ fig : preassembly ] shows the result .
the pre - assembled uw - psl sealing plug on the cable was then pushed into the opening on the side wall . when the marine sealant on the acrylic base had cured and the polyethylene donut was placed the voltage - divider board was plugged onto the socket of the pmt .
the acrylic base and acrylic cap were then joined with weldon-3 and weldon-16 acrylic cements , which have a cure time of less than 30 minutes .
about an hour after the acrylic cements were applied the pmt was placed under a fume hood .
the last step of the assembly was to seal the pmt base with epoxy .
the two components of epoxy 3 m scotch cast sc2130 were mixed thoroughly and air bubbles were removed with a vacuum degassing chamber .
the degassed epoxy was then poured into the acrylic shell through the top filling hole in the cap until the voltage - divider circuit board was immersed .
the epoxy normally cured in 40 minutes .
after four hours to a whole night , the filling hole was sealed with a small acrylic cap and acrylic cement . the opening with the uw - psl sealing plug inserted
was then sealed with a flat acrylic cap which also served as a strain relief for the cable .
once the acrylic cement had cured , mastic tape was applied to the region where the cable exited the acrylic cap .
figure [ fig : laminarhood ] shows the completely assembled pmt units .
+ each potted pmt was subject to a pressure test , mimicking the working conditions in the water pools at daya bay .
the basic idea was to partially submerge each potted assembly in de - ionized ( di ) water inside a 47.3-liter pressure tank ( maximum gauge pressure : @xmath26 ) .
the tank was then pressurized to a pre - determined value for a period of time .
after the test was over , the assembly was inspected for water leakage .
prior to the pressure test we visually inspected all the seals of each potted pmt assembly to ensure the sealants were completely dry and free of defects such as small holes .
+ to make sure that no fluid could get into the assembly during the pressure test , the end of the coaxial cable with an shv connector was encapsulated inside a custom - built cable end - sealing tube . the components of the sealing tube are shown in the left picture of figure [ fig : sealingtube ] .
the right picture of figure [ fig : sealingtube ] shows the assembled sealing tube with the cable in place .
+ the potted pmt assembly was then installed in the pressure tank which was partially filled with di water .
the potted portion of the pmt was submerged in water , with the glass bulb remaining in air .
the cable end - sealing tube was placed on a plastic top with an indentation to prevent the tube from rolling .
once the tank was sealed , pressure was applied slowly and monitored with a pressure gauge on the lid .
usually , the pressure test lasted about 8 to 15 hours for each assembly .
a subset of assemblies was tested for around 40 hours .
after the test was complete , the pmt assembly was visually inspected for any leaks or damage .
+ at the beginning of mass production a small number of assemblies failed the pressure test .
one mode of failure was infiltration of water into the cable end - sealing tube due to damage in the fragile jacket of the cable .
for such a case , with the cable immersed in a bath of soda water , the location of the damaged region was determined by looking for an electrically conducting path between the ground braid of the cable and the water . after the damaged area was patched up with mastic tape and covered with heat - shrink tubing and electrical tape , the pressure test was repeated to confirm that the cable was repaired .
in addition , the screening test was revised to include an integrity check of each cable before assembly . +
another mode of failure was fracturing of the glass envelope .
as specified by the manufacturer , the maximum allowable pressure is @xmath27 .
however , a few pmts imploded at a pressure of @xmath28 during the test
. a few other assemblies did not implode , but their glass dome cracked .
for these cracked tubes , we measured the thickness of the glass .
the top portion of the glass bulb was the thickest , reaching @xmath29 , but at about halfway up the bottom hemisphere from the dynode structure , the glass thickness was about @xmath30 or less .
accordingly , cracking usually occurred in the bottom hemisphere below the equator .
this observation motivated us to weigh the bare pmts before and after potting .
the imploded or cracked pmts were found to have a net weight of less than @xmath31 .
+ since the pmt assemblies would be installed at different depths in the water pools , we decided to apply a lower pressure ( p ) for tubes with net mass ( m ) smaller than @xmath31 .
specifically , if m @xmath32 , p @xmath33 ; m @xmath34 , p @xmath35 .
the potted macro assemblies that passed the test at @xmath36 were installed in the upper eighth of the water pools ( static pressure @xmath37 ) , and those that survived a pressure of @xmath28 were installed in the upper half of each pool ( @xmath38 ) .
the lower halves of the water pools were populated exclusively with the new hamamatsu pmts . potted but unweighed macro tubes were temporarily put aside . after potting all the tubes , we found the correlation between the before- and after - potting masses ( left plot of figure [ fig : tubemass1 ] ) .
this correlation allowed us to estimate the net mass of the tubes that we did not weigh before potting , and then apply a proper pressure to test them .
the right plot of figure [ fig : tubemass1 ] shows the distributions of the before - potting net mass of passed and failed assemblies .
the tubes that broke during the test tend to have smaller mass .
+ during mass production , we potted eight tubes per day and performed pressure tests for two to three potted assemblies . by the end of march 2010
, we had produced a total of 386 waterproofed assemblies , of which 52 were assembled before the potting and acceptance testing procedures were revised . of these ,
50 were tested and 36 passed the pressure test , giving a success rate of 72@xmath0 .
the remaining 334 tubes were assembled after the procedures of potting and pressure testing were refined . among them ,
190 were tested with 176 passed . the success rate
improved to about 93@xmath0 .
six of these 176 waterproofed assemblies were sent to rensselaer polytechnic institute for long - term pressure testing . in the end
, we had 352 waterproofed pmt assemblies available for use , including 146 that had not been pressure - tested .
these were later tested in the surface assembly building at daya bay from august through december of 2010 with a success rate greater than 95@xmath0 .
a facility for testing all the pmt assemblies to be used in the daya bay experiment was set up at the dongguan university of technology ( dgut ) , which is about a two - hour drive from daya bay .
the 352 waterproofed macro pmt assemblies were sent to dgut for electrical testing at the end of march 2010 .
( black ) and 2@xmath39(red ) ; top right : peak - to - valley ratio at a gain of 1@xmath39 ; bottom left : after - pulsing ratio at the gain of 1@xmath39 ; bottom right : dark rates at the gain of 1@xmath39(black ) and 2@xmath39(red).,title="fig:",width=226,height=151 ] ( black ) and 2@xmath39(red ) ; top right : peak - to - valley ratio at a gain of 1@xmath39 ; bottom left : after - pulsing ratio at the gain of 1@xmath39 ; bottom right : dark rates at the gain of 1@xmath39(black ) and 2@xmath39(red).,title="fig:",width=226,height=151 ] ( black ) and 2@xmath39(red ) ; top right : peak - to - valley ratio at a gain of 1@xmath39 ; bottom left : after - pulsing ratio at the gain of 1@xmath39 ; bottom right : dark rates at the gain of 1@xmath39(black ) and 2@xmath39(red).,title="fig:",width=226,height=151 ] ( black ) and 2@xmath39(red ) ; top right : peak - to - valley ratio at a gain of 1@xmath39 ; bottom left : after - pulsing ratio at the gain of 1@xmath39 ; bottom right : dark rates at the gain of 1@xmath39(black ) and 2@xmath39(red).,title="fig:",width=226,height=151 ] the test included measuring the single photoelectron spectrum and its peak - to - valley ratio ( p / v ) , gain as a function of the applied hv , dark rate , and after - pulsing ratio ( apr ) . here ,
apr is defined as the probability of observing a pulse excessing a threshold of 0.25 photoelectron within about 20 @xmath40 after the main pulse .
figure [ fig : testresults ] shows the distributions of hv for operating the pmts at gains of 1@xmath39(black ) and 2@xmath39(red ) , p / v at the gain of 1@xmath39 , apr at the gain of 1@xmath39 and the dark rate at the gains of 1@xmath39(black ) and 2@xmath39(red ) , for the 352 macro pmts .
there were 17 assemblies that did not meet our requirements . among them , one showed no output , two had p / v less than two , four had an apr larger than 10@xmath0 , and ten had dark rates higher than @xmath41 at gains of [ 0.3 - 3]@xmath39 .
we decided that six of the ten noisy pmts having dark rates just above @xmath41 and two others with p / v slightly lower than two could still be used . in total
, there were 343 waterproofed macro pmts available to the experiment . by the end of 2011 ,
all waterproof pmt assemblies were installed and commissioned in the three experimental sites . among them , 104 of the 288 in both eh1 and eh2 , and 133 of the 384 in eh3 were waterproofed macro pmts . taking eh1 as an example ,
the singles rates of the pmts were 2 - 8 @xmath42 in the inner partition of the water pool , and 10 - 16 @xmath42 in the outer one .
there was no noticeable difference in the singles rates between the new hamamatsu and refurbished macro waterproof pmts @xcite .
over the course of nearly one year of data - taking , a number of the waterproofed assemblies failed .
the cumulative number of failures over this time is shown in the left plot of figure [ fig : tubefail ] .
the mass distribution of the pmts that failed in the water pools is very similar to that of the pmts that failed the pressure test during fabrication , the latter of which is shown in figure [ fig : tubemass1 ] .
+ , title="fig:",width=226,height=151 ] , title="fig:",width=226,height=151 ] from the end of july through september 2012 , steady data - taking was paused to perform special ad calibration measurements and install the last two ads , one in eh2 and one in eh3 . during this time
, the failed pmts in the water pools were investigated and all but one were replaced with spare hamamatsu pmts . in total , 6 , 5 , and 2 replacements were made in eh1 , eh2 , and eh3 respectively .
the failed pmts were tested for cable leaks using a custom - made cable / air pressurization tube : no cable leak was found .
all but two of the failed pmts were found to have cracks in the glass extending from the equator of the dome near the support ring upward away from the base .
one of the two remaining failed pmts imploded after data - taking was paused and it was removed from eh1 , while the other was removed from eh2 because it had a high dark rate that might have been caused by a leak in the base . upon refilling the pools , 2 and 6 new failures occurred in eh2 and eh3 , respectively .
it is hypothesized that the changes in pressure across the pmt glass during draining and filling the pools triggered the failures .
it was decided not to re - drain the pools and replace these new failures due to the risk of causing more failures .
after all , the very small number of such failures has minimal impact on the performance of the water cherenkov counters . in the subsequent two years of steady data - taking ,
20 additional macro pmt assemblies failed in the three pools ( see the right plot of figure [ fig : tubefail ] ) . since october 2012 ( corresponding to 70 weeks in figure [ fig : tubefail ] ) , the failure rate appears to have stablized to @xmath430.5 pmt / month .
we have successfully fabricated 386 waterproofed pmt assemblies by recycling some of the macro pmts .
the motivation of this effort was to avoid the expense of purchasing new waterproof pmt assemblies .
however , the complexity and cost of this effort were , in the end , higher than expected .
we have installed 341 of these waterproofed pmts in the muon system of the daya bay experiment , corresponding to slightly more than one third of all pmts in the water pools . during the first twelve months of operation , 15 of these 341 pmts malfunctioned in the water pools but only one is thought to have failed due to the imperfect sealing of the pmt base .
the electrical performance of the waterproofed macro pmts has been very similar to that of the new hamamatsu waterproof r5912 pmts .
we would like to thank istituto nazionale di fisica nucleare ( infn ) for offering the macro pmts .
we are grateful for the support provided by the director , office of science , office of high energy physics , of the u.s .
department of energy under contract no .
de - ac02 - 05ch11231 . | in the daya bay reactor neutrino experiment 960 20-cm - diameter waterproof photomultiplier tubes are used to instrument three water pools as cherenkov detectors for detecting cosmic - ray muons . of these 960 photomultiplier tubes ,
341 are recycled from the macro experiment .
a systematic program was undertaken to refurbish them as waterproof assemblies . in the context of passing the water leakage check , a success rate better than 97@xmath0 was achieved .
details of the design , fabrication , testing , operation , and performance of these waterproofed photomultiplier - tube assemblies are presented .
daya bay , reactor , anti - neutrino , waterproof , photomultiplier tube , macro 25.30.mr , 29.40.mc , 29.40.vj , 95.55.vj , 96.40.tv |
it has been known for some time that quantum field theory ( qft ) allows violations of all known pointwise energy conditions ( epstein et al . , 1965 ) such as the `` weak energy condition '' ( wec ) : @xmath0 for all timelike vectors @xmath1 , and the `` null energy condition '' ( nec ) @xmath2 for all null vectors @xmath3 .
( for these pointwise conditions , the latter condition follows from the wec by continuity . )
some known examples of these violations are the casimir effect [ ( casimir , 1948 ) , ( brown & maclay , 1969 ) ] , squeezed states ( slusher et al , 1985 ) , and the evaporation of black holes due to hawking radiation [ ( hawking , 1975 ) , ( davies et al , 1976 ) , ( candelas , 1980 ) ] .
the first two have been experimentally observed ( although the energy density itself is too small to be directly measurable ) .
the third is theoretically predicted , but necessary to insure the consistency of black hole thermodynamics .
all of these imply that we must take negative energy seriously .
on the other hand , if there are _ no _ constraints on negative energy , we could see gross macroscopic effects , such as * possible violations of 2nd law of thermodynamics ( ford , 1978 ) * traversable wormholes [ ( morris and thorne , 1988 ) , ( visser , 1995 ) ] * warp drive [ ( alcubierre , 1994 ) , ( krasnikov , 1998 ) ] * time machines [ ( morris et al . , 1988 ) , ( visser , 1995 ) , ( hawking , 1992 ) ] . over the last ten years , considerable progress has been made on the development of `` quantum inequality ( qi ) '' constraints on negative energy , which were first introduced by l. ford ( ford , 1978 , 1991 ) . as a simple example , consider the massless scalar field in d - dimensional minkowski spacetime ( @xmath4 units ) .
it obeys a qi of the form : @xmath5 where @xmath6 is a smooth sampling function of width @xmath7 , and normalized to have unit integral , and @xmath8 is a constant of order @xmath9 or less , which depends only on @xmath10 and @xmath11 .
the average is taken over a timelike geodesic , for all physically reasonable states @xmath12 and all sampling times @xmath13 .
the class of sampling functions includes both ( smooth ) compactly and non - compactly supported sampling functions . a variety of such bounds have been established over the last decade for massless and massive fields in flat and curved spacetimes for general worldlines and arbitrary hadamard states .
these include various extensions of the qi bounds to free dirac , maxwell , proca , and rarita schwinger fields .
in addition , it has recently been proved that all unitary , positive energy conformal field theories in two - dimensional minkowski space obey qi bounds , thus providing the first examples of qis for interacting quantum field theories . [ for more details and references , see the review articles by fewster ( 2003 ) and roman ( 2004 ) , as well as fewster & roman ( 2005 ) , which is a longer version of the present article . ]
the qis have been used to place strong constraints on exotic spacetimes , such as traversable wormhole and warp drive spacetimes [ ( ford & roman , ( 1996 ) , ( pfenning & ford , 1997 ) , ( everett & roman , 1997 ) , ( fewster , 2004 ) , ( nandi et al . ,
the basic idea is to * obtain the stress - energy tensor required to support a given spacetime from the metric and the einstein equations ; * assume that this stress tensor is generated by quantum fields ; * test it for consistency with the qi bounds , leading to constraints on various parameters arising in the metric .
the flat spacetime qi bounds have been applied to curved spacetimes under the following assumption .
let @xmath14 be the smallest proper radius of curvature ( or smallest proper distance to any boundary in the spacetime ) .
spacetime is approximately minkowskian in regions small compared to @xmath14 , hence one can apply the flat spacetime qi - bounds in the `` short sampling time limit '' : @xmath15 where @xmath16 and the average is taken along timelike geodesics . strictly speaking , this is an _ assumption _ , but it is one for which good justification can be provided .
the equivalence principle implies that physics `` in the small '' should be approximately minkowskian as far as freely falling observers are concerned . the assumption is also borne out by specific examples in four dimensions by taking the short - sampling time limit of various curved spacetime qis [ ( pfenning & ford , 1998 ) , ( fewster & teo , 2000 ) ] .
its validity has recently established for massless scalar fields in general two - dimensional spacetimes ( fewster , 2004 ) .
morris and thorne introduced a class of wormholes described by the metric : @xmath17 where @xmath18 is the radial proper length , and @xmath19 is the `` red - shift function '' .
one can also write the metric in the form @xmath20 where @xmath21 is called the `` shape function '' .
the throat of the wormhole is located at @xmath22 and @xmath23 is the radial proper distance from @xmath24 to any @xmath25 .
the curvature tensor components are @xmath26\ , + \ , { { \phi } ' \over { 2r^2 } } \ , ( b - b'r ) \,,\ ] ] @xmath27 these are the components in a static orthonormal frame .
the relevant stress tensor components for our discussion are @xmath28 @xmath29 \ , .
\label{eq : trr}\ ] ] the quantities @xmath30 and @xmath31 are the mass - energy density and radial pressure , respectively , as measured by a static observer .
let @xmath32 be the magnitude of the maximum curvature component .
then the smallest proper radius of curvature is @xmath33 if we work in a small spacetime volume around the throat whose size is @xmath34 , then the flat qi bounds should then be applicable in this region .
one typically finds ( ford & roman , 1996 ) that the qi bound implies that _ either _ the wormhole is only a little larger than planck size _ or _ there is a large discrepancy in the length scales which characterize the wormhole , e.g. , the ( - ) energy must be concentrated in a thin band _ many orders of magnitude smaller _ than the throat size .
similar constraints have been obtained for warp drives [ ( pfenning & ford , 1997 ) , ( everett & roman , 1997 ) ] .
most qis to date have involved an average of the energy density over a timelike worldline . recently
a `` null - contracted '' qi was proven ( fewster & roman , 2003 ) in which one averages the _ null_-contracted stress tensor @xmath35 over a _
timelike worldline_. [ the techniques used in the proof are those first used by fewster ( 2000 ) to prove a general qi on energy density in arbitrary globally hyperbolic spacetimes . ] as a simple case , consider the massless scalar field in 4d minkowski spacetime where the average is taken over a timelike geodesic with ( constant ) four - velocity @xmath36 , and where @xmath3 is a smooth constant null vector field .
the new qi takes the form @xmath37 for all hadamard states @xmath12 and any smooth compactly supported @xmath38 . as before
, we can apply this qi to wormhole spacetimes in the short sampling time limit , @xmath39 , where @xmath16 .
suppose @xmath40 , for @xmath41 .
then @xmath42 so @xmath43 taking the infimum over smooth @xmath38 compactly supported in @xmath44 . solving the variational problem one gets @xmath45 ( in units where @xmath4 ) . to apply our qi to wormhole spacetimes let us choose @xmath46 , and our worldline to be that of a static observer at the throat .
then for this trajectory , @xmath47 , and @xmath48 where @xmath49 , and @xmath50 is the planck length .
assume that the stress - energy tensor is generated by a hadamard state of a free scalar quantum field .
then our qi bound implies : @xmath51 for @xmath52 we get @xmath53 as a reasonably generous interpretation of @xmath54 , let us choose @xmath55 . since @xmath56 , we have that @xmath57 we can use this bound to obtain strong constraints on wormholes , in general . let us now look at some specific cases .
visser , kar , and dadich ( vkd ) introduced the expression : @xmath58\ , dv = 2\int_{r_0}^{\infty}[\rho + p_r]\,4\pi r^2 \,dr \label{eq : total}\ ] ] as a suitable measure of the `` amount of exotic matter '' required to maintain a traversable wormhole ( visser et al . , 2003 ) .
[ other possible volume measures were discussed by nandi et al .
( 2004 ) . ]
vkd then considered a class of wormholes for which the `` averaged null energy condition anec '' line integral is finite and negative , but for which the volume integral above can be made as small as one likes .
they concluded that one can construct traversable ( in principle ) wormholes using only arbitrarily small amounts of exotic matter .
all of the models discussed by vkd are `` spatially schwarzschild '' , that is , @xmath59
so the @xmath60 spatial slices are the same as those of schwarzschild .
therefore @xmath61 , which implies that @xmath62 , i.e. , the energy density is zero in the static observer s frame . to apply the original qis derived for
energy density one would have to go to a boosted frame in which the energy density is negative .
( such a frame can always be found when the wec is violated . )
the disadvantage is that in so doing , one might have to insure that the sampling time does not exceed the proper time that the observer remains in the region of exotic matter .
this problem does not arise using the _ null_-contracted qi , because the latter involves not only the energy density but also the radial pressure .
hence we can take our observers to be static and located at the throat .
as a specific example of qi constraints on vkd wormholes , we consider the `` piecewise @xmath63 ( zero scalar curvature ) '' wormhole .
for this type of wormhole @xmath64= \epsilon + \lambda \sqrt{1 - 2m / r } \ , , \,\,\,\hfil\break { \rm for } \,\,\,r \in \ , ( r_0=2m , a ) \,,\ ] ] and @xmath64= \sqrt{1 - 2m / r } \ , , \ , \,\,\,{\rm for } \,\,\,r \in \,(a,\infty ) \,,\ ] ] where @xmath65 is essentially the coordinate thickness of the region of exotic matter .
continuity of the metric coefficients implies that @xmath66 where @xmath67 . in this case ,
applying our qi bound gives @xmath69 with @xmath70 , we have @xmath71 so macroscopic `` piecewise @xmath63 '' wormholes are ruled out .
it should also be noted that quite apart from the qi constraints , macroscopic vkd wormholes would be subject to other problems .
the smaller the amount of exotic matter in these wormholes , the closer they are to being vacuum schwarzschild wormholes .
this could lead to problems with traversal , since the smaller the amount of @xmath72 energy , the longer the traversal time , as measured by external clocks .
possibly more serious are stabilization problems . the smaller the total amount of @xmath72 energy , the more unstable the wormhole is to even very small amounts of infalling @xmath73 energy radiation ( e.g. , infalling cmb photons ) , due to their enormous blueshifts .
kuhfittig introduced a series of wormhole models ( kuhfittig , 1999 , 2002 , 2003 ) which were also designed to minimize the total amount of exotic matter . in fewster & roman ( 2005 )
, we used the qi to put severe constraints on one such class . here
we consider a recent model which he claims to be traversable , uses an arbitrarily small amount of exotic matter , and is consistent with the qis ( kuhfittig , 2003 ) . in this model
he writes the metric in the form @xmath74 so that @xmath75 he then assumes that * \(1 ) @xmath76 , so that @xmath77 . *
\(2 ) @xmath78 is finite at @xmath79 .
assumption ( 1 ) implies that the wormhole flares outward very slowly .
assumption ( 2 ) is made to avoid an event horizon at the throat @xmath24 .
in particular , he chooses @xmath80 where here @xmath81 is a ( positive ) constant and @xmath82 where @xmath83 is another positive constant , and @xmath84 .
the choice @xmath85 is made to obtain @xmath86 near @xmath79 .
the condition on @xmath87 is imposed to avoid an event horizon at the throat @xmath24 . these choices for @xmath88 and @xmath89 lead to the following problems : * radially ingoing light rays
reach the throat only after an _ infinite lapse of affine parameter _ ;
* radially infalling particles take an _ infinite amount of proper time _ to reach the throat .
therefore , contrary to his claims , kuhfittig s wormhole is _ not _ traversable .
this failure is related to the very `` slow - flaring '' property , and an apparent confusion between proper and coordinate distance .
as this contribution was being finalised , kuhfittig released a preprint ( kuhfittig 2005 ) which modifies the form of @xmath88 and @xmath89 near the throat to @xmath90 and @xmath91 for positive constants @xmath92 and @xmath83 , with @xmath93 and @xmath84 . with these choices
the wormhole is now traversable , and has @xmath94 , thereby evading the general analysis given above .
however , it suffers from a different problem : one may show that points at radius @xmath95 lie in the nec - violating region for all possible values of the parameters above , and that the radial tidal constraint @xmath96 ( morris and thorne 1988 ) implies , for this value of @xmath97 , that @xmath98 ( ignoring factors of order 1 ) .
but computing the proper distance to this radius , @xmath99 , from the throat yields @xmath100 so this model , as it stands , involves an extensive region of nec - violation , if it is to respect the radial tidal constraint necessary for human traversal .
while the model can doubtless be modified further , to cut off the nec violation before radius @xmath99 , such a modification would have to satisfy the quantum inequalities and tidal constraints at all radii @xmath25 in the nec - violating region if it is to be a reasonable model . whether this is possible or not remains to be seen .
in this work we employed a recently derived qi bound on the _ null_-contracted stress tensor averaged over a _ timelike _
worldline to give a simplified analysis of general wormhole models , not just those with small quantities of exotic matter .
in particular our results imply that _ macroscopic _ vkd wormholes are ruled out or severely constrained , i.e. , there is a large discrepancy between throat size and curvature radius .
we also derived constraints on the kuhfittig models , and in fact showed that one of the recent models presented is _ not _ traversable .
the concentration of exotic matter to an arbitrarily small region around the wormhole throat , in a purely classical analysis , is by itself , not sufficient to guarantee both traversability and consistency with the qi bounds .
the authors would like to thank matt visser , larry ford , and lutz osterbrink for helpful comments .
this work was supported in part by the national science foundation under grant phy-0139969 ( to tar ) .
one of us ( tar ) would like to thank the mathematical physics group at the university of york for their kind hospitality during the course of some of this work .
roman t.a . , some thoughts on energy conditions and wormholes , to appear in the proceedings of the 10th marcel grossmann meeting on general relativity and gravitation , rio de janeiro , brazil ( 2004 ) , gr - qc/0409090 | a quantum inequality bound on the expectation value of the null - contracted stress tensor in an arbitrary hadamard state is used to obtain constraints on the geometries of traversable wormholes .
particular attention is given to the wormhole models of visser , kar , and dadhich ( vkd ) and to those of kuhfittig .
these are models which use arbitrarily small amounts of exotic matter for wormhole maintenance .
it is shown that macroscopic vkd models are either ruled out or severely constrained , while a recent model of kuhfittig is shown to be , in fact , _ non_-traversable .
[ 2001/04/25 1.1 ( pwd ) ] |
the energy relaxation time @xmath1 of superconducting qubits is affected by dielectric loss , nonequilibrium quasiparticles @xcite , and charge or bias noise , and varies between a few nano- to several microseconds , depending on qubit type , material , and device layout .
superconducting qubits are commonly based on @xmath2 thin films , and their central element , the non - linear inductor given by a josephson tunnel junction ( jj ) , is formed either by overlap @xcite or window - type geometries @xcite .
qubit spectroscopy reveals coupling to stochastically distributed two - level systems ( tlss ) in the tunnel oxide @xcite which provide a channel for qubit decoherence .
while the physical nature of tlss is still under debate , their number was shown to decrease with junction size and their density with higher atomic coordination number of the tunnel oxide @xcite .
the number of coherent oscillations in the qubit is limited by , among other decoherence mechanisms such as nonequilibrium quasiparticles , the _ effective _ dielectric loss tangent @xmath3 @xcite .
the overlap geometry provides jjs with amorphous barriers with no need for isolation dielectrics , being itself a source for additional tlss and dielectric losses .
the window geometry is used for higher quality , e.g. epitaxial , trilayer jjs with in - situ grown barriers . besides complex fabrication
, they have the drawback of requiring additional isolation dielectrics @xcite .
the importance of keeping the total dielectric volume in qubits small to reduce the additional loss was shown in ref .
@xcite .
in this paper we give an overview of our standard technology for junction fabrication , and present an alternative junction based on sputtered trilayer stacks , which provide an intrinsically cleaner tunnel oxide and is well suited for micron - sized trilayer qubit junctions .
the so - called _ side - wall passivated jjs _ provide contact to the top electrode without adding too much lossy dielectric to the circuitry , which would negatively affect the loss tangent .
the trilayer isolation is achieved via an electrolytic process .
these novel jjs were realized in a flux - biased phase qubit and characterized by i ) current transport measurements on reference junctions and ii ) spectroscopy and time - domain measurements of the qubit . by systematically replacing only the josephson junction ,
being central to any superconducting qubit , we aim to analyze the loss contributions of this specific element , and , ideally , develop low - loss josephson junctions for superconducting qubits and improve our qubit performance .
we found performance comparable to the current generation of overlap phase qubits .
figure [ sidewallanod ] depicts the patterning process for our standard overlap ( a ) and trilayer junctions ( b ) .
our standard process has an oxide layer grown on an ion mill cleaned aluminum edge , which was previously chlorine etched .
the top wiring is then etched back below the oxide layer using argon with @xmath4 chlorine mixture . for the trilayer process ,
the in - situ sputtered @xmath2-@xmath5-@xmath2 trilayer has a thermally grown tunnel oxide barrier , formed for 10 min at @xmath6 at room temperature .
after deposition of the trilayer stack an edge is etched .
the bottom electrode of the trilayer stack is isolated from the top electrode wiring by a self - aligned nanometer thin dielectric layer , grown for @xmath2 ( or other suitable electrode metals such as @xmath7 ) by anodic oxidation @xcite .
the metallic aluminium serves as partly submerged anode in a liquid electrolytic mixture of @xmath8 ammonium pentaborate , @xmath9 ethylene glycol and @xmath10 @xmath11 at room temperature .
a gold - covered metal served as cathode and the electric contact was made outside the electrolyte to the anode . by protecting parts of the aluminum electrode with photoresist only a well - defined area was oxidized by passing a constant current through the al film and converting the metallic surface to its oxide form .
the oxide thickness can be controlled by the voltage drop across the electrolyte . after a light ion clean and top wiring deposition the resist
is patterned to define the junction area .
finally , the trilayer is etched below the tunnel barrier , yielding josephson junctions with planar tunnel barrier and isolation dielectric on just one side of the tunnel area .
the in - situ grown tunnel oxide avoids sources of residual impurities such as hydrogen , hydroxide or carbon at the interface vicinity , which may remain even after ion - milling in our standard process .
these trilayer junctions are fully compatible with our standard process using overlap patterning and no junction side - wall
. + transport measurements on a @xmath12 reference junction at @xmath13 are shown in fig .
the critical current @xmath14 is @xmath15 , with normal resistance @xmath16 yielding @xmath17 , close to the calculated ambegaokar - baratoff value of @xmath18 for the measured superconducting gap of @xmath19 .
the retrapping current of @xmath20 indicates a very small subgap current .
the current transport is consistent with tunneling , and we can exclude transport via metallic pinholes , located in the @xmath21 thin side - wall dielectric . as a further check ,
the @xmath22 dependence is as expected , see inset in fig .
[ ivc ] , with a critical temperature @xmath23 of @xmath24 . + and @xmath22 dependence ( lower inset ) of a @xmath25 side - wall passivated trilayer junction .
top inset : dielectric circuit elements of the junction .
the tunnel oxide capacitance @xmath26 is connected in parallel with the capacitor formed by the side - wall oxide @xmath27.,width=325 ]
the qubit is a flux - biased phase qubit that is coupled via a tunable mutual inductance to the readout - squid @xcite . , provided by a parallel plate capacitor with relative permittivity @xmath28 made from hydrogenated amorphous silicon ( a - si : h ) .
the measurement process follows the standard phase qubit characterization @xcite . when operated as a qubit , spectroscopy over a range of more than @xmath29 revealed clean qubit resonance spectra with just two avoided level crossings of @xmath30@xmath31 coupling strength ( at @xmath32 and @xmath33 , as shown in fig .
[ fig:2dspectro ] ) .
the excitation pulse length is @xmath34 , and the qubit linewidth is about @xmath35 in the weak power limit .
the qubit visibility , measured in a separate experiment , is about @xmath36 , which is in the range we found for our standard phase qubits .
+ qualitatively , the tls number and coupling strength per qubit is lower than in other trilayer systems @xcite , that have larger tunnel areas .
the tls density per qubit has roughly the same order of magnitude as in conventional overlap qubits with similar tunnel area dimensions @xcite . .
two avoided level crossings due to qubit - tls coupling are observed at @xmath32 and @xmath33 ( arrows ) .
top inset : dielectric circuit schematics of the qubit .
bottom inset : qubit relaxation measurement.,width=325 ] qubit relaxation measurements via @xmath37 pulse excitation and time - varied delay before readout pulse were obtained when operated outside the avoided level structures .
we measured a relaxation time @xmath1 of about @xmath38 , as shown in the lower inset of fig .
[ fig:2dspectro ] .
this relaxation time is similar to that observed in the overlap qubits , which consistently have @xmath39 for @xmath40-@xmath41 jj size .
apart from the change to trilayer junctions , no modification from the previous design was made .
.dielectric parameters for anodic oxide , shunt capacitance , and tunnel barrier .
@xmath42 is given for low temperature and low power at microwave frequencies .
the capacitance for the tunnel oxide @xmath5 is taken and corrected for @xmath2 electrodes from ref .
@xcite for the dimensions given in the text . for qubits the loss tangent is calculated away from tls resonances , as the losses in small size anodic side - wall oxide and tunnel barrier are smaller than the bulk value considered for the specific loss contribution @xmath43 .
the @xmath44 is a factor 2 - 3 smaller than @xmath45 , the weighted sum of all specific loss contributions .
[ tab : lossestimation ] [ cols="<,<,<,<,<,^ " , ] we estimate the additional dielectric loss due to the sidewall oxide .
the _ effective _ loss tangent of a parallel combination of capacitors is given by @xmath46 with @xmath47 and @xmath48 being the real and imaginary part of the individual permittivity for capacitor @xmath49 with area @xmath50 and dielectric thickness @xmath51 .
now , we discuss the individual loss contributions for all dielectrics .
we design the qubit so that the dominant capacitance comes from the shunt capacitor made from a - si : h , which has a relatively low loss tangent of @xmath52 . including the non - negligible capacitance of the tunnel junction
, this gives an effective loss tangent to the qubit of @xmath53 .
because the junction capacitance is about 10% of the shunting capacitance , the effective junction loss tangent is 10 times less than the loss tangent of the junction oxide .
we statistically avoid the effects of two - level systems by purposely choosing to bias the devices away from the deleterious resonances .
the loss tangent of the junction is smaller than the value for bulk aluminum oxide , approximately @xmath54 , and probably smaller than @xmath55 since long energy decay times ( @xmath56 ) have been observed for an unshunted junction when operated away from resonances @xcite .
+ the anodic side - wall oxide contributes a small capacitance of about 3.2ff , which can be calculated assuming a parallel plate geometry . here
, we use the dielectric constant @xmath57 for aluminum oxide , assume an area given by @xmath58 m , the width of the overlap , multiplied by @xmath59 m the thickness of the base layer , and estimate the thickness of the oxide @xmath60 nm as determined by the anodic process @xcite . the anodic oxide is assumed to have a bulk loss tangent of @xmath54 @xcite , which gives a net qubit loss contribution of @xmath61 , about 5 times lower than for the a - si : h capacitor .
note that we expect the loss from this capacitance to be even lower because of statistical avoidance of the tls loss @xcite .
the small volume of the capacitor , equivalent to a @xmath62 volume tunnel junction , implies that most biases do not put the qubit on resonance with two - level systems in the anodic oxide .
from the measured energy decay time @xmath63 , we determine the loss tangent of the qubit to be @xmath64 , using a qubit frequency @xmath65 .
this is 3 - 5 times larger than our estimation of our dielectric losses , as shown in table [ tab : lossestimation ] .
we believe the dissipation mechanism comes from some other energy loss sources as well , such as non - equilibrium quasiparticles @xcite .
in conclusion , we have shown that the use of a anodic oxide , self - aligned to the junction edge , does not degrade the coherence of present phase qubits @xcite .
we found performance comparable to the current generation of overlap phase qubits . the new junction geometry may provide a method to integrate submicron sized , superior quality junctions ( lower tls densities ) grown , for example , by mbe epitaxy to eliminate the need for shunt dielectrics .
also , our nanometer - thin , three dimensional - conform anodic passivation layer can be replaced by a self - aligned isolation dielectric at the side - wall , which could be used for all types of trilayer stacks .
the authors would like to thank d. pappas for stimulating discussions .
this work was supported by iarpa under grant w911nf-04 - 1 - 0204 .
m.w . acknowledges support from avh foundation and m.m . from an elings postdoctoral fellowship .
matthias steffen , m. ansmann , r. mcdermott , n. katz , radoslaw c. bialczak , erik lucero , matthew neeley , e. m. weig , a. n. cleland , and john m. martinis .
state tomography of capacitively shunted phase qubits with high fidelity .
, 97(5):050502 , aug 2006 .
jrgen lisenfeld , clemens mller , jared h. cole , pavel bushev , alexander lukashenko , alexander shnirman , and alexey v. ustinov .
rabi spectroscopy of a qubit - fluctuator system .
, 81(10):100511 , mar 2010 .
john m. martinis , k. b. cooper , r. mcdermott , matthias steffen , markus ansmann , k. d. osborn , k. cicak , seongshik oh , d. p. pappas , r. w. simmonds , and clare c. yu .
decoherence in josephson qubits from dielectric loss . , 95(21):210503 , nov 2005 .
matthew neeley , m. ansmann , radoslaw c. bialczak , m. hofheinz , n. katz , erik lucero , a. oconnell , h. wang , a. n. cleland , and john m. martinis . transformed dissipation in superconducting quantum circuits .
, 77(18):180508 , may 2008 .
aaron d. oconnell , m. ansmann , r. c. bialczak , m. hofheinz , n. katz , erik lucero , c. mckenney , m. neeley , h. wang , e. m. weig , a. n. cleland , and j. m. martinis .
microwave dielectric loss at single photon energies and millikelvin temperatures .
, 92(11):112903 , 2008 . | we have developed a novel josephson junction geometry with minimal volume of lossy isolation dielectric , being suitable for higher quality trilayer junctions implemented in qubits .
the junctions are based on in - situ deposited trilayers with thermal tunnel oxide , have micron - sized areas and a low subgap current . in qubit
spectroscopy only a few avoided level crossings are observed , and the measured relaxation time of @xmath0 is in good agreement with the usual phase qubit decay time , indicating low loss due to the additional isolation dielectric . |
the study of the dynamics of phase transitions started with the question posed by kibble @xcite .
he noted that , in the rapidly cooling post big bang universe , phase transitions must lead to disparate local choices of the broken symmetry vacuum .
the resulting mosaic of domains with independently chosen vacua will in turn precipitate formation of topological defects with observable consequences .
for instance , kibble theorized that the presently observed cosmological structures have been seeded by cosmic strings
an example of such defects .
while the original estimate @xcite of defect density was incorrect , as it relied on an equilibrium argument @xcite ( hence , densities did not depend on the phase transition rate ) , the key question whether transitions leave relics such as topological defects in their wake can be also posed in the condensed matter setting and studied in the laboratory .
the theory @xcite developed to estimate defect density for the second order phase transition relies on their non - equilibrium nature but also on the universality class of the transition : it uses equilibrium critical exponents to predict scaling of the density of defects and other excitations with the quench timescale .
the size of domains within which symmetry breaking can be coordinated is given by the size of the `` sonic horizon '' that estimates how far the local choice of the broken symmetry can be communicated ( and plays a similar role as the causal horizon used bykibble @xcite to set lower bounds on defect density ) the usual estimate of the sonic horizon size relies on a power - law scaling of the relaxation time and the healing length with a distance from the critical point that is characterized by the dynamical and correlation length critical exponents @xmath0 and @xmath1 .
a characteristic time - scale @xmath2 and a length - scale @xmath3 are predicted , where the quench time @xmath4 quantifies the rate of the transition .
the correlation length @xmath5 determines the number of excitations as a function of @xmath4 .
the kibble - zurek mechanism ( kzm ) has been confirmed by numerical simulations @xcite and by experiments @xcite in a variety of settings , with most recent results providing evidence of the kzm scaling laws @xcite .
refinements and extensions of kzm include phase transition in inhomogeneous systems ( see for a recent overview ) , generation of winding numbers in the limit when flux trapping is a rare event @xcite ( and kzm scaling must be suitably adjusted @xcite ) , and applications that go beyond topological defect creation ( see e.g. ) .
recent reviews related to kzm are also available @xcite .
in this paper we consider a zero - temperature quantum phase transition . despite important differences with respect to thermodynamic transitions where thermal rather than quantum fluctuations act as seeds of symmetry breaking the kzm can be generalized to quantum phase transitions @xcite , see also for reviews .
the quantum regime was also addressed in some of the recent experiments @xcite .
recently a fully fledged scaling hypothesis was proposed @xcite as a generalization and extension of the predictive power of kzm , though some of its basic ingredients were known from the beginning @xcite . since in the adiabatic limit , when @xmath6 , both scales @xmath7 and @xmath8 diverge
, they should become the only relevant time and length scales in the regime of low frequencies and long wavelengths .
this in turn suggests a dynamical scaling hypothesis , similar to the ( static ) one in equilibrium phase transitions , that during the quench all physical observables depend on a time @xmath9 through the scaled time @xmath10 and on a distance @xmath11 through the scaled distance @xmath12 .
what makes it really powerfull is the fact that the two scales are not independent : @xmath13 .
this space - time renormalization hypothesis was confirmed in a precise experiment @xcite where it proved useful in organizing the experimental data . in the following
we also find it useful in organizing our numerical results .
in this paper we consider the mott - superfluid quantum phase transition in the 1d bose - hubbard model that belongs to the kosterlitz - thouless universality class .
this problem was touched upon in ref . where it was argued that , since the correlation length diverges exponentially near the critical point , one can not ascribe a definite scaling exponent @xmath14 to @xmath15 except for @xmath16 in the limit of exceedingly slow @xmath4 that are beyond any realistic experiment . however , when the range of @xmath4 is restricted to one or two orders of magnitude , then an effective scaling @xmath15 with an effective exponent @xmath17 can be a convenient approximation . in this paper
, we readdress the problem , this time with fully fledged numerical dmrg simulations . like in a real experiment ,
there are limitations that restrict the range of available quench times and the kz scaling hypothesis with effective exponents is a convenient approximation .
we simulate linear quenches from the mott insulator to superfluid , where the range of correlations builds up as the tunnelling rate between nearest - neighbor sites is turned on .
the spatial profile of the correlators and their time dependence satisfy the scaling hypothesis .
we also simulate reverse linear quenches from the superfluid to mott insulator .
somewhat surprisingly , we find the excitation during a ramp across the gapless superfluid to be negligible as compared to the excitation that builds up after crossing the critical point to the mott phase .
apparently , the different critical superfluid ground states that are crossed by the linear ramp are similar enough for the excitation to be negligible despite their vanishing gap .
what matters here is that a relatively large change of the tunneling rate during the ramp across the superfluid phase corresponds to a relatively small change of the luttinger liquid parameter @xmath18 that determines the ground state of the liquid .
when measured by a distance between different ground states , the superfluid phase can be effectively identified as a single critical point .
it is only after crossing to the mott phase that the ground state begins to change fundamentally .
this inevitably makes the excitations build up until their growth is halted by the opening mott gap .
the last crossover , that takes place at the time @xmath7 after crossing to the mott phase , is the essence of the quantum kzm .
our simulations confirm this simple scenario by demonstrating that the excitation energy in the mott phase satisfies the kz scaling hypothesis .
the paper is organized as follows .
we begin with a general discussion of the textbook version of the quantum kibble - zurek mechanism in section [ sec : qkz ] .
this generic version assumes gapfull phases on both sides of the transition and a power law divergence of the correlation length at the critical point .
these assumptions are relaxed in section [ sec : kzinkt ] where we introduce the 1d bose - hubbard model and consider the phase transtion from mott insulator to superfluid that belongs to the kosterlitz - thouless universality class . here
the divergence of the correlation length on the mott side is exponential and the whole superfluid phase is gapless and critical .
we recall relevant predictions of ref . for this type of transition . in this paper
these results are corroborated by numerical simulations described in sections [ sec : misf ] and [ sec : sfmi ] . in section
[ sec : misf ] we describe simulations of the mott - to - superfluid transition and in section [ sec : sfmi ] those of the reverse superfluid - to - mott quench .
finally , we briefly conclude in section [ sec : concl ] .
( -4,0)(4,0 ) ; ( 0,0)(0,4 . ) ; at ( 4,0 ) @xmath9 ; ( -1.5,-.1 ) node[below]@xmath19 ( -1.5,0.1 ) ; ( 0,-.1 ) node[below]@xmath20 ( 0,0.1 ) ; ( 1.5,-.1 ) node[below]@xmath21 ( 1.5,0.1 ) ; at ( -3,3 ) adiabatic ; at ( + 3,3 ) adiabatic ; at ( 0,4.3 ) impulse ; at ( -3.7,0.8 ) ; at ( + 3.5,1.75 ) ; plot ( , -4./ ) ; plot ( , + 4./ ) ; plot ( , ) ; plot ( - , ) ; ( -1.5,0 ) ( -1.5,4 ) ; ( 1.5,0 ) ( 1.5,4 ) ; ( -1.5,0 ) ( 1.5,0 ) ( 1.5,4 ) ( -1.5,4 ) cycle ; a distance from a quantum critical point can be measured with a dimensionless parameter @xmath22 .
the ground state of the hamiltonian @xmath23 undergoes a fundamental change at @xmath24 when the correlation length in its ground state diverges like ~||^- [ xi ] and the relevant gap closes like ~||^z .
[ delta ] the system , initially prepared in its ground state , is driven across the critical point by a linear quench , ( t)= , [ quench ] with a quench time @xmath4 .
the evolution sufficiently far from the critical point is initially adiabatic .
however , the rate of change of epsilon , ||= , [ rate ] diverges at the gapless critical point .
therefore , the evolution can not be adiabatic in its neighborhood between @xmath25 and @xmath7 , see fig . [
fig : kzm1 ] . here
@xmath7 is the time when the gap ( [ delta ] ) equals the rate ( [ rate ] ) , so that : t ~_q^z/(1+z ) .
[ hatt ] just before the adiabatic - to - non - adiabatic crossover at @xmath25 , the state of the system is still approximately the adiabatic ground state at @xmath26 , where = _
q^-1/(1+z ) , [ hatepsilon ] with a correlation length ~^- ~_q^/(1+z ) .
[ hatxi ] in a first - order impulse approximation , this state `` freezes out '' at @xmath25 and does not change until @xmath7 . at @xmath7 the frozen state is no longer the ground state but an excited state with a correlation length @xmath5 .
it is the initial state for the adiabatic process that follows after @xmath7 .
no matter how quantitatively accurate is the above `` freeze - out scenario '' , this simple scaling argument establishes @xmath5 and @xmath7 , interrelated via t~^z , [ hattz ] as the relevant scales of length and time , respectively .
what is more , in the adiabatic limit , when @xmath6 , both scales diverge becoming the unique scales in the regime of long - wavelength and low - frequency . in analogy to static critical phenomena , this uniqueness suggests a scaling hypothesis @xcite : ( t)| o(x ) |(t)=
^-_o f_o(t/^z , x/ ) .
[ oscaling ] here @xmath27 is the quantum state during the quench , @xmath28 is an operator depending on a distance @xmath11 , @xmath29 is its scaling dimension , and @xmath30 its scaling function .
the same conclusions about scaling are reached when one follows the narrative based on the sonic horizon @xcite . in ( [ hatt ] )
becomes a linear function @xmath31 , where @xmath32 is a microscopic energy scale . for comparison , in ( a )
, we plot @xmath7 for the kosterlitz - thouless transition in the 1d bose - hubbard model ( [ hattkt ] ) in function of @xmath4 over many decades of the argument .
this function becomes linear only asymptotically for @xmath6 , but it may appear linear locally , i.e. , in a range of one or two decades . indeed , in ( b ) , we focus on the narrow range of @xmath33 that is small enough for a realistic cold - atom experiment .
these plots can be reasonably approximated by linear functions , especially when experimental error bars are present . in ( c ) , we plot a local slope @xmath34 of the log - log plot in panel ( a ) in function of @xmath35 .
the slope @xmath36 in eq .
( [ hatxinaive ] ) is achieved asymptotically but only for @xmath4 that are `` cosmological '' in magnitude . in ( d ) , a focus on the realistic @xmath4 shows that the local slope can be significantly less than @xmath36 . ]
the power laws ( [ xi],[delta ] ) are not directly applicable in the kosterlitz - thouless ( kt ) transition @xcite , where , on the disordered / mott side of the transition , the correlation length s divergence is exponential : @xmath37 here @xmath38 and @xmath39 is a microscopic scale of length .
this faster - than - polynomial divergence can be captured by stating that @xmath40 , see _ e.g. _ , but it may tempt one to misuse eqs .
( [ hatt ] , [ hatepsilon ] , [ hatxi ] ) by inserting @xmath40 together with @xmath41 to obtain : t ~_q^1 , ~_q^0 , ~_q^1 .
[ hatxinaive ] as shown in ref .
, these equations are valid asymptotically for @xmath6 but this asymptote is achieved for unrealistically slow @xmath4 ( and , hence , astronomically large @xmath5 of the order of kilometers @xcite ) .
below we briefly recount the argument . in the 1d bose - hubbard model at commensurate
filling , where @xmath41 , the gap @xmath42 on the mott - insulator side of the transition closes like @xmath43 where @xmath32 is a microscopic energy scale . for the linear ramp ( [ quench ] ) , driving a quench from the mott insulator to superfluid , this gap equals the rate ( [ rate ] ) at @xmath44 when @xmath45 a solution of this transcendental equation is @xmath46 where @xmath47 is the lambert function @xcite .
the above solution is plotted for different values of @xmath48 in figs .
[ fig : hatt]a , b .
a similar relation has been derived for the first time and confirmed by numerical simulations in a dissipative classical model @xcite .
it has been also tested in a recent experiment @xcite .
figure [ fig : hatt ] shows that the exponent of unity for the dependence of @xmath7 on @xmath4 in eq .
( [ hatxinaive ] ) is attained only for exceedingly slow quenches that are unlikely to be experimentally or numerically accessible . for any reasonably slow quenches
the effective exponent would be significantly less than @xmath36 .
the above argument determines the correlation length @xmath5 imprinted on the quantum state on the mott side of the transition .
it characterizes the excited state when the quench ramp enters the superfluid phase . in this critical phase
the imprinted correlations are spreading with a velocity limited by the speed of quasiparticles .
the 1d bose - hubbard model at the commensurate filling of @xmath36 particle per site is described by a hamiltonian h= -j_s=1^l-1 ( b_l^b_l+1+b_l+1^b_l)+ _ l=1^l n_l(n_l-1 ) , where @xmath49 is the hopping rate and @xmath50 is the on - site interaction strength .
the kosterlitz - thouless quantum phase transition from the mott - insulator to superfluid takes place at @xmath51 ( see _ e.g. _ ref . ) .
we consider a quench driven by a linear ramp j(t)=j_c \ { ll 0 , & for t-_q , + 1+t/_q , & for t > -_q . .
[ eq : jt ] from @xmath52 to @xmath53 .
the initial state is the ground state at @xmath54 : |(0)=|1,1,1, ... ,1 .
[ fock ] figure [ fig : dmrgvbog ] shows that there is a finite density of excitation energy @xmath55 above the adiabatic ground state during a quench , hence the evolution with the ramp is not adiabatic .
furthermore , the system seems to be excited in two stages by two different mechanisms .
the early oscillations visible at small @xmath49 are excited by the discontinuous time derivative of the ramp at @xmath54 that is proportional to @xmath56 . since this sudden initial
jolt excites states well above the mott gap with a probability proportional to @xmath57 , the amplitude of the early excitation is roughly @xmath58 , as can be demonstrated by the quantum perturbation theory @xcite . for small enough @xmath49
they compare well with predictions of the bogoliubov doublon - holon model @xcite , see figure [ fig : dmrgvbog ] . for larger @xmath49 , as their energy gap closes with @xmath49 approaching @xmath59 , the early oscillations are adiabatically suppressed .
closer to @xmath59 the kz mechanism steps in , see fig .
[ fig : kzm2 ] . at the earlier stages of the project we attempted to mitigate the effect of the initial jolt by initiating the quench more smoothly . for rapid quenches
such efforts proved only party successful .
however , similarly as for the linear ramp in fig .
[ fig : dmrgvbog ] , long quench times suppress the effect of these early `` jolt '' excitations compared to these caused by the crossing of the critical region .
( -4,0)(4,0 ) ; ( 0,0)(0,4 . ) ; at ( 4,0 ) @xmath9 ; ( -1.5,-.0 ) node[below]@xmath19 ( -1.5,0.1 ) ; ( 0,-.0 ) node[below]@xmath20 ( 0,0.1 ) ; at ( -3,3 ) adiabatic ; at ( 0,4.3 ) impulse ; at ( -2.7,0.3 ) ; at ( -3.0,1.2 ) ; plot ( , -3.8/ ) ; plot ( , 2*exp(-0.5/sqrt(- ) ) ) ; plot ( , 0.1 ) ; ( -1.5,0 ) ( -1.5,4 ) ; ( -1.5,0 ) ( 3.8,0 ) ( 3.8,4 ) ( -1.5,4 ) cycle ; the quench times in our numerical simulations are far below the `` astronomical '' standards that would demand @xmath5 of the order of kilometers to approach the scaling @xmath60 in eq . ( [ hatxinaive ] ) . limited by the system size
@xmath61 they span a narrow range of magnitude .
therefore , we can assume a phenomenological power - law for the ground - state correlation length in the mott phase : ~^-_l .
[ nul ] given the exponential divergence ( [ xikt ] ) , we expect the effective exponent @xmath62 to be large . to account for finite - size effects , we allow for its dependence on @xmath61 .
furthermore , we found that the data can be accurately parametrized by letting the gap scale with an effective dynamical exponent : ~^-z_l ~^z_l_l .
[ zl ] we expect to recover the exact @xmath63 for sufficiently large @xmath61 . in function of the scaled distance @xmath12 measured in the bulk of @xmath64 lattice sites at the scaled times @xmath65 . with increasing @xmath66 the plots collapse to the scaling function @xmath67 . in ( b ) ,
the same as in ( a ) but for @xmath68 and @xmath69 .
, title="fig : " ] in function of the scaled distance @xmath12 measured in the bulk of @xmath64 lattice sites at the scaled times @xmath65 . with increasing @xmath66 the plots collapse to the scaling function @xmath67 . in ( b ) , the same as in ( a ) but for @xmath68 and @xmath69 .
, title="fig : " ] the correlation length in the ground state is obtained from the best fit to the tail of the correlator : c(x ) = b_s^b_s+x ~. the lengths and gaps for different @xmath22 and @xmath61 are collected in fig .
[ fig : zlnul ] .
they fit well the phenomenological power - laws ( [ nul],[zl ] ) . as a self - consistency check ,
figure [ fig : zlnul ] shows how the dynamical exponent @xmath70 decays to @xmath36 with increasing @xmath61 . for each @xmath61
we use the effective exponents to find the kz scales [ hattxil ] t ~_q^ , ~_q^. these scales are applied to verify the kz scaling hypothesis , c(t , x ) = ^-1/4 f_c(t/^z_l , x/ ) , in fig .
[ fig : misfscaling ] . the plots for different @xmath4 collapse demonstrating validity of the hypothesis .
the collapsed plots in fig . [ fig : misfscaling ] depend on the scaled time @xmath10 .
this effect reveals that the impulse approximation is not quite correct : since the correlations are spreading , the state can not be frozen .
this behavior suggests a `` sonic horizon '' paradigm we have also noted earlier .
nevertheless , as discussed and demonstrated by an exact solution of the quantum ising chain @xcite , the scaling hypothesis still holds because the velocity @xmath71 of the spreading must also be a combination of the two kz scales : @xmath72 . for a linear dispersion at the critical point , @xmath41 ,
@xmath73 is expected to be bounded from above by twice the speed @xmath74 of quasiparticles at the critical point . in figure
[ fig : misfhatv ] , we attempt an estimate of the spreading velocity @xmath71 .
since accurate tails of the correlators can not be accessed on a finite lattice , our estimate is not robust but at least it is less than the @xmath75 estimated from the quasiparticle dispersion in figure [ fig : misfhatv ] . in function of the quench time @xmath4 at the center of @xmath64 lattice at the time @xmath76 when the quench is crossing the critical point .
the solid line is the best fit @xmath77 consistent with @xmath78 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]in ( b ) , the same as in ( a ) but for @xmath68 sites .
the solid line is the best fit @xmath79 consistent with @xmath80 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]b . in ( c ) , scaled excitation energy density in function of scaled time for @xmath64 sites . the plots collapse to the scaling function @xmath81 in the kz regime close to @xmath59 .
, title="fig : " ] in function of the quench time @xmath4 at the center of @xmath64 lattice at the time @xmath76 when the quench is crossing the critical point .
the solid line is the best fit @xmath77 consistent with @xmath78 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]in ( b ) , the same as in ( a ) but for @xmath68 sites .
the solid line is the best fit @xmath79 consistent with @xmath80 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]b . in ( c ) , scaled excitation energy density in function of scaled time for @xmath64 sites
. the plots collapse to the scaling function @xmath81 in the kz regime close to @xmath59 .
, title="fig : " ] in function of the quench time @xmath4 at the center of @xmath64 lattice at the time @xmath76 when the quench is crossing the critical point .
the solid line is the best fit @xmath77 consistent with @xmath78 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]in ( b ) , the same as in ( a ) but for @xmath68 sites .
the solid line is the best fit @xmath79 consistent with @xmath80 predicted by eqs .
( [ hattxil],[wtscaling ] ) with the exponents fitted in fig .
[ fig : zlnul]b . in ( c ) , scaled excitation energy density in function of scaled time for @xmath64 sites .
the plots collapse to the scaling function @xmath81 in the kz regime close to @xmath59 .
, title="fig : " ] neglecting remnants of the early excitation , a distribution of quasiparticle excitations should have a scaling form @xmath82 . in the kz regime close to @xmath59
, their dispersion can be approximated by the gapless linear dispersion at the critical pojnt , @xmath83 , and the excitation energy ( also sometimes referred to as work @xcite ) density should satisfy a scaling hypothesis : w / l = c_0|k| f(t / t , k ) = ^-2
f_w(t / t ) .
[ wtscaling ] here @xmath81 is a scaling function . in particular , at @xmath59 ( where @xmath84 ) we expect @xmath85 .
this scaling law and the more general scaling hypothesis ( [ wtscaling ] ) are confirmed by the data in fig .
[ fig : misfenergy ] . between a block of @xmath86 sites at the end and the rest of the chain of @xmath68 sites .
panel ( a ) , ( b ) , and ( c ) show the entropy at @xmath44 , @xmath87 , and @xmath20 , respectively .
all plots are in function of the scaled block size , @xmath88 , and the entropy is divided by @xmath89 this logarithmic function of @xmath4 is obtained as the best fit to @xmath90 with fitting parameters @xmath91 and @xmath92 . the best central charge @xmath92 for @xmath44 , @xmath87 , @xmath20 is @xmath93 , @xmath94 , @xmath95 , respectively .
the plots collapse demonstrating the scaling hypothesis in eq .
( [ sscaling ] ) . ]
finally , we considered the entropy of entanglement between a block of @xmath86 sites at the end of the chain and the rest of it . ignoring boundary effects , the entropy should satisfy a scaling hypothesis : = f_s(t / t , l/ ) . [ sscaling ] here
@xmath91 and @xmath96 are scaling functions and @xmath92 is the central charge @xmath97 at the kosterlitz - thouless transition .
this hypothesis is tested in fig .
[ fig : misfentropy ] . except for @xmath44 , where the state is still close to the ground state , the fitted @xmath92 , though close to @xmath36 , are not quite satisfactory , but we have to bear in mind that the range of @xmath4 at hand is too narrow for a better fit with a logarithmic function . however , we find that in accordance with the scaling hypothesis
the plots collapse to a scaling function @xmath96 after rescaling the block size by @xmath5 .
in this section we reverse the quench . now the linear ramp begins when @xmath98 deep in the superfluid phase at @xmath99 and ends at @xmath54 when @xmath100 : j(t)=j_c(1-t/_q ) .
[ eq : jt2 ] the initial ground state has a quasi - long - range order characterized by a power - law decay of its correlation function @xmath101 .
small excitations in the superfluid described by a luttinger liquid are gapless , hence it may be tempting to treat the whole evolution in the superfluid phase as impulse .
however , different critical ground states in the superfluid phase are sufficiently similar to each other for a significant excitation to be postponed until after the ramp crosses the boundary with the mott phase where the gap begins to open and the ground state begins to change fundamentally . from this perspective , the similarity between different superfluid critical states means that on the superfluid side of the transition there is little difference between the impulse and adiabatic approximations .
the whole superfluid critical phase can be effectively collapsed to the mott critical point .
indeed , at @xmath76 when the quench is leaving the superfluid at @xmath59 the correlator @xmath102 quickly converges with increasing @xmath4 to the correlator in the adiabatic ground state @xmath103 at @xmath59 , see figure [ fig : sfmicxt0 ] .
this fast convergence is consistent with the quick decay of the excitation energy density shown in fig .
[ fig : sfmicxt0 ] .
its decay is steep , almost as steep as predicted for a linear ramp in a luttinger liquid @xcite : @xmath104 with @xmath105 standing for the speed of luttinger quasiparticles .
it turns out to be steep enough for the excitation in the superfluid phase to give negligible contribution to the final excitation deep in the mott phase , see fig .
[ fig : sfmiwj0 ] .
its panel shows the final excitation energy density after the full ramp from @xmath99 to @xmath54 and panel after a shorter ramp from @xmath106 to @xmath54 .
the two panels are almost indistinguishable and in both the excitation energy density is two orders of magnitude higher than in fig . [
fig : sfmicxt0 ] .
therefore , the final excitation after the full ramp originates almost exclusively from the evolution in the mott phase when the gap opens and the ground state undergoes a fundamental change .
( 0,0)(8,0 ) ; ( 0,0)(0,4 . )
; at ( 8,0 ) @xmath9 ; ( 0,-.1 ) node[below]@xmath20 ( 0,0.1 ) ; ( 3.0,-.1 ) node[below]@xmath107 ( 3.0,0.1 ) ; at ( + 6,4.3 ) adiabatic ; at ( 1.4,4.3 ) impulse ; at ( + 1.2,3.5 ) ; at ( + 6.0,0.8 ) ; plot ( , + 4./ ) ; plot ( , exp(-0.5/sqrt ( ) ) ) ; ( 3,0 ) ( 3,4 ) ; ( 0,0 ) ( 3,0 ) ( 3,4 ) ( 0,4 ) cycle ; the mott part of the full ramp , from @xmath59 to @xmath20 , falls within the kz framework , as shown schematically in figure [ fig : kzmhalf ] .
the evolution in the mott phase crosses over from impulse to adiabatic near @xmath108 when the correlation length in the ground state is @xmath109 .
as long as quasiparticle excitations can be considered non - interacting , their distribution @xmath82 satisfies the kz scaling hypothesis .
this scaling form has at least two important consequences . in function of
scaled time @xmath10 for different quench times @xmath4 .
the plots collapse to the scaling function @xmath110 in eq .
( [ wtscaling ] ) . ] before @xmath7 , when the quasiparticle dispersion can be approximated by its critical form @xmath111 , the excitation energy density should conform to the scaling law ( [ wtscaling ] ) .
indeed , the scaled plots in fig .
[ fig : sfmiwt ] demonstrate a collapse to the scaling function @xmath81 . as predicted , the collapse is perfect up to @xmath7 . in the adiabatic stage after @xmath7 ,
the quasiparticle distribution freezes out , @xmath112 , and their dispersion @xmath113 for the `` excited '' @xmath114 between @xmath115 can be approximated by the finite gap @xmath116 .
consequently , the final excitation energy density at @xmath54 should scale as = _ -^ _ k f(k ) ~ ^-1 .
[ wjscaling ] since in this regime the excited quasiparticles are approximately dispersionless , @xmath117 , the excitation energy density is simply proportional to their density @xmath118 . with our best fits for @xmath62 and @xmath70 on the @xmath64 lattice
we obtain @xmath119 .
this is roughly consistent with the best fit @xmath120 in fig .
[ fig : sfmiwj0 ] .
in a linear quench from mott insulator to superfluid the excitation energy density , the entropy of entanglement , and the correlations that build up as the system is crossing the critical point satisfy the kz scaling hypothesis with effective power laws accurate for a limited range of quench times @xcite . in particular , the range of correlations scales with an effective power of the transition time .
all superfluid ground states are qualitatively similar .
therefore , in a reverse quench from superfluid to mott insulator the excitation in the gapless superfluid turns out to be negligible as compared to the excitation that begins to build up just after crossing the critical point when the gap opens and the ground state begins to change fundamentally .
the last excitation also falls into the kz framework .
the final excitation energy deep in the mott phase proportional to the number of empty and doubly - occupied sites decays with an effective power of the quench time .
work of j.d . and b.g .
was supported in part by narodowe centrum nauki ( national science center ) under project no .
2013/09/b / st3/01603 and 2016/20/s / st2/00152 , respectively . work of w.h.z . was supported by the us department of energy under the los alamos ldrd program .
this research was supported in part by pl - grid infrastructure .
101ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty http://stacks.iop.org/0305-4470/9/i=8/a=029 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/0370-1573(80)90091-5 [ * * , ( ) ] http://stacks.iop.org/1751-8121/49/i=41/a=411001 [ * * , ( ) ] \doibase doi:10.1038/317505a0 [ * * , ( ) ] http://www.actaphys.uj.edu.pl/vol24/abs/v24p1301.htm?series=reg&vol=24&page=1301 [ * * , ( ) ] \doibase doi:10.1016/s0370 - 1573(96)00009 - 9 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.78.2519 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.80.5477 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.82.4749 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.82.2824 [ * * , ( ) ] link:\doibase 10.1103/physrevd.62.065005 [ * * , ( ) ] http://www.actaphys.uj.edu.pl/vol31/abs/v31p2937.htm?series=reg&vol=31&page=2937 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.075701 [ * * , ( ) ] http://stacks.iop.org/1367-2630/12/i=11/a=115003 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.106.135301 [ * * , ( ) ] link:\doibase 10.1103/physreva.76.043613 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.101.115701 [ * * , ( ) ] link:\doibase 10.1103/physrevb.93.014106 [ * * , ( ) ] \doibase doi:10.1038/srep00352 [ * * , ( ) ] link:\doibase 10.1103/physrevx.5.021015 [ * * , ( ) ] http://dx.doi.org/10.1038/ncomms8406 [ * * , ( ) ] , link:\doibase 10.1103/physrevb.93.075134 [ * * , ( ) ] link:\doibase 10.1126/science.251.4999.1336 [ * * , ( ) ] link:\doibase 10.1126/science.263.5149.943 [ * * , ( ) ] link:\doibase 10.1038/382334a0 [ * * , ( ) ] link:\doibase 10.1038/382332a0 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.84.4966 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.89.080603 [ * * , ( ) ] link:\doibase 10.1103/physrevb.80.180501 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.91.197001 [ * * , ( ) ] link:\doibase 10.1038/nature05094 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.247002 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.167603 [ * * , ( ) ] link:\doibase 10.1103/physrevx.2.041022 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.110.133004 [ * * , ( ) ] link:\doibase 10.1103/physreva.87.051401 [ * * , ( ) ] \doibase doi:10.1038/ncomms3290 [ * * , ( ) ] http://dx.doi.org/10.1038/ncomms3291 [ * * , ( ) ] link:\doibase 10.1038/nature07334 [ * * , ( ) ] http://dx.doi.org/10.1038/nphys2734 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.113.135302 [ * * , ( ) ] http://dx.doi.org/10.1038/ncomms7162 [ * * , ( ) ] http://dx.doi.org/10.1038/nphys3142 [ * * , ( ) ] link:\doibase 10.1126/science.1258676 [ * * , ( ) ] @noop `` , '' ( ) , link:\doibase 10.1103/physrevlett.116.155301 [ * * , ( ) ] link:\doibase 10.1126/science.aaf9657 [ * * , ( ) ] http://stacks.iop.org/0953-8984/25/i=40/a=404210 [ * * , ( ) ] http://stacks.iop.org/0953-8984/25/i=40/a=404209 [ * * , ( ) ] link:\doibase 10.1103/physreva.83.062104 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.102.105702 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.160404 [ * * , ( ) ] link:\doibase 10.1103/physreva.75.052321 [ * * , ( ) ] http://link.springer.com/book/10.1007{%}2f978-94-007-1029-0[__ ] ( , , ) \doibase http://dx.doi.org/10.1063/1.2784684 [ * * , ( ) ] link:\doibase 10.1080/00018732.2010.514702 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.83.863 [ * * , ( ) ] link:\doibase 10.1142/s0217751x1430018x [ * * , ( ) ] link:\doibase 10.1103/physrevlett.88.167001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.95.035701 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.95.105701 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.95.245701 [ * * , ( ) ] link:\doibase 10.1103/physrevb.72.161201 [ * * , ( ) ] link:\doibase 10.1103/physrevb.76.174303 [ * * , ( ) ] http://stacks.iop.org/1742-5468/2009/i=02/a=p02007 [ * * , ( ) ] link:\doibase 10.1103/physrevb.78.144301 [ * * , ( ) ] link:\doibase 10.1103/physreve.81.012101 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.100.077204 [ * * , ( ) ] link:\doibase 10.1103/physrevb.78.045101 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.224408 [ * * , ( ) ] link:\doibase 10.1103/physrevb.77.214427 [ * * , ( ) ] http://stacks.iop.org/1742-5468/2009/i=05/a=p05005 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.054306 [ * * , ( ) ] http://stacks.iop.org/1742-5468/2009/i=03/a=p03032 [ * * , ( ) ] http://stacks.iop.org/0295-5075/89/i=6/a=67001
[ * * , ( ) ] link:\doibase 10.1103/physrevb.81.165431 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.107.140402 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.106.235304 [ * * , ( ) ] link:\doibase 10.1073/pnas.1408861112 [ * * , ( ) ] http://dx.doi.org/10.1038/nphys3695 [ * * , ( ) ] , http://dx.doi.org/10.1038/srep33381 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.109.015701 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.064304 [ * * , ( ) ] http://stacks.iop.org/0295-5075/84/i=6/a=67008 [ * * , ( ) ] http://stacks.iop.org/1367-2630/11/i=6/a=063014 [ * * , ( ) ] http://stacks.iop.org/1367-2630/12/i=5/a=055007 [ * * , ( ) ] http://dx.doi.org/10.1038/srep05950 [ * * , ( ) ] http://stacks.iop.org/0022-3719/6/i=7/a=010 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.47.1542 [ * * , ( ) ] link:\doibase 10.1038/nature04851 [ * * , ( ) ] @noop `` , '' ( ) https://books.google.com/books?id=tumzhaaacaaj[__ ] , research report ( , ) http://stacks.iop.org/1742-5468/2011/i=02/a=p02032 [ * * , ( ) ] link:\doibase 10.1073/pnas.1500763112 [ * * , ( ) ] link:\doibase 10.1088/1367 - 2630/14/12/125015 [ * * , ( ) ] link:\doibase 10.1103/physreva.75.023603 [ * * , ( ) ] @noop `` , '' ( ) , @noop `` , '' ( ) , link:\doibase 10.1103/physreve.91.032119 [ * * , ( ) ] link:\doibase 10.1103/physreve.92.042126 [ * * , ( ) ] link:\doibase 10.1103/physrevb.84.214522 [ * * , ( ) ] | the ground state of the one - dimensional bose - hubbard model at unit filling undergoes the mott - superfluid quantum phase transition .
it belongs to the kosterlitz - thouless universality class with an exponential divergence of the correlation length in place of the usual power law .
we present numerical simulations of a linear quench both from the mott insulator to superfluid and back .
the results satisfy the scaling hypothesis that follows from the kibble - zurek mechanism ( kzm ) . in the superfluid - to - mott quenches
there is no significant excitation in the superfluid phase despite its gaplessness .
since all critical superfluid ground states are qualitatively similar , the excitation begins to build up only after crossing the critical point when the ground state begins to change fundamentally .
the last process falls into the kzm framework . |
in this section we introduce clique colourings and geometric graphs ; and we present our main results , on clique colourings of deterministic and random geometric graphs . recall that a _ proper colouring _ of a graph is a labeling of its vertices with colours such that no two vertices sharing the same edge have the same colour ; and the smallest number of colours in a proper colouring of a graph @xmath10 is its _ chromatic number _ , denoted by @xmath11 .
we are concerned here with another notion of vertex colouring . a _ clique _
@xmath12 is a subset of the vertex set such that each pair of vertices in @xmath13 is connected by an edge ; and a clique is _ maximal _ if it is not a proper subset of another clique .
a _ clique colouring _ of a graph @xmath14 is a colouring of the vertices such that no maximal clique is monochromatic , ignoring isolated vertices .
the least number of colours in such a colouring is the _ clique chromatic number _ of @xmath14 , denoted by @xmath15 .
( if @xmath14 has no edges we take @xmath15 to be 1 . )
clearly , @xmath16 but it is possible for @xmath15 to be much smaller than @xmath11 .
for example , for any @xmath17 we have @xmath18 but @xmath19 .
note that if @xmath14 is triangle - free then @xmath20 . for several graph classes
the maximum clique chromatic number is known to be 2 or 3 . for maximum value 2
we have for example : comparability graphs @xcite , claw - free perfect graphs @xcite , odd - hole and co - diamond free graphs @xcite , claw - free planar graphs @xcite , powers of cycles ( other than odd cycles longer than three , which need three colours ) @xcite , and claw - free graphs with maximum degree at most @xmath21 ( again , except for odd cycles of length more than three ) @xcite . for maximum value 3
we have for example : planar graphs @xcite , co - comparability graphs @xcite , circular - arc graphs ( see @xcite ) and generalised split graphs ( see @xcite ) .
further related results can be found in @xcite , @xcite and @xcite .
it was believed for some time that perfect graphs had bounded clique chromatic number , perhaps with maximum value 3 ( see @xcite or for example @xcite ) ; but it was shown very recently that in fact such clique chromatic numbers are unbounded @xcite .
the behaviour of the clique chromatic number for the binomial ( known also as erds - rnyi ) random graph @xmath22 is investigated in @xcite and @xcite . on the algorithmic side
, it is known that testing whether @xmath23 for a planar graph can be performed in polynomial time @xcite , but deciding whether @xmath23 is @xmath24-complete for perfect graphs @xcite and indeed for @xmath25-free perfect graphs @xcite , and for graphs with maximum degree @xmath26 @xcite .
we are interested here primarily in clique colourings of geometric graphs in the plane , but we shall also briefly consider geometric graphs in @xmath27 for any positive integer @xmath28
. given @xmath0 points @xmath29 in @xmath27 and given a threshold distance @xmath2 , the corresponding ( euclidean ) geometric graph has vertex set @xmath3 , and for @xmath30 , vertices @xmath4 and @xmath5 are adjacent when the euclidean distance @xmath31 .
we call a graph @xmath14 _ geometric _ or _ geometric in _ @xmath27 if there are points @xmath7 and @xmath2 realising @xmath14 as above . by rescaling by a factor @xmath32 we may assume , without loss of generality , that @xmath33 . a geometric graph in @xmath34 is also called a _ unit disk graph_. our first theorem shows that the clique chromatic number is uniformly bounded for geometric graphs in the plane . (
in contrast , bacs et al .
@xcite observed that @xmath15 is unbounded even for line graphs of complete graphs , and recall that @xmath15 is unbounded for perfect graphs . )
[ thm.detr2 ] if @xmath14 is a geometric graph in the plane @xmath34 then @xmath35
. let @xmath36 denote the maximum value of @xmath15 over geometric graphs @xmath14 in @xmath27 .
clearly @xmath37 is at least 3 ( consider @xmath38 ) so we have @xmath39 : it would be interesting to improve these bounds . in section [ sec.detproofs ]
we shall see that more generally @xmath36 is finite for each @xmath28 , but ( perhaps unsurprisingly ) @xmath40 as @xmath41 ; and we shall see further related deterministic results . for random geometric graphs the upper bound in theorem [ thm.detr2 ] can often be improved .
given a positive integer @xmath0 and a threshold distance @xmath2 , we consider the _ random geometric graph _
@xmath42 on vertex set @xmath43 obtained as before by starting with @xmath0 random points sampled independently and uniformly in the square @xmath44 ^ 2 $ ] , see @xcite .
( we could equally work with the unit square @xmath45 ^ 2 $ ] . )
note that , with probability @xmath46 , no point in @xmath47 is chosen more than once , so we may identify each vertex @xmath48 with its corresponding geometric position @xmath49 .
the ( usual ) chromatic number of @xmath9 was studied in @xcite , see also @xcite .
we say that events @xmath50 hold _ with high probability _
( _ whp _ ) if the probability that @xmath50 holds tends to @xmath46 as @xmath0 goes to infinity .
also , we use @xmath51 to denote natural logarithm . it is known that the value @xmath52 is a sharp threshold function for connectivity for @xmath42 ( see , for example , @xcite ) .
this means that for every @xmath53 , if @xmath54 , then @xmath14 is disconnected whp , whilst if @xmath55 , then @xmath14 is connected whp .
the next two results summarise what we know about the clique chromatic number @xmath56 of a random geometric graph @xmath14 in the plane ; but first here is an overview .
as @xmath8 increases from 0 we have whp the following rough picture : @xmath15 is 1 up to about @xmath57 , then 2 up to about @xmath58 , then at least 3 ( and at most @xmath59 ) up to about @xmath60 ( roughly the connectivity threshold ) , when it drops back to 2 and remains there .
[ thm.rggsummary ] for the random geometric graph @xmath42 in the plane : 1 .
if @xmath61 then @xmath62 whp , 2 . if @xmath63 then @xmath64 and @xmath65 , 3 . if @xmath66 and @xmath67 then @xmath68 whp , 4 . if @xmath69 then @xmath70 and @xmath71 , for a suitable constant @xmath72 ( see below ) , 5 . if @xmath73 and @xmath74 then @xmath75 whp , 6 .
if @xmath76 then @xmath68 whp . the constant @xmath77 in part ( 4 ) above may be expressed explicitly as an integral , see equation ( 3.2 ) in @xcite .
it is the asymptotic expected number of components @xmath38 in the case when @xmath78 .
we can say more within the interval in ( 5 ) above where @xmath79 : at the low end of the interval we have @xmath80 whp ; and higher up , within a suitable subinterval , @xmath15 is whp as large as is possible for a geometric graph .
[ prop.rgg4 ] for the random geometric graph @xmath42 in the plane : 1 .
if @xmath73 and @xmath81 then @xmath80 whp , 2 .
there exists @xmath53 such that , if @xmath82 then @xmath83 whp .
the only random geometric graphs we consider here are those described above , where the points are independently and uniformly distributed over a square in the plane .
see @xcite for more general models of random geometric graphs , and see @xcite in particular for models in high dimensions .
in this section , we start by proving theorem [ thm.detr2 ] , and then consider geometric graphs in dimensions greater than 2 .
after that we give lemma [ lem.detsmall ] , concerning the maximum value of @xmath15 for general @xmath0-vertex graphs , for small values of @xmath0 : this result will be used in the next section in the proof of proposition [ prop.rgg4 ] .
fix @xmath84 with @xmath85 .
divide the plane into horizontal strips @xmath86 for @xmath87 .
suppose we are given a finite set of points in the plane , and let @xmath14 be the corresponding unit disk graph .
consider one strip , let @xmath88 be the subset of the given points which are in the strip ( which we may assume is non - empty ) , and @xmath89 be the geometric graph corresponding to @xmath88 .
we claim that @xmath90 . for @xmath91
we write @xmath92 if @xmath93 and @xmath94 . if @xmath92 then @xmath95 so @xmath96 . thus if also @xmath97 then @xmath98 , so @xmath99 .
thus @xmath100 is a ( strict ) partial order on @xmath88 .
further , @xmath101 is the corresponding comparability graph , since if @xmath94 then @xmath102 ( for if @xmath103 then @xmath104 so @xmath105 is in @xmath106 not @xmath107 ) .
thus @xmath89 is a co - comparability graph .
hence , by the result of duffus et al .
@xcite mentioned earlier , we have @xmath90 , as claimed .
now label the strips cyclically @xmath108 moving upwards say , and use 3 colours to properly clique colour the @xmath109-strips , a new set of 3 colours for the @xmath110-strips and similarly a new set of 3 colours for the @xmath111-strips , using 9 colours in total .
a monochromatic maximal clique with at least 2 vertices could not have points in two different strips since @xmath112 , and could not be contained in one strip since we have a proper clique - colouring there .
thus @xmath113 .
theorem [ thm.detr2 ] shows that the clique chromatic number is at most 9 for any geometric graph in the plane .
we next see that , for a given dimension @xmath28 , there is a uniform bound on the clique chromatic number for all geometric graphs in @xmath27 .
[ thm.detrd ] let @xmath14 be a geometric graph in @xmath27 .
then @xmath114 our simple proof uses a tessellation into small hypercubes which induce cliques . in the case
@xmath115 it is better to use hexagonal cells , and then the bound improves from 18 to 14 . in @xcite ,
hexagonal cells are used in pairs to show that @xmath116 , nearly matching the upper bound 9 in theorem [ thm.detr2 ] .
we may assume that the threshold distance @xmath8 is 1 .
let @xmath117 , let @xmath118 , and let @xmath119 be the hypercube @xmath120 .
observe that @xmath119 has diameter @xmath121 , so the subgraph of @xmath14 induced by the points in @xmath119 is complete .
we partition @xmath27 into the family of translates @xmath122 of @xmath119 , for @xmath123 .
( here @xmath124 is the set of all points @xmath125 for @xmath126 . )
consider the subfamily @xmath127 .
let @xmath128 and @xmath129 be distinct points in @xmath130 , and let @xmath131 and @xmath132 be points in the cells @xmath133 and @xmath134 in @xmath135 respectively . without loss of generality , we may assume that @xmath136 .
then @xmath137 thus the subgraph @xmath138 of @xmath14 induced on the vertices corresponding to the points in the cells of @xmath135 consists of disjoint cliques , with no edges between them .
hence @xmath139 , since we just need to ensure that each cell with at least two points gets two colours .
finally , let @xmath140 denote the translate by @xmath141 of the family @xmath135 , so @xmath142 ( and @xmath143 ) .
let @xmath144 be the union of the cells in @xmath140 , and let @xmath145 be the subgraph of @xmath14 induced by the vertices corresponding to the points in @xmath144 .
then the @xmath146 sets @xmath144 for @xmath147 partition @xmath27 ; and so @xmath148 as required for the first inequality . for the second inequality
, we have @xmath149 and the proof is finished .
for example , we may deduce from this result that @xmath150 .
it is not hard to make small improvements for each @xmath28 , but let us focus on the case @xmath151 .
[ thm.detr3 ] if @xmath14 is a geometric graph in @xmath152 then @xmath153 .
let @xmath154 denote the unit triangular lattice in @xmath34 , with vertices the integer linear combinations of @xmath155 and @xmath156 ( and where the edges have unit length ) .
consider the hexagonal packing in the plane , as in figure 1 , formed from the hexagonal voronoi cells of @xmath154 . at distance 1 , and solid lines bound the hexagonal voronoi cells , width=240,height=240 ] the sublattice @xmath157 of @xmath154 with vertices generated by @xmath158 and @xmath159 is a triangular lattice with edge - length @xmath160 , and 7 translates of @xmath161 partition @xmath162 ( for example translate by @xmath163
see figure [ fig:7colors ] , and for example @xcite ) .
we thus obtain a 7-colouring of the vertices of @xmath154 , and this gives a 7-colouring of the cells . since the cells have diameter @xmath164 , the distance between any two cells centred on distinct points in @xmath157 is at least @xmath165 .
( in fact , the minimum distance occurs for example between the cells centred on @xmath166 and on @xmath158 , and equals @xmath167 . )
thus our 7-colouring of the cells is such that , for any two distinct cells of the same colour , the distance between them is at least @xmath168 ( see also theorems 3 and 4 of @xcite for related results ) .
, width=240,height=240 ] rescale by multiplying by @xmath169 , so that the diameter of a hexagonal cell is now @xmath170 .
the distance between distinct rescaled cells corresponding to centres in @xmath157 has now been reduced to at least @xmath171 , still bigger than 1 .
suppose that we are given any finite set of points in @xmath152 , take @xmath33 , and let @xmath14 be the corresponding geometric graph .
think of @xmath152 as @xmath172 .
consider any cell @xmath173 , and let @xmath174 be the geometric graph corresponding to the points in the cylinder @xmath175 , with threshold distance @xmath33 .
we may now argue as in the proof of theorem [ thm.detr2 ] : for clarity we spell this out .
observe that for @xmath176 , if @xmath177 then @xmath178 so @xmath179 .
for @xmath176 we write @xmath92 if @xmath180 and @xmath181 . if @xmath92 then @xmath182 and so @xmath183 .
if also @xmath97 then similarly @xmath184 ; and then @xmath185 and so @xmath99 .
it follows that @xmath100 is a ( strict ) partial order , and @xmath174 is the co - comparability graph .
thus , once more by the result of duffus et al .
@xcite , we have @xmath186 . consider the 7-colouring of the cells . for each colour @xmath187 and each cell @xmath173 of colour @xmath188 , properly clique colour the points in @xmath175 using colours @xmath189 . if two points in distinct cylinders have the same colour , then the distance between them is at least @xmath190 , so the corresponding vertices are not adjacent in @xmath14 . thus the colourings of the cylinders fit together to give a proper clique colouring of @xmath14 using at most 21 colours , as required .
the next result shows that , if we do not put some restriction on the dimension @xmath28 , then we can say nothing about a geometric graph in @xmath27 .
[ prop.allgeom ] for each graph @xmath14 there is a positive integer @xmath28 such that @xmath14 is a geometric graph in @xmath27 . indeed if @xmath14 has @xmath17 vertices we can take @xmath191 .
let @xmath192 and let @xmath14 have vertex set @xmath193 $ ] . if @xmath14 is complete we may take @xmath194 trivially , so suppose not .
let @xmath195 be the set of unordered non - adjacent pairs @xmath196 of vertices .
we shall work in @xmath197 , with a co - ordinate for each @xmath198 .
let the vertex @xmath199 have degree @xmath200 in the complementary graph @xmath201 .
we specify the point @xmath202 corresponding to vertex @xmath199 as follows , by specifying its value @xmath203 at co - ordinate @xmath204 , where @xmath205 :
@xmath206 suppose vertices @xmath207 and @xmath208 are adjacent .
then @xmath209 on the other hand , if @xmath207 and @xmath208 are non - adjacent then @xmath210 hence the unit disk graph on these points @xmath202 is @xmath14 .
let @xmath211 be the maximum value of @xmath15 over all @xmath0-vertex graphs .
since the ramsey number @xmath212 satisfies @xmath213 , there exist @xmath0-vertex triangle - free graphs @xmath214 with stability number @xmath215 ( see @xcite for the best known bounds ) and thus with chromatic number and hence clique chromatic number @xmath216 . hence @xmath217 it now follows from proposition [ prop.allgeom ] that @xmath218 this shows explicitly
that @xmath40 as @xmath41 , though the lower bound here is rather a long way from the upper bound ( roughly @xmath219 ) provided by proposition [ thm.detrd ] .
( see also section [ sec.concl ] , where we discuss @xmath36 in paragraph ( 2 ) , and @xmath211 in paragraphs ( 5 ) and ( 6 ) . )
it is convenient to give one more deterministic result here , which we shall use in the proofs in the next section .
let us note first that the grtzsch graph is triangle - free on @xmath220 vertices and has chromatic number @xmath221 , and thus clique chromatic number @xmath221 .
[ lem.detsmall ] let the graph @xmath14 have @xmath0 vertices . if @xmath222 then @xmath223 except if @xmath14 is isomorphic to @xmath38 when @xmath80 .
if @xmath224 then @xmath225 .
suppose that @xmath222 .
if @xmath226 then colouring @xmath227 with colour 1 and the other vertices with colour 2 shows that @xmath223 : thus we may assume that each degree is at most 2 . if @xmath14 has a triangle then @xmath14 consists of a triangle perhaps with one additional disjoint edge ,
so @xmath223 . if @xmath14 does not have a triangle , then either @xmath14 is isomorphic to @xmath38 or @xmath228 . also ,
since @xmath38 has no triangles , @xmath229 .
this completes the proof of the first statement .
now let us prove that @xmath225 for @xmath230 .
now suppose for a contradiction that @xmath230 and @xmath231 , and @xmath0 is minimal such that this can happen .
the minimum degree in @xmath14 is at least 3 ( for if @xmath232 and @xmath233 then @xmath225 ) .
suppose that @xmath234 for some vertex @xmath208 , and let @xmath235
. then @xmath236 and @xmath237 ( since @xmath238 ) .
thus by the first part of the lemma , @xmath239 and @xmath138 is isomorphic to @xmath38 .
but now @xmath208 has neighbours in @xmath138 , a contradiction .
it follows that @xmath14 is cubic .
hence @xmath0 is even , and so @xmath240 .
now let @xmath208 be any vertex and as before let @xmath235 .
arguing as before , we must have @xmath237 so @xmath138 is isomorphic to @xmath38 and @xmath208 has neighbours in @xmath138 , a contradiction .
it remains only to show that @xmath225 when @xmath241 .
as above , we may assume that @xmath14 is connected and the minimum degree in @xmath14 is at least @xmath26 .
if @xmath14 has a vertex @xmath208 with @xmath242 , then by the case @xmath243 of the lemma , @xmath235 satisfies @xmath244 , since @xmath138 consists of an isolated vertex and a 4-vertex graph : but now , using the third colour for each vertex in @xmath227 , we see that @xmath245
. if each vertex has degree at most 3 then @xmath246 by brooks theorem ( since @xmath14 is connected and is not @xmath25 ) .
now we may assume , without loss of generality , that @xmath14 has a vertex @xmath208 with @xmath239 . since @xmath208 is isolated in @xmath235 , by the case @xmath247 of the lemma , @xmath244 ( and thus as before @xmath245 ) unless @xmath138 is the disjoint union of the vertex @xmath208 and the @xmath248-cycle @xmath249 with edges @xmath250 ( where @xmath251 means @xmath252 ) .
assume that @xmath138 is indeed of this form .
we now have two cases .
* case 1 : * _ there are adjacent vertices @xmath4 , @xmath253 in the cycle @xmath173 that form a triangle with some vertex @xmath254 . _
+ we may @xmath26-clique colour @xmath14 as follows . without loss of generality , assume that @xmath255 .
give colour 1 to @xmath208 , @xmath256 , @xmath257 and @xmath258 ; give colour 2 to @xmath252 and @xmath259 ; and give colour 3 to each vertex in @xmath227 .
let @xmath260 be a monochromatic clique of size at least 2 .
if @xmath260 has only colour 1 , then @xmath260 can not contain @xmath208 or @xmath258 ( since they have no neighbours coloured 1 ) , so we can add @xmath207 to @xmath260 ; @xmath260 can not have only colour 2 ( since the vertices coloured 2 form a stable set ) ; and if @xmath260 has only colour 3 then we can add @xmath208 to @xmath260 .
* case 2 : * _ no two vertices in the cycle @xmath173 form part of a triangle .
_ + each vertex @xmath254 can be adjacent to at most two ( non - adjacent ) vertices in @xmath173 , and every vertex @xmath4 in @xmath173 has at least @xmath46 and at most @xmath261 neighbours in @xmath227 .
hence some vertex in @xmath173 has exactly one neighbour in @xmath227 : without loss of generality , assume that @xmath252 has exactly one neighbour , say , @xmath262 in @xmath227 . note that @xmath262 is not adjacent to @xmath256 or @xmath258 : since @xmath262 is adjacent to at most one of @xmath263 we may assume , without loss of generality , that @xmath262 is not adjacent to @xmath264 . give colour 1 to @xmath265 ; give colour 2 to @xmath266 ; and give colour 3 to @xmath252 and each vertex in @xmath267 . as in the first case ,
let @xmath260 be a monochromatic clique of size at least 2
. then @xmath260 can not be only colour 1 or only colour 2 , since the vertices coloured 1 and the vertices coloured 2 both form stable sets ; and if @xmath260 has only colour 3 then @xmath268 ( since @xmath252 has no neighbours coloured 3 ) so we can add @xmath208 to @xmath260 .
the grtzsch graph is triangle - free on @xmath220 vertices and has chromatic number @xmath221 , and thus has clique chromatic number @xmath221 . since @xmath269 by the last result
, it follows that @xmath270 .
indeed , we may deduce easily that @xmath271 in order to see it , suppose @xmath14 is connected and has @xmath272 vertices : we must show that @xmath273 .
if each vertex has degree at most 4 then @xmath274 by brooks theorem , and so @xmath273 .
if some vertex @xmath208 has degree at least 5 then @xmath275 $ ] has at most 10 vertices , so @xmath276 .
in this section we prove theorem [ thm.rggsummary ] and proposition [ prop.rgg4 ] .
we use one preliminary lemma that concerns the appearance of small components in the random geometric graph @xmath14 .
it is taken from chapter 3 of @xcite , where it is proved using poisson approximation techniques .
[ lem.comps ] let @xmath277 be an integer , let @xmath89 be a connected unit disk graph with @xmath278 vertices , and let @xmath279 be the constant defined in equation ( 3.2 ) in @xcite . 1 .
if @xmath280 then whp @xmath14 has no component with @xmath278 or more vertices .
2 . if @xmath281 where @xmath282 then the expected number of components @xmath89 tends to @xmath283 , and the probability that @xmath14 has a component @xmath89 tends to @xmath284 .
3 . if @xmath285 and @xmath286 then whp @xmath14 has a component @xmath89 .
( in part ( 2 ) above , the number of components @xmath89 converges in distribution to poisson@xmath287 , but we shall not use that fact here . ) we may now prove theorem [ thm.rggsummary ] , taking the parts in order .
we shall use the last lemma several times , sometimes without explicit reference .
_ proof of theorem [ thm.rggsummary ] _ _ part _ ( 1 ) .
the expected number of edges is asymptotic to @xmath288 .
thus by markov s inequality , if @xmath61 then whp @xmath14 has no edges so @xmath289 .
( this also follows from lemma [ lem.comps ] part ( 1 ) with @xmath89 as the complete graph @xmath290 . ) _ part _ ( 2 ) . if @xmath63 where @xmath282 , then the expected number of edges tends to @xmath283 , where @xmath291 ( edge - effects are negligible ) . also , since @xmath292 , whp each component has size at most 2 , and so @xmath223 .
hence @xmath293 , and @xmath294 . _
part _ ( 3 ) .
if @xmath66 then whp @xmath14 has an edge ( and indeed @xmath14 has at least one component that is an isolated edge ) , so @xmath295
. if @xmath67 then whp each component of @xmath14 has size at most 4 , and then @xmath223 by lemma [ lem.detsmall ] .
these two results combine to prove part ( 3 ) . _
part _ ( 4 ) .
if @xmath69 ( where @xmath282 ) , then the probability there is a component @xmath38 tends to @xmath296 , where @xmath297 . also whp @xmath14 has edges and each component has size at most 5 . hence @xmath298 ; and , using also lemma [ lem.detsmall ] , @xmath299 . _
part _ ( 5 ) .
if @xmath73 and @xmath286 , then whp @xmath14 has a component @xmath38 , and so @xmath79 .
the following lemma covers the remainder of the relevant range of values for @xmath8 .
[ lem.rgg2 ] let @xmath42 with @xmath300 and @xmath301 . then @xmath79 whp . in order to simplify the proof of lemma [ lem.rgg2 ]
we will make use of a technique known as poissonization , which has many applications in geometric probability ( see @xcite for a detailed account of the subject ) .
here we sketch all we need .
consider the related model of a random geometric graph @xmath302 , where the set of points is given by a homogeneous poisson point process of intensity @xmath46 in the square @xmath47 of area @xmath0 .
in other words , we form our graph from @xmath303 points in the square @xmath304 chosen independently and uniformly at random , where @xmath303 is a poisson random variable of mean @xmath0 .
the main advantage of generating our points by a poisson point process arises from the following two properties : ( a ) the number of points that lie in any region @xmath305 of area @xmath109 has a poisson distribution with mean @xmath109 , and the numbers of points in disjoint regions of @xmath47 are independently distributed ; and ( b ) by conditioning @xmath302 on the event @xmath306 , we recover the original distribution of @xmath9 .
therefore , since @xmath307 , any event holding in @xmath302 with probability at least @xmath308 must hold whp in @xmath9 .
our plan is to show that whp @xmath14 contains a copy of @xmath38 such that no edge of this copy is in a triangle in @xmath14 , and so @xmath79 . in order to allow @xmath8 to be as large as possible we consider a configuration of 5 points such that the corresponding unit disk graph is @xmath38 , and the area @xmath309 that must contain no further points ( to avoid unwanted triangles ) is as small as possible .
we work in the poisson model @xmath302 . within @xmath47
choose @xmath310 disjoint square cells which are translates of @xmath311 .
for each of these cells , we shall consider a regular pentagon @xmath119 centered at the center of the cell and contained well within the cell .
consider first the square @xmath312 .
start with a regular pentagon , with extreme points listed clockwise as @xmath313 around the boundary , centred on the origin @xmath314 , and scaled so that the diagonals ( for example @xmath315 ) have length 1 .
the angle @xmath316 is @xmath317 , and the line @xmath318 is orthogonal to the line @xmath315 and bisects it . hence the radius ( from the centre @xmath319 to each extreme point @xmath320 ) is @xmath321 .
( we give numbers rounded to 6 decimal places . )
if @xmath154 is the midpoint of the side @xmath322 , then the line @xmath323 is orthogonal to @xmath322 and the angle @xmath324 is @xmath325 .
hence the side length @xmath326 ( the length of @xmath322 for example ) satisfies @xmath327 , so @xmath328 . for each successive pair @xmath329 of extreme points ( including @xmath330 ) ,
let @xmath331 be the intersection of the unit radius disks centred on @xmath320 and @xmath332 ; and let the ` controlled region ' @xmath333 be the union of the @xmath331 , with area @xmath309 .
for the value of @xmath309 , we have the following claim .
* claim : * @xmath334 .
we may calculate @xmath309 as follows .
let us take @xmath335 to be on the @xmath84-axis above the origin @xmath336 , so @xmath337 .
now @xmath338 .
let us denote @xmath335 by @xmath88 and @xmath339 by @xmath340 .
suppose that the circle @xmath173 of radius 1 centred on @xmath340 meets the lines @xmath341 ( on which @xmath88 lies ) and @xmath342 ( bisecting the angle between @xmath343 and @xmath344 ) above the @xmath345-axis at @xmath119 and @xmath346 respectively .
then the area @xmath309 is @xmath347 , where @xmath348 is the area bounded by these two straight lines and the arc of the circle @xmath173 between @xmath119 and @xmath346 see the shaded area on figure [ fig3 ] .
we may calculate @xmath348 as the area @xmath349 of the sector of the circle bounded by the arc between @xmath119 and @xmath346 and the radii @xmath350 and @xmath351 , less the area @xmath352 of triangle @xmath353 , plus the area @xmath354 of triangle @xmath355 . recall that @xmath154 is the point of intersection of the lines @xmath356 and @xmath357 , and note that @xmath358 and @xmath356 and @xmath357 are orthogonal .
now ( by pythagoras theorem ) @xmath359 and @xmath360 ; and so @xmath361 and @xmath362 .
drop a perpendicular from @xmath119 to the ( extended ) line @xmath356 , meeting the line at @xmath363 .
note that the angle @xmath364 is @xmath365 , and so @xmath366 .
hence , by considering the triangle @xmath367 in which @xmath368 , @xmath369 is the positive solution @xmath370 of the quadratic equation @xmath371 and thus we obtain @xmath372 .
it follows that the angle @xmath373 is @xmath374 , the angle @xmath375 is @xmath376 , and so @xmath377 moreover , @xmath378 @xmath379 and so @xmath380 .
we continue with the proof of lemma [ lem.rgg2 ] . let @xmath381 ; and let @xmath382 satisfy @xmath300 and @xmath383 . as indicated earlier , we shall show that whp @xmath14 contains a copy @xmath384 of @xmath38 such that no edge of @xmath384 is in a triangle in @xmath14 , and so @xmath79 .
choose @xmath385 sufficiently small that @xmath386 , @xmath387 is contained in the ball centred on @xmath336 with radius @xmath261 , and @xmath388 .
scale up by a factor @xmath389 , and use the notation @xmath390 , @xmath391 , @xmath392 to refer to the rescaled case .
note that @xmath392 is contained in @xmath312 ( by our assumption on @xmath393 ) . put small open balls of radius @xmath394 around the five extreme points @xmath390 of the pentagon , and note that these small balls are all disjoint ( since @xmath395 ) . if @xmath131 and @xmath141 are points in the small balls at non - adjacent vertices @xmath390 and @xmath396 then @xmath397 ( since @xmath398 ) .
if @xmath131 and @xmath141 are points in the small balls at adjacent vertices @xmath390 and @xmath399 ( where @xmath400 means @xmath401 ) then @xmath402 ( since @xmath403 ) ; and if @xmath404 then either @xmath405 or similarly @xmath406 , so we do not get triangles involving a point @xmath407
. now rescale by @xmath8 , and call the rescaled controlled region @xmath408 .
note that the area of @xmath408 is @xmath409 .
if exactly one poisson point @xmath131 lies in each rescaled small ball and there are no other such points in @xmath408 then we have a copy of @xmath38 as desired . setting @xmath410
, the probability @xmath411 of this happening satisfies @xmath412 since events within different cells are independent , the probability @xmath413 that @xmath414 has no @xmath38 as desired satisfies @xmath415 observe that @xmath416 if @xmath417 then @xmath418 and if @xmath419 then @xmath420 thus in both cases @xmath421 .
it follows that the failure probability in the original @xmath9 model is @xmath422 , as required .
_ part _ ( 6 ) ( of theorem [ thm.rggsummary ] ) .
the next lemma proves part ( 6 ) , and thus completes the proof of theorem [ thm.rggsummary ] .
[ lem.rgg ] let @xmath42 with @xmath76 . then @xmath23 whp
clearly @xmath14 has an edge whp , and so @xmath295 whp .
hence , we only need to show that @xmath223 whp . as in the proof of proposition [ thm.detr3 ] ,
start with a hexagonal packing in the plane , as in figure 1 , formed from the voronoi cells ( with vertical left and right sides ) of the unit triangular lattice @xmath154 ( where the edges have unit length ) .
the hexagonal cells have area @xmath423 and diameter @xmath164
. now rescale by the factor @xmath424 for some suitably small @xmath425 . as a result
, each cell has area @xmath426 and diameter @xmath427 .
( for orientation , note that the lower bound on @xmath8 is ( for small @xmath428 ) more than @xmath429 . ) by shrinking slightly in the @xmath345 and the @xmath84 directions , we may ensure that the left and right sides of the square @xmath47 lie along vertical sides of cells ( more precisely , we may ensure that , as we move up the left side of the square , every second internal cell has its vertical left boundary along the side of the square , and every second one is bisected by the side of the square ; and similarly for the right side of the square ) , and each cell which meets a horizontal side of @xmath47 is at least half inside @xmath47 .
we then obtain ( at least for large @xmath0 ) a partition of the square @xmath47 such that each cell has diameter at most @xmath430 , each ` internal ' cell not meeting the boundary has area at least @xmath431 , and each ` boundary ' cell meeting the boundary has area at least @xmath432 .
there are @xmath433 internal cells and @xmath434 boundary cells .
the probability that a given internal cell contains at most one point in its interior is at most @xmath435 since there are @xmath433 internal cells , the expected number of such cells is @xmath436 .
similarly , the probability that a given boundary cell contains at most one point in its interior is at most @xmath437 ; and since there are @xmath434 boundary cells , the expected number of such cells is @xmath438 .
it follows from markov s inequality that whp all cells have at least two points in their interior .
it suffices now to show ( deterministically ) that for each set of points in @xmath47 with at least two in the interior of each cell , the corresponding graph @xmath14 has @xmath223 .
to do this , we colour the vertices of @xmath14 arbitrarily as long as both colours are used in every cell : we shall show that this gives a proper clique - colouring .
observe that @xmath14 has no isolated vertices since @xmath8 is more than the diameter @xmath430 of a cell ( indeed , @xmath439 and so
assuming @xmath0 is large the minimum degree may be shown to be at least 95 , since in the triangular lattice there are 48 lattice points @xmath440 within graph distance 7 of @xmath166 , and thus within euclidean distance @xmath441 , so each point in each of these cells is at euclidean distance @xmath442 from each point in the cell corresponding to @xmath166 ) .
consider any maximal clique @xmath260 in @xmath14 with corresponding euclidean diameter @xmath443 ( so @xmath444 ) , and suppose that @xmath443 is attained for the euclidean distance between the points @xmath445 and @xmath446 corresponding to vertices @xmath207 and @xmath208 in @xmath260 .
let @xmath141 be the midpoint of the line joining @xmath445 and @xmath446 , and let the cell @xmath173 contain @xmath141 .
since for each vertex @xmath199 in the clique @xmath260 , the corresponding point @xmath202 is at distance at most @xmath443 from both @xmath445 and @xmath446 , it follows that @xmath202 is at distance at most @xmath447 from @xmath141 . hence if @xmath448 then every point of the cell @xmath173 is at distance at most @xmath8 from all points of @xmath260 . since @xmath260 is maximal , all vertices corresponding to points that belong to the cell must be in @xmath260 , and so @xmath260 is not monochromatic . since @xmath449 , the desired inequality holds as long as @xmath450 , which is equivalent to @xmath451 . but
@xmath452 to 6 decimal places .
thus , by choosing @xmath385 sufficiently small , we see that it suffices to have @xmath453 .
we have completed the proof of theorem [ thm.rggsummary ] .
it remains to prove proposition [ prop.rgg4 ] .
the first part of that result follows directly from lemmas [ lem.detsmall ] and [ lem.comps ] , since we already know that @xmath79 whp , and the latter lemma shows that whp @xmath14 has no components with more than 10 vertices .
the second part will follow easily from the next lemma , by considering a connected geometric graph @xmath89 such that @xmath454 .
[ lem.allcomps ] let @xmath455 and let @xmath89 be any given connected geometric graph with @xmath370 vertices .
suppose that @xmath456 and @xmath457 .
then for @xmath42 , whp @xmath14 has a component isomorphic to @xmath89 . if @xmath458 and @xmath286 then whp @xmath14 has a component @xmath89 by lemma [ lem.comps ] . to handle larger values of @xmath8 , we now work in the poisson model @xmath302 .
assume from now on that @xmath459 say ( and still @xmath457 ) .
fix distinct points @xmath460 such that , for each distinct @xmath188 and @xmath461 , @xmath462
if @xmath463 and @xmath464 if @xmath465 . thus the unit disk graph generated by these points is @xmath89 .
let @xmath466 , let @xmath467 , and let @xmath468 .
let @xmath469 .
put a small open ball @xmath470 of radius @xmath471 around each point @xmath6 .
observe that these balls are pairwise disjoint , and if @xmath472 for each @xmath188 then @xmath473 yield the same geometric graph @xmath89 .
let @xmath474 be the set of points within distance 1 of the points @xmath6 ( so @xmath474 is the union of the balls @xmath475 ) , and let @xmath476 be the area of @xmath474 .
observe that @xmath477 since @xmath455 and @xmath89 has an edge .
let @xmath173 be the set of points within distance @xmath478 of the @xmath6 , and let @xmath173 have area @xmath309 .
let @xmath479 . if @xmath471 is chosen sufficiently small then @xmath480 ; assume we have done this .
if each ball @xmath470 contains exactly one poisson point and there are no other such points in @xmath173 , then we have a copy of @xmath89 forming a component of @xmath14 .
now scale by @xmath8 , note that we can pack @xmath481 disjoint copies of the configuration in @xmath304 , and we may argue as in the proof of lemma [ lem.rgg2 ] , as follows .
set @xmath482 .
let @xmath411 be the probability that each small ball contains exactly one poisson point and there are no such points where they should not be .
then @xmath483 since events within different cells are independent , for some constant @xmath484 the probability @xmath413 that @xmath414 has no component @xmath89 satisfies @xmath485 now , for @xmath486 , we have @xmath487 , and so @xmath488 thus , since @xmath489 , we have @xmath421 .
it follows that the failure probability in the original @xmath9 model is @xmath422 , as required .
let us pick up a few points for further thought . 1 .
recall that @xmath37 is the maximum value of @xmath15 over geometric graphs @xmath14 in the plane , and we saw that @xmath490 .
can we improve either bound ? + observe that if a geometric graph @xmath14 is triangle - free then @xmath14 is planar ( if in the embedding of a geometric graph two edges cross , then this induces a triangle in @xmath14 , see for example @xcite ) and so @xmath491 by grtzsch s theorem .
we saw in lemma [ lem.detsmall ] that @xmath225 for all graphs with at most 10 vertices .
the grtzsch graph showed that this bound does not extend to @xmath492 ( see also equation ( [ eqn.chicmax4 ] ) , and point ( 4 ) below ) .
but the grtzsch graph is not a geometric graph in the plane , so perhaps the upper bound 3 extends to larger values @xmath0 when we restrict our attention to geometric graphs ?
any extension for geometric graphs would lead to an improvement in proposition [ prop.rgg4 ] part ( 1 ) . if it turns out that @xmath493 , then theorem [ thm.rggsummary ] is tighter than it currently seems , and proposition [ prop.rgg4 ] is redundant .
if @xmath494 then it would be interesting to refine part ( 5 ) of theorem [ thm.rggsummary ] .
2 . more generally
, can we say more about @xmath36 ?
we saw in proposition [ thm.detr3 ] that @xmath495 : can we improve this upper bound ? can we find a geometric graph in @xmath152 with @xmath496 ?
we have seen that @xmath36 is at most @xmath497 and is @xmath498 as @xmath41 .
can we improve these bounds ?
3 . in the light of the last two parts of theorem [ thm.rggsummary ] ( and proposition [ prop.rgg4 ] ) , it is natural to ask if there is a constant @xmath499 , where @xmath500 , such that for @xmath42 and any @xmath53 , we have whp @xmath501 4 . recall that @xmath211 is the maximum value of @xmath15 over all @xmath0-vertex graphs .
trivially @xmath502 .
we saw in lemma [ lem.detsmall ] and equation ( [ eqn.chicmax4 ] ) that @xmath503 what about larger values of @xmath0 ? + now consider asymptotic behaviour .
we saw in equation ( [ eqn.chicmaxn ] ) that @xmath504 . on the other hand ,
we claim that @xmath505 we may see this as follows .
_ repeatedly _ , pick greedily a maximal independent set , give all the vertices in the set the same fresh colour and remove them , _ until _ we find a maximal independent set @xmath506 of size less than @xmath507 .
such a set @xmath506 is a dominating set in the remaining graph @xmath89 , so @xmath508 , see @xcite . thus if @xmath89 has @xmath370 vertices , then at most @xmath509 further colours are needed .
+ in the first phase we use at most @xmath510 colours .
if @xmath511 then we use at most @xmath512 colours in total .
if @xmath513 then we use at most @xmath514 colours , and hence @xmath515 .
this proves the claim ( [ eqn.chicmaxub ] ) . +
we know that @xmath211 is @xmath516 and @xmath517 .
can we say more about the asymptotic behaviour of @xmath211 ?
see also @xcite , and problem 1 there in particular .
our upper bound on @xmath15 gives an upper bound on the _ clique transversal number _ @xmath518 , which is defined to be the minimum size of a set @xmath13 of vertices which meets all maximal cliques ( ignoring isolated vertices ) . for each @xmath0-vertex graph @xmath14 , since the maximum size of a set of vertices containing no maximal clique is at least @xmath519 , we have @xmath520 the result noted above that @xmath521 yields @xmath522 , which may be compared with the best known bound @xmath523 ( see @xcite ) . it is not likely to be easy to improve our upper bound by say a factor @xmath221 to @xmath524 , since that would strictly improve the upper bound on @xmath525 ( to @xmath526 ) .
consider again high dimensions .
let @xmath527 be the least value @xmath28 such that every graph with @xmath0 vertices is geometric in @xmath27 .
then @xmath528 by proposition [ prop.allgeom ] .
we claim that @xmath529 for , let @xmath385 , and let @xmath530 for @xmath531 .
suppose that @xmath532 for arbitrarily large values @xmath0 .
we shall obtain a contradiction .
+ note first that @xmath533 is increasing for @xmath534 .
define @xmath535 . clearly @xmath536 for @xmath537 .
now @xmath538 as @xmath41 : hence , for some constant @xmath539 we have @xmath540 for each @xmath541 .
+ let @xmath542 be arbitrarily large .
there exists @xmath543 such that @xmath544 for some @xmath0 with @xmath532 .
now @xmath545 and so @xmath546 but by ( [ eqn.chicmaxn ] ) , for some constant @xmath484 we have @xmath547 hence , @xmath548 but this contradicts the upper bound on @xmath36 in proposition [ thm.detrd ] , and so we have established the claim ( [ eqn.dstar ] ) .
+ now we know that @xmath549 and @xmath528 .
our bounds are wide apart .
what more can be said about @xmath527 ? | a clique colouring of a graph is a colouring of the vertices such that no maximal clique is monochromatic ( ignoring isolated vertices ) .
the least number of colours in such a colouring is the clique chromatic number . given @xmath0 points @xmath1 in the plane , and a threshold @xmath2
, the corresponding geometric graph has vertex set @xmath3 , and distinct @xmath4 and @xmath5 are adjacent when the euclidean distance between @xmath6 and @xmath7 is at most @xmath8 .
we investigate the clique chromatic number of such graphs .
we first show that the clique chromatic number is at most 9 for any geometric graph in the plane , and briefly consider geometric graphs in higher dimensions .
then we study the asymptotic behaviour of the clique chromatic number for the random geometric graph @xmath9 in the plane , where @xmath0 random points are independently and uniformly distributed in a suitable square .
we see that as @xmath8 increases from 0 , with high probability the clique chromatic number is 1 for very small @xmath8 , then 2 for small @xmath8 , then at least 3 for larger @xmath8 , and finally drops back to 2 .
[ section ] [ theorem]lemma [ theorem]definition [ theorem]conjecture [ theorem]proposition [ theorem]algorithm [ theorem]corollary [ theorem]observation [ theorem]open problem [ theorem]remark |
it has long been known that the major axis of central cluster galaxies tends to be aligned with the major axis of the cluster galaxy distribution and to point towards similar structures on scales of typically a few tens of mpc ( e.g. , @xcite ; trevese , cirimele & flin 1992 ; fuller , west & bridges 1999 , @xcite and references therein ) .
the most likely mechanism for the origin of this ` alignment effect ' invokes _ collimated infall _ of galaxies along the filaments that make up the large scale structure in which clusters are embedded and from which clusters grow @xcite , suggesting a connection between the growth of the clusters and the formation of their massive dominant galaxies ( e.g. , @xcite ) . it is unclear whether fainter cluster galaxies should show the alignments that are expected ( and observed ) for the first - ranked members .
clusters and galaxies are believed to form through a process of hierarchical accretion . in these models ,
predictions for isotropy ( or deviations ) of galaxy position angles are not well defined ( west , villumsen & dekel 1991 ) but the consensus seems to be that any primordial alignments would be quickly erased by dynamical interactions @xcite .
conversely , it is expected that groups and clusters show preferential alignments in the direction of the last major accretion episode @xcite .
searches for alignments among the fainter cluster galaxies have generally returned ambiguous results , with some positive detections @xcite as well as null results ( @xcite ; aryal , kandel & sauer 2006 ) .
however , it is possible that the presence of anisotropy in galaxy position angles can only be detected in less evolved clusters @xcite , while it is also conceivable that only near neighbor galaxies are aligned with each other because they exert reciprocal tidal torques @xcite , or that radial alignments are induced by the cluster tidal field @xcite . with this in mind , we re - examine the issue of isotropy in galaxy position angles in the coma cluster .
we use a unique multicolor optical - infrared ccd mosaic of the inner 750 @xmath0 mpc of this cluster and , within this region , a complete ( magnitude - limited to @xmath1 ) sample of 111 spectroscopic members plus 77 additional fainter members .
most previous work has relied on photographic plates and identified cluster members statistically or via the red sequence .
the results we present in the following are most consistent with an isotropic distribution for galaxy position angles , other than for the two brighter members .
we then explore the implications of these findings for the formation of clusters and their members , particularly as regards the brighter central cluster galaxies .
we use the @xmath2 images of the core of the coma cluster and the catalog published by @xcite , which includes photometry and a compilation of spectroscopy from the literature .
the spectroscopic sample of coma galaxies is complete to @xmath3 and less so for fainter magnitudes . galaxy major axis position angles and ellipticities obtained independently from @xmath4 , @xmath5 and @xmath6 images were compared and found to be similar ; consequently only the @xmath5-band images were used for further analysis .
we verified the sextractor position angles by isophotal modelling of the galaxies using iraf ellipse and also by comparing with the isophotal position angles from the sdss @xcite .
the values produced by sextractor are in good agreement with the measurements from iraf ellipse and the sdss position angles , except for a few objects with small ellipticities .
we tested the observed distribution of position angles against a uniform distribution using the kuiper test .
figure 1 shows the results of this comparison : we find no significant deviation from isotropy .
this holds if we limit our sample to galaxies with larger ellipticity or to the brighter members ( e.g. , @xcite ) .
subsamples of galaxies with ellipticities @xmath7 and @xmath8 were examined , containing 128 , 94 and 61 galaxies , respectively , but in all cases no preferred orientations were found . likewise culling the sample of galaxies to those brighter than @xmath9 and @xmath10 ( 10 , 33 and 58 galaxies , respectively ) also yielded a null detection of position angle anisotropies .
+ + while the above constitutes strong evidence that there are no large - scale galaxy alignments ( other than for the brightest cluster galaxies ) , we now consider whether there are alignments between nearest neighbors ( in projection ) due to reciprocal tidal torques @xcite , or due to the presence of substructure whose members are aligned with each other @xcite .
the result of this test is shown in figure 2 , where we find no significant deviation from isotropy .
neither can we confirm the detection of radial galaxy alignments by @xcite .
the above results are perhaps understandable if we consider that there is no reason to assume that cluster members that are close in projection are actually physically close , because clusters have intrinsic depth and an extended radial velocity distribution .
therefore , any true alignment signal between neighbors will be diluted by the overwhelming number of galaxy ` asterisms ' .
the results of the analysis presented above are consistent with random orientations of the position angles of coma galaxies , with the possible exception of the two brightest cluster members , ngc 4889 and ngc 4874 , whose major axis position angles are quite similar to that of the cluster s galaxy and x - ray gas distributions @xcite .
the line connecting the two galaxies is also aligned with the galaxy and x - ray gas distributions , as also found by @xcite .
while this is consistent with the null results found for some clusters , there are other objects where significant alignments are detected .
possibly the clearest case is abell 521 @xcite , which shows detectable alignments out to 5 @xmath0 mpc .
this cluster is known to be dynamically young and to show considerable substructure @xcite .
coma , on the other hand , is not a fully relaxed cluster and shows substructure in the x - ray gas and in the position - velocity distribution of its members ( e.g. , @xcite ) but it is clearly ( comparatively ) more evolved than either a521 or virgo , where also there is evidence of a non - isotropic distribution of galaxy alignments @xcite . as suggested by @xcite
, the lack of alignments in coma may be due to its more advanced state of dynamical evolution with respect to other clusters ( see , e.g , @xcite ) . in order to test this
, we considered galaxy alignments in the filaments that connect coma to the large scale structure towards a1367 and a2197/a2199 ( figure 3 ) .
we retrieved these data from the sdss and tested for position angle anisotropy .
all galaxies with available data were used , irrespective of their magnitudes .
because complete sdss spectroscopy is not available for this region , redshifts were taken from the nasa extragalactic database and the sample of galaxies restricted to those with velocities in the range 4000 - 10,000 km / s expected for coma and its environs . [
cols="^,^,^ " , ] here too , however , we find no evidence for any significant deviations from isotropy for filament galaxies ( figure 4 ) .
this suggests either that these galaxies formed without any preferred orientations or that any alignments which might have existed are quickly erased even in low density regions . in this context , it is not surprising that alignments can be detected more clearly in a521 and virgo , two clusters that appear to be have recently formed from the coalescence of numerous small groups .
the main inference we can draw from this study is that the first - ranked cluster galaxies are special in yet one more way . unlike all other cluster galaxies they appear to have undergone a peculiar formation mechanism , related to collimated infall and the growth of the cluster from the surrounding large scale structure , and resulting in the observed alignment effects
all other galaxies appear to have had a more random merger history .
while fainter galaxies may have been accreted to the cluster at later epochs , thus explaining the lack of alignments , the brightest cluster member may have been the kernel around which clusters originally formed .
l. torlina was supported by an australian gemini undergraduate summer studentship , funded by the australian research council .
she thanks gemini observatory for its hospitality during her stay in chile .
we thank the referee , manolis plionis , for his very quick review of this paper and for helpful comments that improved its clarity .
adami , c. , biviano , a. , durret , f. , mazure , a. 2005 , , 443 , 17 aryal , b. & saurer , w. 2005 , , 360 , 125 aryal , b. & saurer , w. 2006 , , 366 , 438 aryal , b. , kandel , s. m. & saurer , w. 2006 , , 458 , 357 bertin , e. & arnouts , s. 1996 , , 117 , 393 binggeli , b. 1982 , , 107 , 338 coutts , a. 1996 , , 278 , 87 dubinski , j. 1998 , , 502 , 141 eisenhardt , p. r. , de propris , r. , gonzales , a. , stanford , s. a. , wang , m. c. , dickinson , m. e. 2007 , , 169 , in press faltenbacher , a. , allgood , b. , gottlber , s. , yepes , g. , hoffman , y. 2005 , , 362 , 1099 ferrari , c. , arnaud , m. , ettori , s. , maurogordato , s. , rho , j. 2006 , , 446 , 417 fuller , t. m. , west , m. j. & bridges , t. j. 1999 , , 519 , 22 kitzbichler , m. g. & saurer , w. 2003 , , 590 , l9 pereira , m. j. & kuhn , j. r. 2005 , , 627 , l21 plionis , m. & basilakos , s. 2002 , , 329 , l47 plionis , m. , benoist , c. , maurogordato , s. , ferrari , c. , basilakos , s. 2003 , , 594 , 144 stoughton , c. et al . 2002 , , 123 , 485 strazzullo , v. , paolillo , m. , longo , g. , puddu , e. , djorgovski , s. g. , de carvalho , r. r. , gal , r. r. 2005 , , 359 , 191 trevese , d. , cirimele , g. & flin , p. 1992
, , 104 , 935 west , m. j. 1989 , , 344 , 535 west , m. j. , villumsen , j. v. & dekel , a. 1991 , , 369 , 287 west , m. j. 1994 , , 268 , 79 west , m. j. 1998 , in _ uncovering coma berenices : a new look at an old cluster _ , ed .
a. mazure , f. combes , f. durret , d. gerbal ( singapore : world scientific ) , p. 36
west , m. j. & blakeslee , j. p. 2000 , , 543
, l27 york , d. g. et al .
2000 , , 120 , 1579 | we have examined the orientations of early - type galaxies in the coma cluster to see whether the well - established tendency for brightest cluster galaxies to share the same major axis orientation as their host cluster also extends to the rest of the galaxy population .
we find no evidence of any preferential orientations of galaxies within coma or its surroundings .
the implications of this result for theories of the formation of clusters and galaxies ( particularly the first - ranked members ) are discussed . |
throughout this paper we assume that @xmath5 is injective ( one - to - one ) .
such functions are generic in the space of real functions on @xmath6 .
one may associate a _ gradient flow _ of @xmath5 on the graph @xmath7 , as a map @xmath8 which maps a subset of vertices to its immediate neighbors with lower @xmath5 values .
more precisely , given @xmath9 , define the neighbor set of @xmath10 with lower energy @xmath11 and @xmath12 for any @xmath13 , we define @xmath14 let @xmath15 , _
etc_. we say that @xmath16 is _ reachable _ from @xmath10 , denoted by @xmath17 or @xmath18 , if @xmath19 for some @xmath20 , _ i.e. _ , we can find an energy decreasing path from @xmath10 to @xmath16 .
note that our construction of the gradient flow is related to , but different from the gradient network @xcite , in which each node is only connected to its neighbor with the lowest energy ( _ i.e. _ the neighbor in the steepest descent direction ) .
we also remark that the gradient flow can be viewed as a `` zero temperature '' limit of the stochastic gradient flow introduced in @xcite in the study of network communities .
the _ local minima _ of @xmath5 are those vertices whose @xmath5 value is no larger than the values of its neighbors .
@xmath21 in other words , the set of local minima are precisely the maximal vertex set of _ fixed points _ of the gradient flow @xmath22 . given a local minimum @xmath9 , its _ attraction basin _
is defined to be : @xmath23 these are the points that reach the local minimum @xmath10 but not any other local minima . _ boundary or separatrix _ consists of those nodes which can reach more than one local minimum following the gradient flow @xmath24 it is clear by definition that we have the non - overlapping decomposition @xmath25 our next task is to classify the nodes in @xmath26 .
we do so according to their role in the pathways connecting the different local minima .
in particular , _
index-@xmath27 critical nodes ( saddles ) _ are defined as the maxima on _ local minimum energy paths _ connecting different local minima .
clearly such a definition relies on the notion of _ local minimal energy paths _ , which depends on the topology of the path space .
given two local minima , we examine all the paths connecting them . if a path @xmath28 can be deformed by the gradient flow to another path @xmath29 , we say that @xmath28 is _ deformable _ to @xmath29 .
the _ local minimum energy paths _ are paths which can not be deformed by the gradient flow . to be more precise , given two points @xmath30 , we define a _ path _ from @xmath31 to @xmath32 as @xmath33 such that @xmath34 , @xmath35 , and @xmath36 for @xmath37 .
we denote the collection of paths from @xmath31 to @xmath32 as @xmath38 .
we note the following elementary lemma , whose proof is obvious .
[ lem : path ] let @xmath39 , we can then find a path @xmath40 from @xmath10 to @xmath16 such that @xmath41 for @xmath42 .
given two paths @xmath43 , we say @xmath28 is _ deformable to _ @xmath29 , if there is a map @xmath44 , such that * ( reaching ) every node in @xmath28 reaches some nodes in @xmath29 , _
i.e. _ for any @xmath45 , @xmath46 is not empty and for each @xmath47 , @xmath18 ; * ( onto ) every node in @xmath29 is reachable from @xmath28 , _ i.e. _ for any @xmath48 , there exists @xmath49 , so that @xmath50 , or equivalently , @xmath51 let @xmath52 be two local minima .
we call a path @xmath53 _ local minimum energy path _ , if it is not deformable to any other path in @xmath38 .
we define the energy of a path the maximal energy traversed by the path , _
i.e. _ @xmath54 . from the definition ,
if @xmath29 is deformable to @xmath28 , we have @xmath55 , so in terms of energy barrier , @xmath28 is a more preferable path than @xmath29 .
given a local minimum energy path , we call the node of maximal energy on the path an _ index-@xmath27 critical node_. the set of all index-@xmath27 critical nodes is denoted by @xmath56
. we will also call local minima _ index-@xmath2 critical nodes _ , and hence the notation @xmath57 .
the following fact gives a characterization of index-@xmath27 critical nodes .
the proof can be found in the si .
[ thm : lower ] all local minima in @xmath26 are index-@xmath27 critical nodes .
the other index-@xmath27 critical nodes will reach one of the local minima in @xmath26 by the gradient flow .
we call the index-@xmath27 critical nodes that are also local minima in @xmath26 the nondegenerate index-@xmath27 critical nodes , the set of which will be denoted as @xmath58 .
the other index-@xmath27 critical nodes are called degenerate .
not every index-@xmath27 critical node is a local minimum in @xmath26 , for example in some cluster trees ( see figure [ fig : tree ] ) .
critical node , where node 5 on top of the tree is a degenerate index-@xmath27 saddle while nodes 3 is a nongenerate index-@xmath27 saddle .
right : an example of both degenerate and non - degenerate critical node , where node 7 on top of the tree is a degenerate index-1 saddle as it lies on the minimum energy path connecting local minima 1 ( or 2 ) and 3 ( or 4 ) , and as well a non - degenerate index-2 saddle as it is on the minimum energy path linking index-1 saddles 5 and 6.,title="fig:",scaledwidth=40.0% ] critical node , where node 5 on top of the tree is a degenerate index-@xmath27 saddle while nodes 3 is a nongenerate index-@xmath27 saddle .
right : an example of both degenerate and non - degenerate critical node , where node 7 on top of the tree is a degenerate index-1 saddle as it lies on the minimum energy path connecting local minima 1 ( or 2 ) and 3 ( or 4 ) , and as well a non - degenerate index-2 saddle as it is on the minimum energy path linking index-1 saddles 5 and 6.,title="fig:",scaledwidth=55.0% ] the procedure presented above can be extended to define higher index critical nodes . to define index-@xmath59 critical nodes
, we consider the subgraph with nodes in @xmath26 and edges restricted on this subset , denoted by @xmath60 .
the gradient flow @xmath61 on @xmath26 is defined similarly as for @xmath62 .
we define the attraction basins for @xmath63 as @xmath64 note that for any nondegenerate index-@xmath27 critical node , the attraction basin is nonempty . while for a degenerate index-@xmath27 critical node , the attraction basin is an empty set .
this explains the notion `` degenerate '' for the critical nodes that are not local minima in @xmath26 .
we define the boundary set as @xmath65 as shown in proposition [ thm : lower ] , all local minima on @xmath26 are in @xmath56 .
therefore , we have the decomposition @xmath66 analogously , we define
_ index-@xmath59 critical nodes _ as the maxima on local minimum energy paths connecting different nondegenerate index-@xmath27 critical nodes .
it is clear that index-@xmath59 critical nodes , if exist , must be in @xmath67 .
we remark that under our definition , a degenerate index-@xmath27 critical node can also be an index-@xmath59 critical node , as shown in figure .
this ambiguity is actually quite natural from the network point of view , as these points play multiple roles in the structure of the network .
the degenerate index-@xmath27 critical node can lie either in the basin of a nondegenerate critical node or link together two different nondegenerate critical nodes .
higher index critical nodes can be defined recursively through further decomposition of @xmath67 .
classification for high index critical points can be done following similar arguments as above .
combining these , we obtain : [ thm : decomp ] @xmath6 admits the following decomposition @xmath68 where @xmath69 here @xmath70 is the attraction basin of local minima restricted on the @xmath71-th boundary set @xmath72 and @xmath73 is the set of nondegenerate index-@xmath3 critical nodes .
the theorem gives us a hierarchical representation of the network associated to the energy landscape .
it actually leads to a hypergraph representation whose hypernodes are made up of critical nodes with their attraction basins .
the landscape introduced above can be naturally formulated in terms of a flooding procedure , from low to high values of the height function @xmath74 .
flooding starts from local minima , followed by the attraction basins .
once the relevant index-@xmath27 saddle is passed , basins of local minima are merged together .
this procedure then continues on to critical points of higher indices .
more precisely , this procedure can be described in terms of persistent homology .
persistent homology , firstly proposed by @xcite and developed afterwards largely in @xcite , is an algebraic tool for computing the betti numbers and homology groups of a simplicial complex when its faces are added sequentially . to work with persistent homology ,
we extend the graph @xmath7 into a simplicial complex up to dimension 2 , and also define a filtration which consists of such simplicial complexes , in a spirit close to @xcite for pl - manifolds .
an abstract simplicial complex @xmath75 is a collection of subsets of @xmath6 , which is closed under deletion or inclusion , i.e. if @xmath76 , then @xmath77 for any @xmath78 .
we define _ the flooding complex of network @xmath7 associated with the function @xmath5 _ , @xmath79 as follows : * @xmath2-simplex : the vertex set @xmath6 ; * @xmath27-simplex : the vertex pairs @xmath80 that @xmath81 , _
i.e. _ , @xmath82 for some @xmath3 ; * @xmath59-simplex : collections of triangles @xmath83 , such that @xmath84 and @xmath85 .
one can similarly extend the definition above to general @xmath3-simplex .
however for our purpose it suffices to define up to dimension @xmath59 simplices .
a filtration of flooding complex @xmath86 is a nested family @xmath87 with @xmath88 which respects the order of deletion or inclusion in @xmath86 , i.e. if @xmath89 and @xmath90 then @xmath91 .
assume that @xmath1 is injective or one - to - one , which is generically the case . by taking the maximum over vertices
, one can extend @xmath5 from the vertex set to simplicies , and thus to the simplicial complex @xmath86 .
for a simplex @xmath92 let @xmath93 .
this implies that a face s @xmath5-value is always no more than that of its associated simplex , i.e. @xmath94 . a filtration @xmath95 respecting the order of @xmath5 can be defined in the following way : 1 .
@xmath96 ; 2 .
@xmath97 , _ i.e. _ there is precisely one node being added into the filtration for each step ; 3 .
@xmath98 , where @xmath99 , _ i.e.
_ when a node is added into the filtration , all the simplices of the same energy are added into the filtration simultaneously . note that under this construction
, @xmath100 consists of the global minimum of @xmath101 . in this construction , we consider the filtration corresponding to the flooding procedure from low to high @xmath5 values .
the change of betti numbers identifies the index-@xmath2 and index-@xmath27 critical nodes .
once the filtration is defined , persistent homology computes the betti numbers of the simplicial complex in @xmath102 for each @xmath103 , and draws the barcodes of betti number versus the @xmath104 or @xmath5 values , _
e.g. _ using jplex toolbox .
the proof of the following theorem is in the si .
[ thm : persist1]consider the filtration @xmath105 . for all @xmath106 , @xmath107 contains an index-@xmath2 critical node if and only if @xmath108 increases from @xmath109 to @xmath110 ; @xmath107 contains an index-@xmath27 critical node if and only if either @xmath108 decreases or @xmath111 increases from @xmath109 to @xmath110 . to find higher index saddles ,
we restrict on the subgraph @xmath112 where @xmath113 and @xmath114 consists of edges restricted on @xmath115
. we can analogously construct the filtration corresponds to the flooding procedure @xmath116 on the subgraph @xmath117 .
similar identification holds for higher index saddles .
[ thm : persistk]consider the filtration @xmath118 on subgraph @xmath117 for @xmath119 . for all @xmath120 such that @xmath121 contains an index-@xmath3 critical node if either @xmath108 decreases or @xmath111 increases from @xmath122 to @xmath123 .
clearly our characterization of high order critical nodes above only exploits simplicial complex up to dimension @xmath59 , whose persistent homology computation is recently improved to be of complexity @xmath124 @xcite with @xmath125 the total number of simplices and @xmath126 the number of nodes .
such a complexity does not suffer the curse of dimensionality as the computation of high order betti numbers in general .
as we know from proposition [ thm : lower ] that nondegenerate critical nodes are actually local minimum in sub - graphs @xmath117 , this leads to an efficient algorithm for finding nondegenerate critical nodes .
in fact , all the examples shown in this paper have only nondegenerate critical nodes and thus can be found efficiently using this algorithm .
given an injective function @xmath5 on the vertices , we obtain the local mimina and nondegenerate index-@xmath3 saddles using algorithm [ alg : nondegenerate ] . the bottleneck in this algorithm
is in finding the attraction basins of local minima , whose complexity can be @xmath127 where @xmath126 is the number of vertices and @xmath128 is the maximum degree a node has .
the total complexity is @xmath129 where @xmath130 is the maximum index of critical points .
the algorithm is much faster than the previous algorithm for finding all critical nodes .
sort the nodes according to @xmath5 in increasing order ; set @xmath131 ; find neighbors of @xmath10 with lower energy , @xmath132 ; add @xmath10 to @xmath73 and set the color of @xmath10 as its node index ; set the color of @xmath10 as the single color ; leave the color of @xmath10 as blank ; : + ( 1 ) local minima @xmath73 as nondegenerate critical nodes ; + ( 2 ) attraction basins @xmath133 ( @xmath134 ) as color components ; + ( 3 ) boundary @xmath135 as the blank nodes ; set @xmath136 where @xmath137 are edges restricted on @xmath135 ;
zachary s karate club network @xcite consists of 34 nodes , representing 34 members in a karate club with node 1 being the instructor and node 34 being the president ( figure [ fig : karate ] ) .
an edge between two nodes means that the two members join some common activities beyond the normal club classes and meetings .
conflicts broke out between the instructor and the president when the instructor sought to raise the fee and the president opposed the proposal .
the club eventually split into two , one formed by the president ( blue nodes in figure [ fig : karate](a ) ) and another one led by the instructor ( red nodes in figure [ fig : karate](a ) ) .
a lot of information about this fission can be disclosed by looking at the graph structure of this social network .
+ + versus @xmath104 .
node 34 with the lowest energy is added at @xmath138 which creates a connected component which never disappears .
node 1 with the second smallest energy is added at @xmath139 which creates a new connected component disappeared when index-1 saddle 3 is added at @xmath140 .
bottom : @xmath111 versus @xmath104 .
the loop is created by index-@xmath27 saddle 32 added at @xmath141 then cancelled by index-@xmath59 saddle 29 at @xmath142.,title="fig : " ] versus @xmath104 .
node 34 with the lowest energy is added at @xmath138 which creates a connected component which never disappears .
node 1 with the second smallest energy is added at @xmath139 which creates a new connected component disappeared when index-1 saddle 3 is added at @xmath140 .
bottom : @xmath111 versus @xmath104 .
the loop is created by index-@xmath27 saddle 32 added at @xmath141 then cancelled by index-@xmath59 saddle 29 at @xmath142.,title="fig : " ] let @xmath143 be the degree of node @xmath144 , and define @xmath145 . to avoid the same degree between two nodes in neighbor , a small enough random perturbation is added such that @xmath146 is injective .
figure [ fig : karate](b ) shows the gradient flow of @xmath5 .
the arrows on the edges point from low degree nodes to high degree ones . note that nodes 24 and 25 both have degree 3 , hence a small random perturbation is added resulting in the arrow from 25 to 26 . the same is done for nodes 5 and 11 .
figure [ fig : karate](c ) shows the node decomposition for karate club network with each color component for a critical node and its attraction basin .
two local minima , nodes 1 and 34 , are in oval shape together with their attraction basins marked in red and blue , respectively .
two index-@xmath27 saddles , nodes 3 and 32 , are yellow and green diamond nodes , whose basins are in yellow ( nodes 3 ) and green ( node 32 ) correspondingly .
node 3 is the lowest energy node connecting the local minima nodes 1 and 34 via a minimum energy path @xmath147 .
node 32 links the two local minima by another local minimum energy path , @xmath148 .
two index-@xmath59 saddles , nodes 25 ( in light blue diamond ) and 29 ( in cyan diamond ) , which connect two index-@xmath27 saddles via two non - deformable minimal energy paths @xmath149 and @xmath150 .
figure [ fig : karate](d ) further depicts a transition path analysis of a markov chain induced on the graph ( see si ) from local minimum node 1 to node 34 , which shows two index-1 saddles capture most of transition currents . figure [ fig : barcode ] shows the barcodes for the flooding complex of this network .
the social network of les misrables , collected by knuth @xcite , consists of 77 main characters in the novel by victor hugo .
the edge weight @xmath151 record the number of co - occurrence of two characters @xmath144 and @xmath152 in the same scene .
thus it is a weighted graph where @xmath153 as the negative logarithmic weighted degree . the original network exhibits a single local ( global ) minimum , valjean , who is the central character as the whole novel was written around his experience . .
two local minima , valjean and enjoras as well as an index-@xmath27 saddle , courfeyrac , are identified.,scaledwidth=70.0% ] however , dropping those edges whose weights are no more than a threshold value ( @xmath154 here ) , there appears a subnetwork which is closely associated with the paris uprising on the 5th and 6th of june 1832 , see figure [ fig : lesmis ] .
the subnetwork consists of two local minima , enjoras and valjean , the former being the leader of the revolutionary students called _ friends of the abc _ , the abaiss .
led by enjolras , its other principal members are courfeyrac , combeferre , and laigle ( nicknamed bossuet ) et al . , who fought and died in the insurrection . among them
is an index-@xmath27 saddle , courfeyrac , a law student and often seen as the heart of the group , who introduced marius to the friends of abc .
marius , a descend of the gillenormands , though badly injured in the battle , was saved by the main character valjean when the barricade fell and married to cosette , the adopted daughter of valjean .
the landscape of this subnetwork highlights these events in the novel .
this application examines the binding of lysine- , arginine- , ornithine - binding ( lao ) protein to its ligand , recently studied in @xcite .
the critical node analysis provides us a concise summary of global structure of networks while preserving important pathways , which enables us to reach a more thorough description than previous approximate analysis . and
index-@xmath59 saddles are shown in diamonds , circular nodes are regular nodes , and rectangular nodes are solvated states @xmath155 .
color components represent the node decomposition .
, scaledwidth=80.0% ] in @xcite a markov state model was constructed with 54 metastable states , using data obtained from molecular dynamics simulation .
more information about these states can be found in si and @xcite .
now we examine the transition network as a weighted directed graph @xmath156 , where @xmath6 consists of 54 nodes , each representing a metastable state , an edge @xmath157 if transitions from node @xmath144 to @xmath152 are observed in simulations with @xmath158 delays ( the implied time scale for approximate markovian behavior ) , and the number of transitions is recorded as the weight @xmath151 .
eleven of the states ( @xmath159 ) are solvated or unbound states .
the binding state is node 10 .
let @xmath160 be the transition probability from state @xmath144 to state @xmath152 .
this defines a markov chain with a unique stationary distribution @xmath161 .
we threshold this graph to an undirected graph by keeping those edges @xmath162 such that @xmath163 , _
i.e. _ average count number is larger than @xmath164 .
one reason for doing this is that small numbers of transitions may be heavily influenced by the noise caused by the way of counting the transition .
note that the mean transition count is about @xmath165 , and the qualitative behavior reported below shows certain stability under the variation of the threshold value .
the energy function is @xmath166 where @xmath161 is the stationary distribution of metastable states .
application of the method above gives rise to a landscape shown in figure [ fig : lao1 ] .
isolated states are dropped in this picture .
colors in this picture illustrate the node decomposition according to theorem [ thm : decomp ] , where each color component represents the attraction basin of a critical node .
below we shall discuss structural properties of these nodes . a complete picture of structural information for all 54 states can be found in si .
there are two major local minima in the landscape , nodes 10 and 18 .
node 10 is the bound state which is the minimum in the most populated energy basin .
its attraction basin is colored in light blue .
nodes 11 ( population @xmath167 ) and 5 ( population @xmath168 ) are two encounter complexes in that basin . in these states ,
the ligand is in or close to the binding site and conformations in this state have a small twist but large opening angles .
the other local minimum is node 18 , a misbound state , where the ligand interacts with the protein outside the binding site and close to the hinge region of two domains of the protein .
state 18 , together with state 4 , 8 , 9 , and 20 , forms a misbound basin marked in red . in these states
, the ligand interacts with the protein from a distance to the binding site .
state 8 and 9 exhibit similar structural properties with a negative twisting angle and a fixed distance to the binding site ( about @xmath169 ) , while state 4 , 18 , and 20 exhibit similar but a different type of structures .
node 19 and 14 are two index-@xmath27 saddles connecting the two basins associated with local minima .
they are metastable intermediate states between misbound and bound states .
but in these saddles ligand interacts with the protein in different ways . in state 19 ( population @xmath170 ) , the ligand is interacting with the protein from one twisting direction ( positive ) and the protein is quite closed . in a contrast , in state 14 ( population @xmath171 ) the ligand is approaching the protein from the opposite twisting direction ( negative ) and the protein is still quite open ( see si ) .
these two saddles actually play different roles in reactive pathways which will be discussed below .
node 2 is an index-@xmath59 saddle , which is essentially a high energy misbound state .
note that high index saddles are unstable with respect to different thresholding values . in the following
we shall focus on index-@xmath27 saddles . for a quantitative analysis on the roles of index-@xmath27 saddles , we conduct two kinds of transition path analysis using transition path theory ( @xcite or see si ) .
first , we study reactive currents from the misbound state 18 to the bound state 10 .
this analysis shows that a majority of flux passes through the saddle 19 .
therefore once the ligand and protein fall in the misbound state 18 , the major pathway to escape and enter the bound state is via saddle 19 .
the other analysis , as was also did in @xcite , studies transition paths from the eleven solvated states marked from 43 to 53 to the bound state 10 .
in particular , we investigate reactive currents from each of the solvated states to the bounded state , respectively .
the results are summarized as follows .
a large part of these details has been ignored in @xcite , since they only examined @xmath172 transition pathways , ignoring the others . 1 .
solvated state 52 lies in the basin of bound state 10 , whence misbound state 18 has little influence on its pathway .
2 . solvated state 53 only passes through index-@xmath27 critical node 19 to enter the bound state 10 , which is heavily influenced by the misbound state 18 .
3 . solvated states @xmath173 lie in the basin of index-1 critical node 14 and enter the bound state 10 directly or via 14 .
they are not much influenced by the misbound state 18 .
other solvated states are in the basin of index-2 critical node 2 .
transition path analysis further shows that misbound state 18 has a stronger influence on them than those in the basin of 14 . in particular state 50
is mostly influenced with near @xmath174 of transition currents trapped by the misbound state 18 . in summary
, the misbound state 18 affects some of the pathways from solvated states to the bound state .
index-1 critical node 14 is a state where ligand starts to interact with protein to enter the encounter complex 11 .
if we can design some mutations to disrupt the stability of this state or even encounter complexes , we may be able to make the binding much more difficult .
finally we note that the critical node analysis here does not rely on the markov model assumption and can thus be applied to the analysis of transition networks in molecular dynamics beyond its markovian time scale .
we have introduced a notion of critical points for network which can be used to reduce a complex network to a coarse - grained representation while preserving structural properties associated with functional gradient flows .
examples have shown that the information obtained this way is of great value in capturing global structure and dynamics of the network , such as diffusive or reactive pathways . moreover
, the critical point analysis leads to a hierarchical decomposition which may enable us to perform multiscale analysis of complex networks .
these perspectives will be systematically pursued in the future .
an interesting question is the stability of these objects against noise . to answer this question
, one has to clarify the source of noise .
there are two types of noise one should consider in landscape analysis of networks
one associated with the energy function @xmath5 and the other associated with the network structure .
the former can be dealt with traditional persistent homology denoising , where critical nodes with shallow basins can be merged with their saddles .
the latter is however more challenging as there are no systematic studies yet on perturbation or bootstrapping of networks . in the examples above
, we used edge thresholding on the les misrables and the protein binding networks , which is equivalent to modeling such networks as a superposition of a signal graph and some erds - rnyi type random graphs as noise .
however there might be better models which lead to different denoising rules .
w.e . acknowledges supports from aro grant w911nf-07 - 1 - 0637 and onr grant n00014 - 01 - 1 - 0674 .
j.l . is grateful to eric vanden - eijnden for helpful discussions .
y.y . thanks xuhui huang for providing figure s-2 in supporting information with helpful discussions , as well as supports from the national basic research program of china ( 973 program 2011cb809105 ) , nsfc ( 61071157 ) , microsoft research asia , and a professorship in the hundred talents program at peking university .
we show first that every local minimum in @xmath26 must be an index-@xmath27 critical node .
let @xmath10 be a local minimum in @xmath26
. then @xmath10 reaches at least two local minima , say @xmath175 .
consider the subgraph with node set @xmath176 clearly , @xmath177 is connected and @xmath10 is the unique maximum node in @xmath177 . by the definition of the attraction basin , the set @xmath178 is not connected .
since @xmath177 is connected , it contains at least a path from @xmath179 to @xmath180 .
let @xmath181 be the local minimal energy path from @xmath179 to @xmath180 in the subgraph @xmath177 .
as @xmath182 is not connected , @xmath181 must pass @xmath10 , so that @xmath183 .
we now show by contradiction that @xmath181 is also a local minimal energy path in the original graph @xmath6 .
suppose we can find another path from @xmath179 to @xmath180 , called @xmath184 , so that @xmath181 is deformable to @xmath184 . for any @xmath185 , we have @xmath186 .
consider the set @xmath187 , which is non - empty .
we distinguish two cases : 1 .
then , @xmath189 , so that @xmath190 . by construction of @xmath181 , we have @xmath191 ; 2 .
if there exists @xmath192 and @xmath193 , we have some point @xmath194 that @xmath195 .
it is easy to see that @xmath196 must be @xmath10 , since other points on @xmath181 are in attraction basins of @xmath179 and @xmath180 . using lemma 1
, there exists a path @xmath197 from @xmath198 to @xmath10 ordered in energy increase . in particular , consider the point @xmath199 , we have @xmath200 so that @xmath201 .
moreover , @xmath202 and @xmath203 .
this contradicts with the fact that @xmath10 is a local minimizer in @xmath26 .
therefore , @xmath181 is a local minimal energy path , and @xmath10 is an index-@xmath27 critical node . let @xmath204 which is not a local minimum in @xmath26 .
then , @xmath198 must reach a local minimum @xmath10 in @xmath26 by the gradient flow . by the first part of the proposition , @xmath205 .
the proposition is proved .
( necessity ) .
we first show that index-@xmath2 and index-@xmath27 critical nodes , when added into the filtration , will change betti numbers in the way above . for index-@xmath2 critical nodes , they are local minima of graph @xmath7 . when a local minima is added into the filtration , it must create a new connected component which increases the @xmath2-th betti number , @xmath108 .
index-@xmath27 saddles will play a more complicated role .
we have two situations * if an index-@xmath27 saddle lies on top of a global minimal energy path , it will decrease @xmath108 upon being added ; * if an index-@xmath27 saddle lies on top of a local minimal energy path other than the global one , it will increase @xmath111 upon being added .
given a pair of index-@xmath2 critical nodes @xmath206 , among all local minimal energy paths connecting them ( if exist ) , there must be a global minimal energy path @xmath207 , so that @xmath208 is less than any other local minimal energy paths between @xmath179 and @xmath180 .
we denote the maximal node of the global minimal energy path as @xmath10 .
such @xmath10 is an index-@xmath27 critical node .
when @xmath10 is added into the filtration , the @xmath2-th betti number @xmath108 will decrease as @xmath10 connects two components contains @xmath179 and @xmath180 respectively . for the other local minimal energy paths connecting @xmath179 and @xmath180
, the associated index-@xmath27 critical nodes will increase the first betti number @xmath111 when added into the filtration .
indeed , let @xmath198 be such an index-@xmath27 critical node .
thus @xmath198 is a maximum of a local minimum energy path @xmath28 such that @xmath209 .
@xmath28 is not deformable to the global minimal energy path @xmath207 between @xmath179 and @xmath180 .
then two paths @xmath207 and @xmath28 forms a loop , and hence the first betti number @xmath111 increases when @xmath198 is added into the filtration .
( sufficiency ) .
we show next that no other nodes when added into the filtration will change the first two betti numbers in the same way . for any node @xmath10 which lies in the attraction basin of a local minima @xmath210 for some @xmath211
, @xmath10 reaches @xmath212 by gradient flow . for any edge @xmath213 with @xmath214 ,
@xmath215 reaches @xmath212 and thus the triangle @xmath216 is included in the simplicial complex .
this implies that @xmath210 is contractible ( star - shape ) , whence no node in @xmath210 other than local minimum @xmath212 will change betti numbers .
it remains to show that any node in boundary @xmath217 will not change betti numbers in the same way .
any such node @xmath218 must reach at least two local minima , say @xmath31 and @xmath32 .
then by lemma [ lem : path ] there is a path @xmath219 for some @xmath220 such that @xmath221 for @xmath222 and @xmath223 for @xmath224 .
moreover @xmath225 implies that @xmath181 is deformable to a local minimal energy path @xmath226 between the same end nodes , for some @xmath227 .
@xmath198 can not decreases number of connected components as the path @xmath161 , which appears first in the filtration , already connects @xmath31 and @xmath32 .
now we show that the path @xmath181 will not create a loop either .
let @xmath228 be the maximal node on @xmath161 .
we must have @xmath229 . to see this ,
as @xmath181 is deformable to @xmath161 , there is a node @xmath230 which reaches @xmath231 .
we may assume @xmath232 ( @xmath233 ) since otherwise we are done .
then , by the construction of the path @xmath181 , we have @xmath234 , and hence @xmath229 . note that both @xmath198 and @xmath235 reach both local minima @xmath31 and @xmath32 , node @xmath236 with @xmath237 ( @xmath238 ) reaches @xmath31 ( @xmath32 , respectively ) , and node @xmath239 with @xmath240 ( @xmath241 ) reaches @xmath31 ( @xmath32 , respectively ) .
these will create a set of triangles such that @xmath181 is homotopy equivalent to @xmath161 , _ i.e. _ loop - free .
the proof is analogous to that of theorem 2 .
the energy landscape gives us a global picture for the different attraction basins on the network .
to understand the dynamics between the different basins , the transition path theory ( tpt ) provides a natural tool .
the transition path theory was originally introduced in the context of continuous - time markov process on continuous state space @xcite and discrete state space @xcite , see @xcite for a review . another description of discrete transition path theory for molecular dynamics can be also found in @xcite .
here we adapt the theory to the setting of discrete time markov chain with transition probability matrix @xmath242 .
we assume reversibility in the following presentation , the extension to non - reversible markov chain is straightforward
. given two sets @xmath243 and @xmath244 in the state space @xmath6 , the transition path theory tells how these transitions between the two sets happen ( mechanism , rates , etc . ) .
if we view @xmath243 as a reactant state and @xmath244 as a product state , then one transition from @xmath243 to @xmath244 is a reaction event .
the reactve trajectories are those part of the equilibrium trajectory that the system is going from @xmath243 to @xmath244 . to make the notion more precise ,
define the ordered family of times @xmath245 such that @xmath246 hence , a reaction happens from time @xmath247 to time @xmath248 . given any equilibrium trajectory @xmath249
, we call each portion of the trajectory of between @xmath247 and @xmath248 a _ @xmath250-reactive trajectory_. we call the time during which the reaction occurs the _ reactive times
_ @xmath251
the central object in transition path theory is the committor function .
its value at @xmath10 gives the probability that a trajectory starting from @xmath10 will hit the set @xmath244 first than @xmath243 , _
i.e. _ , the success rate of the transition at @xmath10 . given two sets @xmath243 and @xmath244 in the state space , @xmath252 satisfies the equation @xmath253 the committor function provides natural decomposition of the graph . if @xmath254 is less than @xmath255 , @xmath10 is more likely to reach @xmath243 first than @xmath244 ; so that @xmath256 gives the set of points that are more attached to set @xmath243 .
once the committor function is given , the statistical properties of the reaction trajectories between @xmath243 and @xmath244 can be quantified .
we state several propositions characterizing transition mechanism from @xmath243 to @xmath244 .
the proof of them is an easy adaptation of @xcite and will be omitted .
the probability distribution of reactive trajectories @xmath257 is given by @xmath258 the distribution @xmath259 gives the equilibrium probability that a reactive trajectory visits @xmath10 .
it provides information about the proportion of time the reactive trajectories spend in state @xmath10 along the way from @xmath243 to @xmath244 .
the reactive current from @xmath243 to @xmath244 , defined by @xmath260 is given by @xmath261 the reactive current @xmath262 gives the average rate the reactive trajectories jump from state @xmath10 to @xmath16 . from the reactive current
, we may define the effective reactive current on an edge and transition current through a node which characterizes the importance of an edge and a node in the transition from @xmath243 to @xmath244 , respectively .
the _ effective current _ of an edge @xmath263 is defined as @xmath264 the _ transition current _ through a node @xmath9 is defined as @xmath265 in applications one often examines partial transition current through a node connecting two communities @xmath266 and @xmath267 , _ e.g. _
@xmath268 for @xmath269 , which shows relative importance of the node in bridging communities .
the reaction rate @xmath270 , defined as the number of transitions from @xmath243 to @xmath244 happened in a unit time interval , can be obtained from adding up the probability current flowing out of the reactant state .
this is stated by the next proposition .
the reaction rate is given by @xmath271 finally , the committor functions also give information about the time proportion that an equilibrium trajectory comes from @xmath243 ( the trajectory hits @xmath243 last rather than @xmath244 ) . the proportion of time that the trajectory comes from @xmath243 ( resp . from @xmath244 ) is given by @xmath272
and target set as bound state @xmath172.,scaledwidth=80.0% ] the first figure is the whole co - appearance network of 77 main characters in the novel , les misrables , by victor hugo @xcite .
it is an undirected weighted graph with edge weights as the number of co - appearances for a pair of characters . without thresholding this network
contains one local minimum , valjean .
however a thresholding with edge weight greater than 7 gives rise to the subnetwork in the main text .
the second figure contains a list of structural information on 54 metastable states .
it contains a typical crystal structure in each state , and some free energy plots on certain reaction coordinates . from these pictures
one can read various structural properties of critical nodes in lao - protein binding transition network discussed in the main text .
more information about this system can be found in @xcite .
the third figure shows the ranking of transition currents out of misbound state 18 over eleven transition pathways .
the experiment selects each of the eleven solvated states @xmath273 as the source set and the misbound state 10 as the common target set . in each of the eleven experiments , relative transition current out of state 18 divided by total transition current from the source ,
is recorded and plotted in a descending order .
donoho dl , grimes c ( 2003 ) hessian eigenmaps : locally linear embedding techniques for high - dimensional data .
_ proceedings of the national academy of sciences of the united states of america _ 100:55915596 .
coifman rr , et al .
( 2005 ) geometric diffusions as a tool for harmonic analysis and structure definition of data : diffusion maps i. _ proceedings of the national academy of sciences of the united states of america _ 102:74267431 .
no f , schtte c , vanden - eijnden e , reich l , weikl tr ( 2009 ) constructing the equilibrium ensemble of folding pathways from short off - equilibrium simulations .
_ proceedings of the national academy of sciences of the united states of america _ 106:1901119016 . | topological landscape is introduced for networks with functions defined on the nodes . by extending the notion of gradient flows to the network setting , critical nodes of different indices are defined .
this leads to a concise and hierarchical representation of the network .
persistent homology from computational topology is used to design efficient algorithms for performing such analysis .
applications to some examples in social and biological networks are demonstrated , which show that critical nodes carry important information about structures and dynamics of such networks .
networks have become ubiquitous tools for describing structures that occur in a variety of fields in the past ten or fifteen years , including biology , social sciences , economics and engineering . to study a network
, one has to endow it with some mathematical structure .
the simplest mathematical structure on a network is the graph structure .
this gives rise to notions such as degrees , paths , connectivity , etc .
the distinctions between scale - free networks and small - world networks , for example , can be studied by examining this structure , see for example @xcite .
but one can endow a network with more sophisticated structures , such as geometric structure as in the theory of manifold learning , @xcite , or topological structure as in the theory of persistent homology @xcite .
these structures allow us to probe more deeply into the nature of the network . in this paper
, we discuss how one can endow a network with a _
landscape _ when we study a function on the node set .
the concept of landscape has been crucial in physics and chemistry in describing complex systems , such as energy landscape @xcite .
the introduction of such a concept into complex networks may equip us with a concise description of global structures of networks and help explain certain dynamics such as information diffusion and transition pathways .
many complex networks in real world carry flows of information , products , power , etc .
, which are driven by local gradients of a scalar or energy @xcite .
for example traffic flows may be driven by congestion function , heat flows are driven by temperature . in biomolecular folding ,
conformational changes are driven by the free energy of states . on internet ,
user s attention may be driven by the centrality or significance of websites such as pagerank . in these cases , communities or groups
emerge as metastable sets of gradient - based dynamics or energy basins .
therefore understanding the landscape of such functions will be crucial to disclose associated dynamics in complex networks . in the core of the landscape
lies the notion of critical nodes . in continuous
setting this meets the classical morse theory in the study of manifolds @xcite , where critical points can be located by vanishing gradients and their indices can be decided by dimensionality of the unstable manifold passing through .
however such an approach can not be applied to the graph settings as there is no unambiguous definition of dimensionality in general .
precisely , consider an undirected graph @xmath0 with a function defined on the node set @xmath1 .
the question we will attempt to address is : _ given a function on its nodes , how can we endow the network with a landscape , so that one can distinguish critical nodes such as the local minima , local maxima , and saddles _ ?
there are several studies in the literature which may lead to critical nodes for graphs by carrying morse theory to discrete settings .
nevertheless , none of them gives a satisfied answer to the question . in computational geometry
one may embed the graph into a 2d - surface and then apply morse theory for 2-manifolds @xcite .
however , such a surface embedding is not natural for general graphs in biological and social networks .
another candidate is discrete morse theory @xcite , which studies functions defined on all faces of cell complexes and is therefore hard to use in the graph setting above .
a related subject is the extension of the poincare - hopf theorem to the graph setting , e.g. in @xcite . in this paper
we present a purely combinatorial approach which starts from a discrete gradient flow induced by the function on graph nodes .
such an approach does not need a surface embedding , and turns out to be closely related to persistent homology in computational topology @xcite and discrete morse theory @xcite without studying functions on high dimensional cells .
in particular , given a function ( often referred to as an energy function ) on a network , we will define a discrete gradient flow associated with that function , as well as minimum energy paths between two disjoint sets of nodes .
this allows us to define critical nodes or saddles . roughly speaking , critical nodes
are associated with minimum energy paths between node pairs : index-@xmath2 critical nodes are simply local minima ; index-@xmath3 critical nodes are the highest energy transition nodes of minimal energy paths connecting index-@xmath4 critical nodes .
such a critical node analysis , as we show by examples in social networks and biological networks , leads to a concise representation of networks while preserving some important structural properties . in short
, the local minima or maxima together with their attraction basins can be interpreted as communities or groups in networks ; saddle points act as transition states between different critical points of lower indices . in particular , in social networks index-1 saddles act as hubs in connecting communities ; in biomolecular dynamics , index-1 saddles play roles as intermediate or transition states connecting misfolded and native states . in the latter ,
such an analysis does not rely on commonly used markov state model , whence can be applied to much more general data analysis .
moreover , this approach leads to a hierarchical classification of nodes in the network and a global visualization of networks adaptive to the landscape of given energy function . in algorithmic aspect , critical nodes in this paper
can be computed at a polynomial time cost with an algorithm based on computational topology by monitoring topological changes over energy evel sets , and in nondegenerate case an almost linear algorithm exists which is scalable for the analysis of large scale networks . |
the nature of the odderon an asymptotic odd-@xmath5 regge pole exchange , counterpart of the pomeron for a long time remains a subject of debate .
although there is little doubt about its existence , we still lack direct evidence of the odderon .
various reactions supposedly dominated by odderon exchanges , called odderon filters " @xcite , may offer only indirect evidence either because of low statistics or contamination by competing exchanges . in quantum chromodynamics
the odderon corresponds to the exchange of an odd number of gluons .
relevant calculations were done in a number of papers ; see @xcite and earlier references therein . the only direct way to see the odderon is by comparing particle and antiparticle scattering at high enough energies .
the high - energy proton - proton and proton - antiproton elastic scattering amplitude is a difference or sum of even and odd @xmath5-parity contributions , @xmath6 where , essentially , the even part consists of the pomeron and @xmath7 reggeon , while the odd part contains the odderon and the @xmath8 reggeon .
it is clear from the above formula that the odd component of the amplitude can be extracted from the difference of the proton - antiproton and proton - proton scattering amplitudes , and , since at high enough energies the contributions from secondary regge trajectories die out , this difference offers a direct way of extracting the odderon contribution .
unfortunatelly , @xmath0 and @xmath9 elasctic scatterings were typically measured at different @xmath10 , with the exception of the isr energies of 31 , 53 , 62 gev ( see fig . [
fig : accelarators ] ) .
[ 1]s accelerator , the @xmath0 and the @xmath11 elastic scattering data were measured at the same @xmath12 31 , 53 and 62 gev .
, title="fig:",width=377 ] at present , the only way to extract the odderon from the difference of the @xmath3 and @xmath0 scattering amplitudes is by means of a reliable interpolation of both amplitudes ( or cross sections ) over the missing energy regions . while the energy dependence of the forward amplitude ( or total cross sections ) is controlled by the regge pole theory , its @xmath1 dependence , especially in the dip - bump region , to a large extent is model dependent and unpredictable .
a simple , general , and reliable parametrization of the complicated diffraction structure at high energies at any fixed energy is a sum of two exponentials in @xmath1 related by a complex phase @xmath13 using this generic expression , phillips and barger ( pb ) @xcite ( for brevity we shall refer to it as the pb ansatz ) obtained good fits to the proton - proton differential cross sections , including the dip - bump region at several cern isr fixed energies . in refs .
@xcite the model was extended and improved , in particular , by accounting for the nonexponential behavior in the low @xmath14 region . in ref .
@xcite the parametrization was shown to be applicable to proton - antiproton scattering as well .
its connection with inelastic reactions is discussed on p. 185 of ref .
@xcite .
the pb ansatz reads @xmath15,\ ] ] where @xmath2 and @xmath1 are the standard mandelstam variables ; @xmath16 and @xmath17 were fitted to each energy independently ; i.e. , energy dependence in the pb ansatz enters parametrically .
the success of the simple pb parametrization motivates its further improvement , extension , and utilization . in papers @xcite the low-@xmath14 behavior of the pb ansatz was improved by modifying its simple exponential behavior by ( a ) inclusion of a two - pion threshold required by analyticity and ( b ) by means of a multiplicative factor reflecting the proton form factor . achieving good fits to the totem data @xcite , at the lhc energies of @xmath18 tev
, the modified ansatz was used to predict the behavior of the observables at future energies as well as the expected asymptotic behavior of the cross sections .
the model was also tested in @xcite against the totem data @xcite , and , contrary to many alternative models , it works reasonably well . in ref .
@xcite the pb ansatz eq .
( [ barger ] ) was improved ; in particular , its small-@xmath1 behavior was modified .
the fitted parameters are collected in table ii of that paper . for example , the authors of @xcite quote the following values for these parameters for @xmath0 scattering at @xmath19 gev : @xmath20 in @xmath21 and @xmath22 in gev@xmath23 .
as already mentioned , in addition to elastic @xmath0 scattering , the pb ansatz also describes @xmath11 data , with a different set of the parameters ( see below ) , thus opening the way to be used as a tool in extracting the odderon from the difference of the two .
however , in its original form , the pb ansatz does not describe the @xmath4 dependence of the model parameters . in this paper
, we try to remedy this limitation by combining the appealingly simple and efficient form of its @xmath1 dependence with energy dependence inspired by the regge pole model .
work in this direction was started in papers @xcite .
we address the following issues : ( 1 ) we smoothly interpolate between the values of the parameters fitted at fixed energy values , ( 2 ) extract the odderon contribution from the difference of the @xmath9 and @xmath0 cross sections , and ( 3 ) compare the energy dependence of this difference with the prediction @xcite based on perturbative quantum chromodynamics , by which the intercept of the odderon trajectory equal to one .
the structure of this paper is as follows : in the next section , we introduce the @xmath4 dependence of the parameters of the pb model , which in the next section are determined from the fits to data .
then we discuss the results , in particular the @xmath4 dependence of the results , including , for example , the odderon contribution and the odderon / pomeron ratio .
finally , we summarize and conclude .
we use the norm where @xmath24\ ] ] and @xmath25.\ ] ] following the regge pole theory , we make the following assignment , @xmath26 inspired by the donnachie and landshoff model @xcite of cross sections [ see eq .
( [ norm1 ] ) ] with effective falling ( subleading reggeons ) and rising ( pomeron ) components .
it follows from our fits that the falling ( subleading reggeon ) components in @xmath27 are small ; hence , they are neglected .
the slopes @xmath28 and @xmath29 in the regge pole theory are unambiguously logarithmic in @xmath2 , providing shrinkage of the cone : @xmath30 in the above formulae a normalization factor @xmath31 gev@xmath32 is implied . the phase @xmath17 is the weakest point of this toy " model or generalized pb model . in regge theory , it should depend on @xmath1 rather than on @xmath2 .
fortunately , at high @xmath10 the dependence of @xmath17 on energy is weak ( see the fits below ) .
however , this is not the case as the energy decreases . the best we can do is to fit the data with @xmath33 the low "- energy behavior is a weak point in any case .
apart from the varying phase , we must account in some way for the subleading ( @xmath7 and @xmath8 ) reggeon contributions .
this is done partly by the inclusion in @xmath34 of a decreasing term ( absent in @xmath27 ) .
a complete treatment of these terms with proper @xmath1-dependent signatures will require a radical revision of the model , and we hope to come back to this issue in the future .
now we proceed with this simple approach that has a chance to be viable at high energies , where the pomeron and odderon dominate @xcite and the above complications may be insignificant . to understand better the existence of any connection between the ansatz ( [ barger ] ) and the regge pole model , we plot the values of the parameters @xmath16 and @xmath17 against @xmath2 and fit their experimental " values to regge - like formulas .
this can be done in two complementary ways : a successive two - step " fit .
first , we acquire the values of the parameters @xmath35 from the fits to the @xmath0 and @xmath9 data , then we fit their regge forms ( see below ) to the obtained experimental " values of @xmath35 .
alternatively , one may determine the parameters of eqs .
( [ ac ] ) and ( [ bd ] ) from a single simultaneous fit to all available data .
we chose the first option ( 5 parameters ) since otherwise there were too many ( at least 12 ) free parameters .
thus , we proceed with a two - step fit , by which the final values are determined from a fit to the experimental " values of @xmath16 and @xmath17 .
we fitted separately @xmath0 and @xmath36 in two variables , @xmath2 and @xmath37 by using @xmath0 and @xmath36 data on total and differential cross sections ranging from the isr to the lhc for @xmath0 and from s@xmath38s to the tevatron for @xmath9 . here
the following remarks are in order : \(1 ) it is clear from eqs .
( [ ac ] ) and ( [ bd ] ) that , while the parameters @xmath39 and @xmath5 are particularly sensitive to the data on total cross sections , @xmath28 and @xmath29 are correlated mainly with the differential cross sections ( the slopes ) .
\(2 ) although we are interested mainly in the high - energy behavior ( the odderon ) , low - energy effects can not be fully neglected .
they are taken into account approximately by including in @xmath39 and @xmath5 subleading terms of the type @xmath40 .
\(3 ) having fitted @xmath16 and @xmath41 we perform a cross - check by calculating the resulting total cross sections .
\(4 ) at high energies , the proton - proton and antiproton - proton total cross sections are supposed to converge .
( we consider only this simple option , although we are aware of alternatives . ) since the existing data are not yet in this asymptotic domain , we introduce an extra constraint .
\(5 ) the most delicate issue is the phase , which in regge phenomenology is expected to be @xmath1 , rather than @xmath2 , dependent .
our fits ( and those of @xcite ) show considerable energy dependence of the phase at low energies but weak dependence at high energies , where we are particularly interested in looking for the odderon signal . postponing the introduction of a true regge - pole - motivated , @xmath1-dependent phase to a further study , here we assume a simple parametrization @xmath42 .
we have calculated the @xmath2 dependence of the parameters using a fitting strategy consisting of three consecutive steps described in detail below . in doing so , the following criteria were applied : * best @xmath43 for each fit ; * the @xmath44 range was set within @xmath45-@xmath46 gev@xmath32 ; * for each fixed energy , the model was fitted simultaneously to @xmath47 and @xmath48 ( focusing on the dip region ) ; * from the resulting fits , @xmath48 was reconstructed in the whole available energy range .
the values of the fitted parameters , in general , are consistent , within about 30% , with those obtained in refs .
@xcite , except for the small value of @xmath49 deviating considerably from those quoted in the above papers , which used a different @xmath1 range for their data analysis .
figure [ fig : bp_fits ] shows a fit to the data on the @xmath0 and @xmath9 differential and total cross sections .
the parameters @xmath16 and @xmath17 were fitted to each energy separately . given the simplicity of the model , the fits look reasonable .
figure [ fig : pp_params1 ] shows the fitted values of the parameters @xmath16 and @xmath17 both for @xmath0 and @xmath9 scattering to be used as experimental " data in the second stage of our fitting procedure , in which the explicit expressions ( [ ac ] ) , ( [ bd ] ) and ( [ phi ] ) are inserted .
the fitted values of the parameters and relevant @xmath51 values are quoted in tables [ tab : results1 ] and [ tab : results2 ] .
note the difference between the present model and that of refs .
@xcite in the low-@xmath14 behavior of the differential cross section , corrected in @xcite and taking account of the deviation from an exponential .
normalization in ref .
@xcite is arbitrary since , in that paper , only the differential cross section is shown .
total cross sections are not shown in the above papers .
[ 2 ] data at discrete energy values .
the @xmath47 data are from refs .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] data at discrete energy values .
the @xmath47 data are from refs .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] data at discrete energy values .
the @xmath47 data are from refs .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] data at discrete energy values .
the @xmath47 data are from refs .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] .values of the parameters from a fit to the @xmath0 data at various @xmath4 .
the quoted errors correspond to the relative errors , as given by cern minuit fitting package ( status= converged , error matrix accurate ) . [ cols=">,<,<,<,<,<,^ " , ] the resulting values of the parameters after the second stage of fitting are @xmath54 @xmath55 [ 3 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] and @xmath36 data .
the data on @xmath48 are from the particle data group database @xcite.,title="fig:",width=264 ] the lowest right icon in fig .
[ fig : pp_params1 ] is a cross - check " , showing the @xmath0 and @xmath9 total cross sections calculated from eq .
( [ norm1 ] ) with the explicit values of the parameters defined by ( [ ac ] ) , ( [ bd ] ) and ( [ phi ] ) .
the @xmath9 total cross section turns down at highest energies , deflecting dramatically from that of @xmath56 the reason for that nonphysical effect is the scarcity of @xmath38 data , leaving too much freedom in the high - energy extrapolation of the cross section , where one expects asymptotic equality @xmath57 for @xmath58 ; see eq . ( 47 ) in ref .
this deficiency should , and can , be cured by imposing an additional constraint on the model .
this will be done in the next subsection , by fixing ( tuning ) the parameter @xmath59 ( the leading powers in @xmath2 of @xmath60 ) to be the same in @xmath0 and @xmath9 scattering .
the above , unbiased fit does not satisfy automatically the required ( see @xcite ) asymptotic constraint @xmath62 since the available freedom ( especially due to the lack of simultaneous @xmath0 and @xmath38 elastic scattering data at @xmath63 540 , 630 , 1800 and 7000 gev ) leaves much freedom for the extrapolation to energies beyond the existing accelerators . to remedy this problem ,
we have tuned the parameters to meet the above constraint in the currently available energy range .
below are the results of the tuned " fit ( see fig .
[ fig : pp_params2 ] ) satisfying the asymptotic condition @xmath62 in the @xmath64 14 tev energy range .
the refitted @xmath2-dependent values of the parameters for @xmath0 and @xmath65 scatterings are @xmath66 @xmath67 [ 4 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ] and @xmath65 data , constrained by @xmath68 as @xmath69 .
the lowest right icon shows a cross - check for the ( asymptotically converting ) total cross sections .
, title="fig:",width=264 ]
the existence of a parametrization for both @xmath0 and @xmath9 scattering offers the possibility to extract the odd-@xmath5 contribution by using the formula @xmath70 where @xmath71 and @xmath72 are , respectively , the @xmath5-even and @xmath5-odd components of the scattering amplitude ; see table 1 in ref .
@xcite .
[ 5 ] cross sections calculated from the difference @xmath73 and from the sum @xmath74 fitted to the data.,title="fig:",width=264 ] cross sections calculated from the difference @xmath73 and from the sum @xmath74 fitted to the data.,title="fig:",width=264 ] cross sections calculated from the difference @xmath73 and from the sum @xmath74 fitted to the data.,title="fig:",width=264 ] cross sections calculated from the difference @xmath73 and from the sum @xmath74 fitted to the data.,title="fig:",width=264 ] [ odd ] calculated from the present model fitted to the data .
as seen from the figure , this difference tends to a ( small ) constant , corresponding to a unit intercept odderon @xcite.,title="fig:",width=377 ] while the @xmath5-even component contains the pomeron and the @xmath7 trajectory ( both known ) , the @xmath5-odd part is made of the poorly known odderon and the familiar @xmath8 trajectory . at the lhc energies , the contribution from secondary trajectories , e.g. , @xmath7 and @xmath8 , is negligible ( see ref .
@xcite ) ; therefore , by taking the difference between the known ( fitted ) @xmath38 and @xmath0 amplitudes , one gets a pure odd-@xmath5 contribution that , in the lhc energy range , is the odderon . from the explicit expressions for @xmath0 and @xmath9 amplitudes ( cross sections ) , we calculate the odderon amplitude ( or its contribution to the cross section ) by taking the difference @xmath75 the result ( the energy dependence for several fixed values of @xmath1 and @xmath1 dependence for several fixed values of @xmath2 ) is shown in fig .
the extracted model parameters are used then to evaluate the even and odd contributions to the forward scattering amplitude . in fig .
[ fig : odderon1 ] we show the pomeron and the odderon contributions as the sum or the difference of the differential cross section of @xmath38 and @xmath0 elastic scattering .
we see that , as expected , the pomeron dominates at large colliding energies , while the odderon contribution is small and at @xmath76 even changes sign .
a particularly interesting feature is shown on the lower right panel of fig .
[ fig : odderon1 ] , where the odderon / pomeron ratio is shown at different values of @xmath1 at various @xmath4 . apparently , at @xmath77 gev , the odderon / pomeron ratio becomes @xmath1 independent and the t - dependent curves pass through the same point of about 0.03 .
an important finding of our paper is the near energy independence of the odd-@xmath5 contribution to the scattering amplitude , which , at high ( e.g. , those of the lhc ) energies , is dominated by the odderon exchange . in fig .
[ fig : odderon2 ] we plot the odd-@xmath5 contribution calculated from the difference ( [ odd ] ) .
both the real and imaginary parts of this difference at @xmath76 tend to a ( small ) constant value , which correspond to a unit - intercept odderon as predicted @xcite from quantum chromodynamics .
the present work is a semiquantitative estimate of the possible odd-@xmath5 contribution to the scattering amplitude ( cross sections ) at high energies .
semiquantitative " implies limitations due to the following : * for the sake of simplicity , we ignore the low-@xmath1 nonexponential behavior ( sharpening ) of the differential cross section .
this simplification has dramatic impact on the low-@xmath1 behavior of the odderon contribution because of the large errors due to the cancellation of the pomeron contribution .
the importance of the low-@xmath1 effects was emphasized , e.g. , in refs . @xcite .
* the @xmath2- ( rather than @xmath1- ) dependent signature factor ( phase ) is in agreement with the data , but it is in contrast to expectations based on regge phenomenology . * the odd-@xmath5 contribution to the amplitude , equal to the difference ( [ odd ] ) , shown in fig .
[ fig : odderon2 ] , tends to a constant limit determined by a unit - intercept odderon , predicted by bartels , lipatov , and vacca @xcite . *
it is an oversimplified treatment of the low - energy ( secondary reggeons " ) contributions ( odderon and pomeron here implies , generally , odd and even exchanges ) .
* there is the absence , for the moment , of any physical interpretation in terms of reggeon exchanges of the components in the pb ansatz .
we strongly recommend to run the lhc accelerator at the injection energy @xmath78 gev and at the tevatron energy of @xmath79-@xmath80 tev , so that the missing energy range of @xmath0 elastic scattering will be covered and elastic @xmath0 scattering data will be measured in the region where elastic @xmath38 scattering is already measured
. it would also be desireable to measure elastic @xmath0 scattering at the @xmath81 gev region the upper energy range of the rhic accelerator .
* rewrite the pb ansatz with correct , @xmath1-dependent regge signature factors @xcite , remembering that in the present study the even and odd parts of the amplitude differ only by the values of the fitted parameters , without the identification of particular regge exchanges ( trajectories ) ; * fit to the data to provide hints concerning the physical meaning of the components ( two or more ) ; for example , our fits ( and those of refs .
@xcite ) indicate that @xmath84 but @xmath5 rises with energy faster than @xmath39 .
similar considerations may help in relating the second term with the odderon or a hard " pomeron , for which the slopes ( @xmath28 and @xmath29 ) may be indicative .
* add lower - lying contributions ( subleading reggeons ) which are inevitable at lower energies , at the prise of giving up of the attractive simplicity of the model ; * use other asymptotic pomeranchuk - like constraints in determining the parameters ( see the work in ref .
@xcite that points to this interesting direction ) ; * calculate and fit the model to other observables , e.g. , the ratio of the real to imaginary parts of the amplitude @xmath85 , of the slope @xmath86 , etc .
its relation to inelastic processes ( see @xcite ) may offer additional information .
l.j . acknowledges cordial hospitality by the wigner rcp , budapest , hungary , where this work was completed .
he is also grateful to d.a .
fagundes and g. pancheri for useful discussions and correspondence .
this work was partially supported by the hungarian otka grant no .
nk101438 .
jenkovszky and b.v .
struminsky , report no
. itp-88 - 170 preprint , kiev , 1988 ; + a. p. contogouris , l. l. jenkovszky , e. martynov , and b. struminsky , phys .
b * 298 * , 432 ( 1993 ) ; + v. p. goncalves , nucl . phys .
* a902 * , 32 ( 2013 ) . c. ewerz , m. maniatis , and o. nachtmann , ann .
( amsterdam ) * 342 * , 31 ( 2014 ) .
r. j. n. phillips and v. d. barger , phys .
* 46b * , 412 ( 1973 ) . g. antchev _ et al . _
( totem collaboration ) , europhys .
lett . * 95 * , 41001 ( 2011 ) .
a. grau , s. pacetti , g. pancheri , and y. n. srivastava , phys .
b * 714 * , 70 ( 2012 ) .
d. a. fagundes , a. grau , s. pacetti , g. pancheri , and y. n. srivastava , phys .
d * 88 * , 094019 ( 2013 ) .
d. a. fagundes , a. grau , s. pacetti , g. pancheri , and y. n. srivastava , _ proc .
sci _ , dis ( 2013 ) 306 .
g. pancheri , d. a. fagundes , a. grau , and y. n. srivastava , nuovo cimento soc .
037 * , 179 ( 2014 ) .
l. jenkovszky and d. lontkovskiy , lessons from the first measurements of elastic @xmath0 scattering at the lhc , in _ new trends in high - energy physics _ , edited by p.n .
bogolubov and l.l .
jenkovszky ( l. l. , kiev , 2011 ) , p. 305
, bogolyubov institute for theor .
physics , nat .
sc . of ukraine , http://crimea.bitp.kiev.ua/crimea2011/ ; l. jenkovszky and d. lontkovskiy , in _ proc . of rencontres
du vietnam , 14th workshop on eds 2011 _ , slac econf c111215 , edited by mary ann rotondo and chung i - tan , stanford national accelerator laboratory .
l. l. jenkovszky , a. i. lengyel , and d. i. lontkovskyi , int .
j. mod .
a * 26 * , 4755 ( 2011 ) .
a. donnachie and p. v. landshoff , nucl
* b231 * , 189 ( 1984 ) .
a. n. wall , l. l. jenkovszky , and b. v. struminsky , fiz .
chastits at .
yadra * 19 * , 180 ( 1988 ) .
u. amaldi and k. r. schubert , nucl .
* b166 * , 301 ( 1980 ) . g. antchev _ et al . _
( totem collaboration ) , europhys .
* 101 * , 21002 ( 2013 ) .
a. breakstone et al .
( ames - bologna - cern - dortmund - heidelberg - warsaw collaboration ) , nucl.phys .
* b248 * , 253 ( 1984 ) .
d. bernard _ et al . _
( a4 collaboration ) , phys .
b * 198 * , 583 ( 1987 ) .
d. bernard _ et al . _
( ua4 collaboration ) , phys .
b * 171 * , 142 ( 1986 ) . v. m. abazov _ et al . _
( d0 collaboration ) , phys .
d * 86 * , 012009 ( 2012 ) .
j. beringer _
_ ( particle data group collaboration ) , phys . rev .
d * 86 * , 010001 ( 2012 ) . | starting from a simple empirical parametrization of the scattering amplitude , successfully describing the dip - bump structure of elastic @xmath0 scattering in @xmath1 at fixed values of @xmath2 , we construct a toy model interpolating between missing energy intervals to extract the odderon contribution from the difference between @xmath3 and @xmath0 elastic and total cross sections .
the model is fitted to data from @xmath4 = 23.5 gev to 7 tev and used to extract the odderon and its ratio to the pomeron . from our fits ,
a unit intercept odderon follows , as predicted by j. bartels , l.n .
lipatov , and g.p .
vacca , on the basis of perturbative quantum chromodynamics . |
ap stars are chemically peculiar stars that range from early - b to early - f spectral types with the majority having detectable magnetic fields with strengths of a few @xmath1 to a few @xmath2 g ( @xcite , according to whom @xmath3 per cent have mean longitudinal fields greater than 400 g ) .
the study of ap stars with strong magnetic fields has ramifications for several other branches of astrophysics .
the interaction among strong magnetic fields , atomic diffusion and energy transfer in ( or above ) the upper atmospheres of non - degenerate stars has direct implications for observed stellar abundances and for the instability of stellar pulsations in @xmath4cephei and sdb stars .
the very different magnetic field strengths , orientations and geometries , as well as the great variety of surface abundance distributions for ap stars , therefore makes them particularly informative .
the magnetic field geometry , determined from maps of longitudinal magnetic field strength versus rotation phase , generally agrees with an axisymmetric dipole , although strong deviations from this do occur , e.g. the quadrupole field configuration of hd37776 ( @xcite ) .
typically , the magnetic axis is significantly inclined to the rotation axis , hence these stars are called oblique rotators .
many ap stars are known to have surface spots , indirectly observable from photometric light variations or spectroscopic line strength variations over a rotation cycle .
these stars are known as @xmath5cvn stars after the prototype of that name @xcite .
their light and line strength variations are a consequence of abundance spots that are often associated with the magnetic poles .
the rotation period of ap stars is typically relatively long of the order of several days to years , and even in some cases decades , as determined from their magnetic field variations .
slow rotation is a requirement for the atmospheric atomic diffusion that gives rise to the observed spectral peculiarities , since rapid rotation generates turbulent meridional circulation that inhibits diffusion .
interestingly , atomic diffusion processes can give rise to observably stratified atmospheres in some ap stars , and probably all rapidly oscillating ap ( roap ) stars .
examples of studies of atmospheric abundance stratification in the cool roap star subgroup are given by @xcite .
these studies are in agreement with each other . in general ,
fe is concentrated by gravitational settling in the deeper observable layers in the photosphere around @xmath6 ; the pr and nd line - forming layers are concentrated by radiative levitation high in the atmosphere at @xmath7 , a level that is chromospheric in the sun , and in the roap stars is above the line - forming layer of the narrow core of the h@xmath0 line at @xmath8 , which is itself above the fe line - forming layer .
many other elements show similar behaviour to these examples .
doppler imaging studies of ap stars show that the spots associated with the magnetic poles are typically caused by abundance concentrations of rare earth elements .
overabundances globally are several orders of magnitude compared to normal abundance stars ( e.g. , @xcite ) , and may reach a factor of @xmath9 in the spots , as shown by the doppler imaging study of the roap star hr3831 by @xcite , and as shown in this paper for hd99563 . as a result of these strong abundance anomalies , line - blocking changes the temperature gradient at the magnetic poles of ap stars ; flux is redistributed usually from the blue to the red , so that rotational light variations in visible light occur in antiphase for short and long wavelength observations , along with obvious spectral line strength variations .
importantly , the rotational variations whether magnetic , spectroscopic or photometric allow the precise determination of the stellar rotation period .
with the surface of some ap stars resolvable by doppler imaging , and atmospheric depth being resolved because of stratification , ap stars and particularly the roap stars offer the only opportunity for 3-d studies of stellar atmospheres for any stars other than the sun . a subgroup of ap stars , the rapidly oscillating ap stars ( roap ) , exhibit rapid oscillations with periods of @xmath10min .
their frequencies are considerably higher than those typical for @xmath11sct stars with whom they overlap in location inside the classical instability strip of the hr diagram . at present ,
only 40 roap stars are known ( see , e.g. , @xcite ) , although several photometric surveys have searched for rapid pulsation in ap stars , such as @xcite . for a ( non - exhaustive ) list of spectroscopic studies of roap stars , see kurtz , elkin & mathys ( 2006a ) .
the roap class characteristics are therefore poorly constrained .
as they share the same region of the hr diagram with the ( photometrically established ) non - oscillating ap ( noap ) stars , the selection of new roap candidates is difficult and mostly leads to null - results for both spectroscopic studies ( see , e.g. , @xcite ; freyhammer , elkin & kurtz 2008b ; @xcite ) and for the photometric surveys listed above .
the pulsations of roap stars are described well by the oblique pulsator model , where the pulsation axis is assumed to be aligned with an oblique magnetic field axis ( @xcite , @xcite ) , or derived to be offset from both the rotational and magnetic axes @xcite .
@xcite used this model and assumed that convection was suppressed locally in the magnetic polar regions .
they showed that pulsation aligned with the magnetic poles is unstable to the axisymmetric , nonradial high - order modes observed in roap stars , mainly through excitation by the opacity ( @xmath12 ) mechanism acting in the hydrogen ionization zones in these regions .
they also considered the effect of local he settling , and concluded it is not the primary process responsible for the high - frequency pulsation .
nevertheless , their model finds he to be deficient in the observable atmosphere at the magnetic poles .
while he is not directly observable at visible wavelengths at the effective temperatures of the roap stars ( @xmath13k ) its gravitational settling implies that the magnetic poles are hydrogen - rich .
we return to this point in section [ hcore ] where we show evidence to support it .
statistical studies of the roap stars can be intriguing and informative , e.g. in the comparison of the roap and noap stars by @xcite .
but with only 40 such stars known the class characteristics are still not well - defined .
therefore , the greatest progress at present is to be made by in - depth studies of individual roap stars , such as that for hr3831 ( e.g. @xcite ; @xcite ) and hr1217 ( @xcite ; @xcite ) .
this is reminiscent of the related problem in the pulsational studies of @xmath11sct stars . despite being the most abundant type of variable star
, the major breakthroughs have been obtained through individual studies such as that of fgvir @xcite .
for the @xmath11sct stars , as for the roap stars , the question of when pulsation modes are exited or not remains unanswered for physically similar stars why do only some of these stars pulsate and what is the underlying physics ?
the many null results among roap candidate stars strongly suggest that either the pulsation properties are not fully known ( i.e. amplitude ranges or pulsation geometry ) or the selection of candidates is too coarse .
individual studies of known roap stars are therefore important ( presently the only option at hand ) to study the roap properties in detail . some examples of enlightening studies include that of @xmath14equ that clearly demonstrated the trends for pulsational amplitude with vertical stratification of chemical elements ( @xcite ; @xcite ) , doppler imaging of hr3831 ( kochukhov 2004 , 2006 ) , and the study of line bisector variability by @xcite and a new type of upper atmospheric pulsation ( @xcite ; @xcite ) . to first order , pulsations in many roap stars are oblique axisymmetric dipole ( @xmath15 , @xmath16 ) p modes .
nevertheless , it has been clear for decades that normal modes described by single spherical harmonics are insufficient to explain fully the mode geometry in these stars .
uniquely , the oblique modes of the roap stars allow them to be viewed from varying aspect , giving detailed information about the mode geometry .
an early example was the study of hr1217 by @xcite who showed the presence of alternating even and odd degree p modes , with the odd modes being consistent with oblique dipole modes , and the even modes not being consistent with either radial or quadrupole modes .
this was confirmed by the more extensive whole earth telescope study of this star by @xcite . even for roap stars with single pulsation modes , in the best observed cases such as hr3831
it can be shown that the single mode is a distorted oblique dipole mode ( @xcite ; @xcite ) .
any deviation from spherical symmetry for a pulsating star will give rise to distorted modes . for
the roap stars the dominant effect is the magnetic field , with contributions from the non - uniform abundance distributions and stellar rotation .
the non - uniform abundance distributions are particularly important for spectroscopic studies such as the one in this paper on hd99563 , and that of @xcite for hr3831 where concentrations of ions in spots mean that the pulsation mode is non - uniformly sampled with the rotation
. however , photometric studies that more uniformly sample the mode geometry with rotation also show the distortions of the modes .
the effects of the magnetic field on the oscillations of roap stars have been extensively studied theoretically , and there are ongoing investigations into this complex problem ( @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) . while generally the magnetic field effect on the oscillations is expected to be small , the theoretical results do lead to expectations of distorted modes .
for example , @xcite studied the influence of the magnetic field on the pulsation geometry and found that not only can a dipole magnetic field explain the suppression of the observationally absent @xmath11sct modes , but it also leads to insufficiency of a single spherical harmonic to describe the pulsations angular dependency ; a series of spherical harmonics of different degrees @xmath17 are required .
inclusion of magnetic fields in the study and interpretation of roap pulsations is therefore necessary and constitutes ` a formidable mathematical problem ' @xcite .
@xcite continued the study of the effects of a dipole magnetic field on roap pulsations and showed again that the field stabilised low - order p modes ( such as those excited in @xmath11sct stars ) for fields stronger than 1 kg , but that nonradial , high order distorted dipole , or quadrupole , p modes remain overstable and are most likely to be excited in roap stars .
this is in excellent agreement with the current observations , as well as with the mounting observational and theoretical evidence for observable pulsation nodes in and above the photosphere in the magneto - acoustic layer .
further , and most interestingly , saio modelled the latitudinal amplitude dependence for the roap case of hr3831 and found reasonable agreement with the observations by @xcite .
saio s non - adiabatic analysis predicts ( his figure 10 ) a latitudinal dependence of amplitude that does not differ from a purely dipole mode photometrically , whereas spectroscopy s ability to distinguish potentially between horizontal and radial pulsation components may detect the relatively smaller horizontal component when studying stellar spots close to ( but not necessarily at ) the magnetic poles .
saio s figure 10 demonstrates a reasonable agreement with observed latitudinal velocity amplitude variation for hr3831 , but predicts higher amplitude toward the magnetic axis than is observed .
however , the formation layer of the nd lines used by kochukhov in his observations is far above the boundary of saio s models .
@xcite also did not take into account the inhomogeneous nd surface distribution .
the oblique pulsator model describes how nonradial pulsation modes seen from varying aspect with the stellar rotation show amplitude and phase modulation ( see the overview by @xcite ) that results in a frequency multiplet in an amplitude spectrum , centred on the pulsation frequency @xmath18 and separated by _
exactly _ the rotation frequency @xmath19 .
this exact splitting by the rotation frequency allowed kurtz ( 1982 ; see also @xcite ) to rule out the possibility that the observed multiplets can be explained by rotationally perturbed @xmath20-modes ( @xmath20 being the azimuthal order of a mode ) , which is the typical interpretation of frequency splitting observed in other ( non - oblique ) nonradially oscillating stars , as such rotational multiplets are split by @xmath21 , where @xmath22 is the rotation frequency , @xmath23 is a small constant dependent on the stellar structure , @xmath24 is the overtone of the mode and @xmath17 is its degree . for oblique pulsation
the amplitude ratio of the multiplet components contains information about the pulsation geometry ( assumed to be aligned with the magnetic geometry ) . in the case of a simple oblique dipole mode @xmath25 , where @xmath26 is the inclination angle of the rotation axis , @xmath4 is the obliquity of the pulsation / magnetic axis relative to the rotation axis , and @xmath27 , @xmath28 are the observed amplitudes of the frequency triplet . for modes that are not purely dipolar , or vary in structure with atmospheric depth in the stratified roap stars atmospheres ,
this constraint can depend on the atmospheric depth of formation of the spectral lines studied ( see , e.g. , @xcite ) . despite the complex nature of roap pulsations in the presence of magnetic fields and an oblique geometry , the observational data have rich potential for asteroseismic inference .
photometrically , multisite observations and space - borne observations allow the resolution of both the large and small frequency spacings and through them constrain the stellar dimensions and the interaction of the magnetic field and rotation with the pulsation modes .
high - resolution time series spectroscopy allows the study of the pulsation geometry over the stellar surface ( through line profile variability ) and also vertically in the stellar atmosphere , through radial velocity measurements at different line bisectors , or through lines of elements located at different depths in the stratified atmosphere ( e.g. , @xcite ) .
it is clear that the observations have begun to resolve the magneto - acoustic nature of the pulsations in these stars high into the atmosphere to levels that are unobservable in any other star but the sun .
radial nodes are resolved in some stars such as hd137949 ( mkrtichian , hatzes & kanaan 2003 ; kurtz , elkin & mathys 2005c ) and possibly hd99563 @xcite .
observed variation of pulsation phase with atmospheric depth shows the presence of running waves , and there is even an indication that shock waves are observed in the upper atmosphere of @xmath14equ ( @xcite ) , and by inference other roap stars with similar line profile variations .
hd99563 was discovered to be a roap star by @xcite , pulsating in a 10.70-min mode .
it is a visual binary with a 1.2mag fainter secondary at 1.79arcsec separation @xcite . a spectroscopic study by elkin et al .
( 2005 ) confirmed the monoperiodic pulsation found from photometry , and found the pulsations to have extremely high radial velocity amplitudes for a roap star , such as 2.6 for and up to 5 for eu and tm .
for sake of clarity , we emphasize that _ monoperiodic _ refers to a single ( physical ) pulsation mode
e.g. , the frequency quintuplet detected by @xcite ( actually 5 components of a frequency septuplet ) are all associated with only a single mode with pulsation frequency @xmath18 ; the rest of the quintuplet describes the amplitude and phase modulation of the mode .
elkin et al .
( 2005 ) measured a projected rotation velocity of @xmath29@xmath30 from a high - resolution spectrum , while from narrow - band strmgren photometric indices they estimated @xmath31k and @xmath32 ( cgs ) .
furthermore , they studied variations in amplitude and phase for line bisectors , and from element to element . the demonstrably high amplitudes ( and
thus high signal - to - noise , @xmath33 ) and rich and varied information content from using lines of different elements or with line depth , show hd99563 to be a particular interesting roap star for detailed study .
@xcite carried out a photometric multisite campaign and found an equally spaced frequency quintuplet in hd99563 , for the pulsation frequency @xmath34mhz .
they determined the rotation period to be @xmath35d and a time of pulsation amplitude maximum to be hjd2452031.29627 .
the photometry showed , through application of the axisymmetric spherical harmonic decomposition method of @xcite , that the mode of hd99563 is dominated by a dipole component , with some contribution of an @xmath36 component ( i.e. , the mode is a distorted dipole ) .
handler and collaborators combined photometry , a hipparcos trigonometric parallax and the temperature and surface gravity estimate by elkin et al .
( 2005 ) to constrain the basic parameters of hd99563 . under the assumption that the secondary star in the visual binary is a physical binary component and that the wide orbit indicates
the stars have evolved separately , the estimated parameters of both stars must fit two stellar evolutionary models of the same age . by adopting @xmath37k and @xmath38 , they found for the primary @xmath39 , @xmath40 , an inclination angle of the stellar rotation axis @xmath41 , and a magnetic obliquity @xmath42 .
this makes hd99563 especially interesting because of its high pulsation amplitude , its rotation that is fast enough for doppler imaging , and its ( for a roap star ) rare favourable orientation where both magnetic poles become visible throughout a rotation cycle .
@xcite were the first to measure the magnetic field of hd99563 , finding a longitudinal field of @xmath43 g on hjd2452494.483 , or rotation phase 0.08 with respect to handler et al.s pulsation ephemeris .
@xcite published another two measurements at rotation phases 0.07 and 0.09 showing @xmath44 and @xmath45 g , respectively .
further , using the russian 6-m telescope at the special astrophysical observatory of the russian academy of sciences ( saoras ) , @xcite obtained 6 additional magnetic longitudinal field measurements near rotation phases 0.25 , 0.5 and 0.75 . with all 9 available measurements
, these authors used the well - known relation for a dipolar magnetic field : @xmath46 where @xmath47 @xmath0 is the angle between the magnetic pole and the line - of - sight , @xmath26 is the rotational inclination , @xmath4 is the angle between the rotation axis and the magnetic axis , @xmath22 is the rotation frequency , @xmath48 is the longitudinal magnetic field strength and @xmath49 is the polar field strength . as can be seen from eq.[eq : mag2 ] , the mean magnetic field strength @xmath50 and the amplitude of the magnetic field variations @xmath51 .
thus @xmath52 , from which @xmath4 can be constrained when @xmath26 is known .
using the measured values for hd99563 , @xmath53 g and @xmath54 g , @xcite find an inclination of @xmath55 , and a magnetic obliquity @xmath56 ( 8490 for a 1@xmath57 range ) , which are similar to those determined by @xcite .
the magnetic minimum coincides with the moment of maximum modulated pulsation amplitude ( @xmath58 hereafter ) , and similarly for the maximum half a rotation phase later ( @xmath59 ) .
this is consistent with both magnetic poles being viewed from a similar angle and coming into view for almost the same duration ( requiring that either @xmath26 or @xmath4 be close to @xmath60 ) , and with the pulsation axis and magnetic axis being aligned .
@xcite used doppler imaging to construct the first tomographic maps of pulsation velocity , surface abundance and magnetic field map for the roap star hr3831 . with observations obtained over a wide range in rotation phase for hd99563 , it will be possible to do this with higher precision for this star because of its very high amplitudes , and it will be possible to look simultaneously at the additional dimension of the atmospheric depth dependence of the velocities .
kochukhov , as a first approximation , treated the analysis of hr3831 as if the atmosphere has a single pulsation layer .
observations of hd99563 and other roap stars show the situation to be much more complex than this . disentangling the vertical and horizontal pulsation characteristics
is a challenge , but one with rich rewards for the fields of asteroseismology , pulsation theory , magnetohydrodynamics , stellar atmosphere theory and atomic diffusion .
hd99563 is one of only a few known roap cases where tomographic mapping through doppler imaging is possible , based on rotation rate and the orientation of the magnetic field .
therefore , observations over three nights ( most of the 2.91-d rotation period ) were scheduled from two sites simultaneously
subaru on mauna kea and vlt on cerro paranal to attempt to collect spectra at high resolution in both time and dispersion to obtain a first picture of this star s photosphere , and to determine precisely its physical parameters . theoretically , the scheduled observations would have given us 70 per cent coverage of the rotation cycle , hence of the visible stellar surface .
but due to bad weather , only one site contributed significantly to the pulsation study , and only some 40 per cent rotational coverage was attained .
in this first of a series of papers on hd99563 , we are studying the horizontal pulsation geometry with the stellar rotation . a doppler imaging study of the surface abundance distribution of many ions and detailed modelling of the horizontal and vertical pulsation geometry is in preparation .
.[tab : obslog]journal of vlt observations indicating nightly observing date ( ut ) , rotation phase with reference to a time of spectroscopic pulsation maximum , hjd2454171.43328 , heliocentric julian date ( hjd ) range , typical exposure time , number of collected spectra , and mean @xmath33 .
the @xmath33 ratios are based on the random noise in the continuum of consecutive spectra , or for single spectra based on predicted ratios .
the 2004 data were re - reduced in same manner as the new data , and included in the analysis as they supplement the covered rotation phases .
an additional 35 spectra were obtained with the hds spectrograph at the subaru telescope on 2007 march 12 and march 13 and combined into two average spectra at rotation phases @xmath61 and 0.56 . [ cols="<,<,^,^,^,>,^ " , ]
we have performed a first analysis of the pulsation of hd99563 based on one of the most comprehensive spectroscopic efforts on a roap star so far .
we show that the pulsation is well - described by a single distorted oblique dipole mode , with different distortion terms for different elements . in the future
, we will use these differences to make further inference concerning the pulsation velocity field ultimately in 3-d . by using doppler imaging maps of the surface distribution of nd and eu ,
we have identified two relatively concentrated spots near the magnetic poles . during rotation phases at magnetic crossover , line doubling in absorption features of these elements ( with each component relating to a spot )
is considerable , and we show that the stellar photosphere at opposite sides of the star have mutual @xmath62 rad offset , again supporting the oblique dipole rotator geometry .
these spots allow the study of the pulsation mode for only parts of the visible hemisphere of the star , and from varying aspect , an opportunity only possible for roap stars .
remarkably , blue - to - red running wave features were found only near the rotation phases where the lines were doubled and the spots were seen at high aspect angle . our study included a short data set we also obtained with uves on the vlt 3 years earlier ( elkin et al . 2005 ) , which after a minor revision of the pulsation frequency showed amplitudes and phases for h , nd and eu that in general are consistent with the new data .
some roap stars exhibit short lifetimes of pulsation modes in the most extreme case , hd60435 , the mode lifetimes are less than a week . the stability of hd99563 over a 3-yr time span simplifies the interpretation of the pulsations .
this also suggests that it will be possible to merge future observations of the rotational phases not covered by our data for a more detailed study .
we found a clear rotational variation in the radial velocities of the h@xmath0 line in addition to the pulsational variations which we interpret to be the first observational evidence of hydrogen overabundance at the magnetic / pulsational poles , as expected theoretically as a result of helium depletion .
while we prefer this interpretation , alternatively , we found the first planet orbiting a roap star , a hot jupiter with a mass of 26 times the mass of jupiter . in future work we will model the h@xmath0 line radial velocity variations in detail to test our preferred hypothesis of hydrogen spots .
lmf , dwk and vge acknowledge support for this work from the particle physics and astronomy research council ( pparc ) and from the science and technology facilities council ( stfc ) .
lmf received support from the danish national science research council through the project ` stellar structure and evolution
new challenges from ground and space observations ' , carried out at aarhus university and copenhagen university .
we thank two anonymous referees for their comments that helped to improve the paper .
we acknowledge extensive usage of the vald , vizier , simbad , ads ( nasa ) databases .
wz is supported by the fp6 european coordination action helas and by the research council of the university of leuven under grant goa/2003/04 .
asplund m. , grevesse n. , sauval a. j. , 2005 , in barnes t. g. , iii , bash f. n. , eds , asp conf .
336 , cosmic abundance as records of stellar evolution and nucleosynthesis .
, san francisco , p. 25 | we have used high - speed spectroscopy of the rapidly oscillating ap ( roap ) star hd99563 to study the pulsation amplitude and phase behaviour of elements in its stratified atmosphere over one 2.91-d rotation cycle .
we identify spectral features related to patches in the surface distribution of chemical elements and study the pulsation amplitudes and phases as the patches move across the stellar disk .
the variations are consistent with a distorted nonradial dipole pulsation mode .
we measure a 1.6 rotational variation in the mean radial velocities of h@xmath0 and argue that this is the first observation of h@xmath0 abundance spots caused by he settling through suppression of convection by the magnetic field on an oblique rotator , in support of a prime theory for the excitation mechanism of roap star pulsation .
we demonstrate that hd99563 is the second roap star to show aspect dependence of blue - to - red running wave line profile variations in ndiii spots .
[ firstpage ] stars : individual : hd99563 stars : pulsating stars : magnetic fields stars : atmospheres stars : chemically peculiar |
a key challenge for the future wireless networks is the increasing video traffic demand , which reached 70% of total mobile ip traffic in 2015 @xcite .
classical downlink systems can not meet this demand since they have limited resource blocks , and therefore as the number of simultaneous video transfers @xmath0 increases , the per - video throughput vanishes as @xmath1 .
recently it was shown that scalable per - video throughput can be achieved if the communications are synergistically designed with caching at the receivers . indeed , the recent breakthrough of _ coded caching _ @xcite has inspired a rethinking of wireless downlink .
different video sub - files are cached at the receivers , and video requests are served by coded multicasts . by careful selection of sub - file caching and exploitation of the broadcast wireless channel ,
the transmitted signal is simultaneously useful for decoding at users with different video requests .
although this scheme theoretically proved to scale well can potentially resolve the future downlink bottleneck , several limitations hinder its applicability in practical systems @xcite . in this work ,
we take a closer look to the limitations that arise from the fact that _ coded caching was originally designed for a symmetric error - free shared link . _
if instead we consider a realistic model for the wireless channel , we observe that a naive application of coded caching faces a _ short - term _ limitation : since the channel qualities of the users fluctuate over time and our transmissions need to reach all users , the transmissions need to be designed for the worst channel quality .
this is in stark contrast with standard downlink techniques , like _ opportunistic scheduling _
@xcite , which serve the user with the best instantaneous channel quality .
thus , a first challenge is to discover a way to allow coded caching technique to opportunistically exploit the fading of the wireless channel .
apart from the fast fading consideration , there is also a _ long - term _ limitation due to the network topology .
the user locations might vary , which leads to consistently poor channel quality for the ill - positioned users .
the classical coded caching scheme is designed to deliver equal video shares to all users , which leads to ill - positioned users consuming most of the air time and hence driving the overall system performance to low efficiency . in the literature , this problem has been resolved by the use of fairness among user throughputs @xcite . by allowing poorly located users to receive less throughput than others , precious airtime is saved and the overall system performance is greatly increased . since the sum throughput rate and equalitarian fairness are typically the two extreme cases , past works have proposed the use of alpha - fairness @xcite which allows to select the coefficient @xmath2 and drive the system to any desirable tradeoff point in between of the two extremes .
previously , the alpha - fair objectives have been studied in the context of ( i ) multiple user activations @xcite , ( ii ) multiple antennas @xcite and ( iii ) broadcast channels @xcite . however ,
here the fairness problem is further complicated by the interplay between scheduling and the coded caching operation . in particular
, we wish to shed light into the following questions : _ what is the right user grouping and how we should design the codewords to achieve our fairness objective while adapting to changing channel quality ? _ to address these questions , we study the content delivery over a realistic block - fading broadcast channel , where the channel quality varies across users and time . in this setting , we design a scheme that decouples transmissions from coding . in the transmission side ,
we select the multicast user set dynamically depending on the instantaneous channel quality and user urgency captured by queue lengths . in the coding side
, we adapt the codeword construction of @xcite depending on how fast the transmission side serves each user set . combining with an appropriate congestion controller , we show that this approach yields our alpha - fair objective .
more specifically , our approaches and contributions are summarized below : * we impose a novel queueing structure which decomposes the channel scheduling from the codeword construction .
although it is clear that the codeword construction needs to be adaptive to channel variation , our scheme ensures this through our _ backpressure _ that connects the user queues and the codeword queues .
hence , we are able to show that this decomposition is without loss of optimality . *
we then provide an online policy consisting of ( i ) admission control of new files into the system ; ( ii ) combination of files to perform coded caching ; ( iii ) scheduling and power control of codeword transmissions to subset of users on the wireless channel .
we prove that the long - term video delivery rate vector achieved by our scheme is a near optimal solution to the alpha - fair optimization problem under the specific coded caching scheme @xcite .
* through numerical examples , we demonstrate the superiority of our approach versus ( a ) opportunistic scheduling with unicast transmissions and classical network caching ( storing a fraction of each video ) , ( b ) standard coded caching based on transmitting - to - all .
since coded caching was first proposed @xcite and its potential was recognized by the community , substantial efforts have been devoted to quantify the gain in realistic scenarios , including decentralized placement @xcite , non - uniform popularities @xcite , and device - to - device ( d2d ) networks @xcite .
a number of recent works replace the original perfect shared link with wireless channels @xcite .
commonly in the works with wireless channels , the performance of coded caching is limited by the user in the worst channel condition because the wireless multicast capacity is determined by the worst user ( * ? ? ?
* chapter 7.2 ) .
this limitation of coded caching has been recently highlighted in @xcite , while similar conclusions and some directions are given in @xcite .
our work is the first to addresses this aspect by jointly designing the transmissions over the broadcast channel and scheduling appropriate subsets of users .
most past works deal with _ offline _ caching in the sense that both cache placement and delivery phases are performed once and do not capture the random and asynchronous nature of video traffic .
the papers @xcite addressed partly the online nature by studying cache eviction strategies , and delay aspects . in this paper
, we explore a different online aspect .
requests for video files arrive in an online fashion , and transmissions are scheduled over time - varying wireless channels .
online transmission scheduling over wireless channels has been extensively studied in the context of opportunistic scheduling @xcite and network utility maximization @xcite .
prior works emphasize two fundamental aspects : ( a ) the balancing of user rates according to fairness and efficiency considerations , and ( b ) the opportunistic exploitation of the time - varying fading channels .
related to our work are the studies of wireless downlink with broadcast degraded channels ; @xcite gives a maxweight - type of policy and @xcite provides a throughput optimal policy based on a fluid limit analysis .
our work is the first to our knowledge that studies coded caching in this setting .
the new element in our study is the joint consideration of user scheduling with codeword construction for the coded caching delivery phase .
we study a wireless downlink consisting of a base station and @xmath0 users .
the users are interested in downloading files over the wireless channel .
the performance metric is the _ time average delivery rate of files _ to user @xmath3 , denoted by @xmath4 .
hence our objective is expressed with respect to the vector of delivery rates @xmath5 .
we are interested in the _ fair file delivery _ problem : @xmath6 where @xmath7 denotes the set of all feasible delivery rate vectors clarified in the following subsection and the utility function corresponds to the _ alpha fair _ family of concave functions obtained by choosing : @xmath8 for some arbitrarily small @xmath9 ( used to extend the domain of the functions to @xmath10 ) . tuning the value of @xmath2 changes the shape of the utility function and
consequently drives the system performance @xmath11 to different points : ( i ) @xmath12 yields max sum delivery rate , ( ii ) @xmath13 yields max - min delivery rate @xcite , ( iii ) @xmath14 yields proportionally fair delivery rate @xcite .
choosing @xmath15 leads to a tradeoff between max sum and proportionally fair delivery rates .
the optimization is designed to allow us tweak the performance of the system ; we highlight its importance by an example .
suppose that for a 2-user system @xmath7 is given by the convex set shown on figure [ fig : example ] .
different boundary points are obtained as solutions to .
if we choose @xmath12 , the system is operated at the point that maximizes the sum @xmath16 .
the choice @xmath13 leads to the maximum @xmath17 such that @xmath18 , while @xmath19 maximizes the sum of logarithms .
the operation point a is obtained when we always broadcast to all users at the weakest user rate and use @xcite for coded caching transmissions .
note that this results in a significant loss of efficiency due to the variations of the fading channel , and consequently a lies in the interior of @xmath7 .
we may infer that the point @xmath13 is obtained by avoiding transmissions to users with instantaneous poor channel quality but still balancing their throughputs in the long run . to analyze the set of feasible rate vectors
@xmath7 we need to zoom in the detailed model of transmissions . * caching model .
* there are @xmath20 equally popular files @xmath21 , each @xmath22 bits long .
the files are available to the base station .
user @xmath3 is equipped with cache memory @xmath23 of @xmath24 bits , where @xmath25 $ ] .
caching placement is performed during off - peak hour , and the goal is to fill the caches up to the memory constraint with selected bits . to this end , we need to select @xmath0 _ caching functions _
@xmath26 which map the files @xmath21 into the cache contents @xmath27 the caching functions can be used to cache a few entire files , or a small fraction from each file , or even coded combinations of subfiles @xcite .
it is important to note that the caching functions are selected once , without knowledge of future requests , and are fixed throughout our system operation . * downlink channel model .
* we consider a standard block - fading broadcast channel , such that the channel state remains constant over a slot of @xmath28 channel uses and changes from one slot to another in an i.i.d . manner .
the channel output of user @xmath3 in any channel use of slot @xmath29 is given by @xmath30 where the channel input @xmath31 is subject to the power constraint @xmath32 \leq pt_{\rm slot}$ ] ; @xmath33 are additive white gaussian noises with covariance matrix identity of size @xmath28 , assumed independent of each other ; @xmath34 are channel fading coefficients @xmath35 independently distributed across time and users , with @xmath36 denoting the path - loss parameter of user @xmath3 . *
encoding and transmissions . *
the transmissions aim to contribute information towards the delivery of a specific vector of file requests @xmath37 , where @xmath38 denotes the index of the requested file by user @xmath3 in slot @xmath29 . here
@xmath20 is the video library size , typically in the order of 10k .
the requests are generated randomly , and whenever a file is delivered to user @xmath3 , the next request of this user will be for another randomly selected file . at each time slot , the base station observes the channel state @xmath39 and the request vector up to @xmath29 , @xmath40 , constructs a transmit symbol using the _ encoding function _ @xmath41 .
@xmath42 finally , it transmits a codeword @xmath43 for the @xmath28 channel uses over the fading broadcast channel in slot @xmath29 .
the encoding function may be chosen at each slot to contribute information to a selected subset of users @xmath44 .
this allows several possibilities , e.g. to send more information to a small set of users with good instantaneous channel qualities , or less information to a large set that includes users with poor quality .
* decoding .
* at slot @xmath29 , each user @xmath3 observes the local cache contents @xmath23 and the sequence of channel outputs so far @xmath45 and employs a _ decoding function _
@xmath46 to determine the decoded files .
let @xmath47 denote the number of files decoded by user @xmath3 after @xmath29 slots .
the decoding function @xmath46 is a mapping @xmath48 the decoded files of user @xmath3 at slot @xmath29 are given by @xmath49 , and depend on the channel outputs and states up to @xmath29 , the local cache contents , and the requested files of all users up to @xmath29 .
a file is incorrectly decoded if it does not belongs to the set of requested files .
the number of incorrectly decoded files are then given by @xmath50 and the number of correctly decoded files at time @xmath29 is : @xmath51 a rate vector @xmath52 is said to be _ feasible _ @xmath53 if there exist functions @xmath54 , [ f_t ] , [ \xi_k])$ ] such that : @xmath55 where the rate is measured in file / slot .
in contrast to past works which study the performance of one - shot coded caching @xcite , our rate metric measures the ability of the system to continuously deliver files to users .
finding the optimal policy is very complex . in this paper , we restrict the problem to specific class of policies given by the following mild assumptions : the admissible policies have the following characteristics : 1 .
the caching placement and delivery follow the decentralized scheme @xcite .
the users request distinct files , i.e. , the ids of the requested files of any two users are different .
since we restrict our action space , the delivery rate feasibility region , @xmath56 , of the class of policies @xmath57 is smaller than the one for the original problem @xmath7 .
however , these restrictions allow us to come up with a concrete solution approach .
note that the optimal cache and transmission design policy is already a very hard problem even in the simple case of broadcast transmissions with a fixed common rate , and the method in @xcite , @xcite are practical approaches with good performance .
in addition , looking at demand ids when combining files would be very complex and , because of the big library sizes , is not expected to bring substantial gains ( it is improbable that two users will make request for the same file in close time instances ) .
in this section we briefly review decentralized coded caching , first proposed in @xcite , and used by all admissible policies @xmath57 .
we set @xmath58 the normalized memory size . under the memory constraint of @xmath24 bits ,
each user @xmath3 independently caches a subset of @xmath59 bits of file @xmath60 , chosen uniformly at random for @xmath61 . by letting @xmath62
denote the sub - file of @xmath63 stored exclusively in the cache memories of the user set @xmath64 , the cache memory @xmath23 of user @xmath3 after decentralized placement is given by @xmath65 , \forall { { \cal j}}\ni k , \forall i = 1,\dots , n \}.\end{aligned}\ ] ] the size of each sub - file measured in bits is given by @xmath66 .
the above completely determine the caching functions .
once the requests of all users are revealed , the offline scheme proceeds to the delivery of the requested files ( delivery phase ) . assuming that user @xmath3 requests file @xmath3 , i.e. @xmath67 , the server generates and conveys the following codeword simultaneously useful to the subset of users @xmath64 : @xmath68 where @xmath69 denotes the bit - wise xor operation .
the main idea here is to create a codeword useful to a subset of users by exploiting the receiver side information established during the placement phase .
it is worth noticing that the _ coded _ delivery with xors significantly reduces the number of transmissions .
compared to uncoded delivery , where the sub - files are sent sequentially and the number of transmissions are equal to @xmath70 , the coded delivery requires the transmission of @xmath71 , yielding a reduction of a factor @xmath72 . in a practical case of @xmath73 , it has been proved that decentralized coded caching achieves the total number of transmissions , measured in the number of files , given by @xcite @xmath74 on the other hand , in uncoded delivery , the number of transmissions is given by @xmath75 since it exploits only _ local _ caching gain at each user .
for a system with @xmath76 users and normalized memory of @xmath77 , the minimum transmissions required by uncoded delivery is @xmath78 and that of decentralized coded caching is @xmath79 , yielding a gain of factor @xmath80 . in order to further illustrate the placement and delivery of decentralized coded caching
, we provide an three - user example . for the case of @xmath81 users in fig.[fig :
ex3 ] , let us assume that user 1 , 2 , 3 , requests file @xmath82 , respectively . after the placement phase , a given file @xmath83 will be partitioned into 8 subfiles .
codewords to be sent are the following * @xmath84 , @xmath85 and @xmath86 to user @xmath87 , @xmath79 and @xmath88 respectively .
* @xmath89 is intended to users @xmath90 .
once received , user @xmath87 decodes @xmath91 by combining the received codeword with @xmath92 given in its cache .
similarly user @xmath79 decodes @xmath92 .
the same approach holds for codeword @xmath93 to users @xmath94 and codeword @xmath95 to users @xmath96 * @xmath97 is intended users @xmath98 .
user @xmath87 can decode @xmath99 by combining the received codeword with @xmath100 given in its cache .
the same approach is used for user @xmath79 , @xmath88 to decode @xmath101 , @xmath102 respectively .
in this section , we address the question on how the transmitter shall convey private and multiple common messages , each intended to a subset of users , while opportunistically exploiting the underlying wireless channel .
we start by remarking that the channel in for a given channel realization @xmath103 corresponds to the gaussian degraded broadcast channel . without loss of generality ,
let us assume @xmath104 so that the following markov chain holds .
@xmath105 the capacity region of the degraded broadcast channel for @xmath0 private messages and a common message is well - known @xcite . in this section ,
we consider a more general setup where the transmitter wishes to convey @xmath106 mutually independent messages , denoted by @xmath107 , where @xmath108 denotes the message intended to the users in subset @xmath109 .
each user @xmath3 must decode all messages @xmath107 for @xmath110 . by letting @xmath111
denote the multicast rate of the message @xmath108 , we say that the rate - tuple @xmath112 is achievable if there exists encoding and decoding functions which ensure the reliability and the rate condition .
the capacity region is defined as the supremum of the achievable rate - tuple , where the rate is measured in bit / channel use .
[ th : capacity_bc ] the capacity region @xmath113 of a @xmath0-user degraded gaussian broadcast channel with fading gains @xmath114 and @xmath106 independent messages @xmath107 is given by @xmath115 for non - negative variables @xmath116 such that @xmath117 .
please refer to appendix [ appendix : bc_capacity_proof ] for the proof .
the achievability builds on superposition coding at the transmitter and successive interference cancellation at receivers . for @xmath81 ,
the transmit signal is simply given by @xmath118 where @xmath119 are mutually independent gaussian distributed random variables satisfying the power constraint and @xmath120 denotes the signal corresponding to the message @xmath108 intended to the subset @xmath121 .
user 3 ( the weakest user ) decodes @xmath122 by treating all the other messages as noise .
user 2 decodes first the messages @xmath123 and then jointly decodes @xmath124 .
finally , user 1 ( the strongest user ) successively decodes @xmath125 and , finally , @xmath126 . later in our online coded caching scheme
we will need the capacity region @xmath113 , and more specifically , we will need to characterize its boundary . to this end
, it suffices to consider the weighted sum rate maximization : @xmath127 we first simplify the problem using the following theorem . the weighted sum rate maximization with @xmath106 variables in reduces to a simpler problem with @xmath0 variables , given by @xmath128 where @xmath129 denotes the largest weight for user @xmath3 @xmath130 the proof builds on the simple structure of the capacity region .
we first remark that for a given power allocation of other users , user @xmath3 sees @xmath131 messages @xmath132 for all @xmath64 such that @xmath133 with the equal channel gain . for a given set of @xmath134 , the capacity region of these messages is a simple hyperplane characterized by @xmath131 vertices @xmath135 for @xmath136 , where @xmath137 is the sum rate of user @xmath3 in the rhs of and @xmath138 is a vector with one for the @xmath60-th entry and zero for the others .
therefore , the weighted sum rate seen is maximized for user @xmath3 by selecting the vertex corresponding to the largest weight , denoted by @xmath139 .
this holds for any @xmath3 .
we provide an efficient algorithm to solve this power allocation problem as a special case of the parallel gaussian broadcast channel studied in ( * ? ? ?
* theorem 3.2 ) .
following @xcite , we define the rate utility function for user @xmath3 given by @xmath140 where @xmath141 is a lagrangian multiplier .
the optimal solution corresponds to selecting the user with the maximum rate utility at each @xmath142 and the resulting power allocation for user @xmath3 is @xmath143_+ = u_k(z ) \right \}/p\end{aligned}\ ] ] with @xmath141 satisfying @xmath144_+.\end{aligned}\ ] ]
this section presents first the queued delivery network and its feasible rate region of arrival rates , then describes the proposed control policy . at each time slot @xmath29
, the controller _ admits _ @xmath145 files to be delivered to user @xmath3 , and hence @xmath145 is a control variable . as our model dictates , the succession of requested files for user @xmath3 is determined uniformly at random
. * queueing model . *
the base station organizes the information into the following types of queues : 1 . * user queues * to store admitted files , one for each user
. the buffer size of queue @xmath3 is denoted by @xmath146 and expressed in number of files .
* codeword queues * to store codewords to be multicast .
there is one codeword queue for each subset of users @xmath147 .
the size of codeword queue @xmath64 is denoted by @xmath148 and expressed in bits . a queueing policy @xmath149 performs the following operations : ( i ) decides how many files to admit into the user queues @xmath146 in the form of @xmath150 variables , ( ii )
then it decides how to combine together files from different user queues to be encoded into the form of multiple codewords which represent the required broadcast transmissions for the reception of this file
these codewords are stored in the appropriate codeword queues @xmath148 , ( iii ) and last it decides the encoding function @xmath151 .
( ii ) and ( iii ) are further clarified in the next section .
a queue @xmath152 is said to be ( strongly ) stable if @xmath153 < \infty.\ ] ] a queueing system is said to be stable if all its queues are stable . moreover , the stability region of a system is the set of all arrival rates such that the system is stable .
the above definition implies that the average delay of each job in the queue is finite . in our problem , if we develop a policy that keeps _ user queues _
@xmath154 stable , then all admitted files will , at some point , be combined into codewords . if in addition _
codeword queues _
@xmath155 are stable , then all generated codewords will reach their destinations , meaning that all receivers will be able to decode the admitted files that they requested .
[ lem : equivalence ] the region of all feasible delivery rates @xmath7 is the same as the stability region of the system ( i.e. the set of all demand arrival rates for which there exists a policy that stabilizes the queueing system ) .
let @xmath156,$ ] denote the time average number of admitted files for user @xmath3 .
lemma [ lem : equivalence ] implies the following corollary .
[ cor : equivalentoptimization ] solving is equivalent to finding a policy @xmath149 such that @xmath157 see appendix [ appendix : equivalence_proof ] contrary to the offline coded caching in @xcite ,
we propose an online delivery scheme consisting of the following three blocks . each block is operated at each slot . 1 .
* admission control : * at the beginning of each slot , the controller decides how many requests for each user , @xmath145 should be pulled into the system from the infinite reservoir .
* routing : * the cumulative accepted files for user @xmath3 are stored in the admitted demand queue whose size is given by @xmath146 for @xmath158 .
the server decides the combinations of files to perform coded caching .
the decision at slot @xmath29 for a subset of users @xmath159 , denoted by @xmath160 , refers to the number of combined requests for this subset of users .
it is worth noticing that offline coded caching lets @xmath161 for @xmath162 and zero for all the other subsets .
the size of the queue @xmath163 evolves as : @xmath164^+ + a_k(t ) \label{eq : codewordqueues}\end{aligned}\ ] ] if @xmath165 , the server creates codewords by applying offline coded caching explained in section [ * * ] for this subset of users as a function of the cache contents @xmath166 .
* scheduling : * the codewords intended to the subset @xmath64 of users are stored in codeword queue whose size is given by @xmath167 for @xmath168 .
given the instantaneous channel realization @xmath169 and the queue state @xmath170 , the server performs scheduling and rate allocation .
namely , at slot @xmath29 , it determines the number @xmath171 of bits per channel use to be transmitted for the users in subset @xmath172 . by letting @xmath173
denote the number of bits generated for codeword queue @xmath174 when offline coded caching is performed to the users in @xmath64 , codeword queue @xmath172 evolves as @xmath175^+ + \sum_{{{\cal j}}:{{\cal i}}\subseteq{{\cal j}}}b_{{{\cal j}},{{\cal i}}}\sigma_{{{\cal j}}}(t ) \end{aligned}\ ] ] where @xmath176 . in
order determine our proposed policy , namely the set of decisions @xmath177 at each slot @xmath29 , we first characterize the feasible region @xmath7 as a set of arrival rates @xmath178 .
we let @xmath179 denote the probability that the channel state at slot @xmath29 is @xmath180 where @xmath181 is the set of all possible channel states .
we let @xmath113 denote the capacity region for a fixed channel state @xmath182 .
then we have the following [ th : feasibilityregion ] a demand rate vector is feasible , i.e. @xmath183 , if and only if there exist @xmath184 , @xmath185 , \forall { { \cal i}}\subseteq \{1,\dots , k\}$ ] such that : @xmath186 constraint says that the service rate at which admitted demands are combined to form codewords is greater than the arrival rate , while implies that the long - term average transmission rate @xmath187 for the subset @xmath172 of users should be higher than the rate at which bits of generated codewords for this group arrive . in terms of the queueing system defined , these constraints impose that the service rates of each queue should be greater than their arrival rates , thus rendering them stable .
theorem [ th : feasibilityregion ] implies that the set of feasible average delivery rates is a convex set . in order to perform the utility maximization
, we need to introduce one more set of queues .
these queues are virtual , in the sense that they do not hold actual file demands or bits , but are merely counters to drive the control policy .
each user @xmath3 is associated with a queue @xmath188 which evolves as follows : @xmath189^+ + \gamma_k(t)\end{aligned}\ ] ] where @xmath190 represents the arrival process to the virtual queue and is given by @xmath191\end{aligned}\ ] ] in the above , @xmath192 is a parameter that controls the utility - delay tradeoff achieved by the algorithm ( see theorem [ th : optimality_infinite ] ) .
the general intuition here is as follows : observe that the number @xmath145 of admitted demands is the service rate for the virtual queues @xmath188 .
the control algorithm actually seeks to optimize the time average of the virtual arrivals @xmath190 . however , since @xmath188 is stable , its service rate , which is the actual admission rate , will be greater than the rate of the virtual arrivals , therefore giving the same optimizer .
stability of all other queues will guarantee that admitted files will be actually delivered to the users .
we present our on - off policy for admission control and routing . for every user @xmath3 ,
admission control chooses @xmath145 demands given by @xmath193 for every subset @xmath109 , routing combines @xmath194 demands of users in @xmath64 given by @xmath195 in order to stabilize all _ codeword queues _ , the scheduling and resource allocation explicitly solve the following weighted sum rate maximization at each slot @xmath29 where the weight of the subset @xmath64 corresponds to the queue length of @xmath196 @xmath197 we propose to apply the power allocation algorithm in section [ section : broadcasting ] to solve the above problem by sorting users in a decreasing order of channel gains and treating @xmath148 as @xmath198 .
in adition , we assume that the number of channel uses in one coherence block is large enough such that the decoding error from choosing channel codes with rate @xmath199 is very small . in this case
, no feedback from the receivers is given .
we conclude this section by providing an example of our proposed online delivery network for @xmath81 users as illustrated in fig .
[ fig : queueing_system_real ] . users .
dashed lines represent wireless transmissions , solid circles files to be combined and solid arrows codewords generated . ] at slot @xmath29 the server decides to combine @xmath200 requested by user @xmath79 with @xmath201 requested by user @xmath79 and to process @xmath202 requested by user @xmath87 uncoded
. therefore @xmath203 and @xmath204 otherwise . given this codeword construction ,
codeword queues have inputs as described in table i. [ tab:1 ] .codeword queues inputs . [ cols="^,^",options="header " , ] in addition , data from queues @xmath205 are transmitted .
in thi section , we present the main result of the paper , that our proposed online algorithm leads to close to optimal performance for all policies in the class @xmath57 : [ th : optimality_infinite ] let @xmath206 the mean time - average delivery rate for user @xmath3 achieved by the proposed policy .
then @xmath207 where @xmath208 is the sum of all queue lengths at the beginning of time slot @xmath29 , thus a measure of the mean delay of file delivery .
the above theorem states that , by tuning the constant @xmath209 , the utility resulting from our online policy can be arbitrarily close to the optimal one , where there is a tradeoff between the guaranteed optimality gap @xmath210 and the upper bound on the total buffer length @xmath211 . for proving the theorem
, we use the lyapunov function @xmath212 and specifically the related drift - plus - penalty quantity , defined as : @xmath213 .
the proposed algorithm is such that it minimizes ( a bound on ) this quantity .
the main idea is to use this fact in order to compare the evolution of the drift - plus - penalty under our policy and two `` static '' policies , that is policies that take random actions ( admissions , demand combinations and wireless transmissions ) , drawn from a specific distribution , based only on the channel realizations ( and knowledge of the channel statistics ) .
we can prove from theorem 4 that these policies can attain every feasible delivery rate .
the first static policy is one such that it achieves the stability of the system for an arrival rate vector @xmath214 such that @xmath215 . comparing with our policy , we deduce strong stability of all queues and the bounds on the queue lengths by using a foster - lyapunov type of criterion . in order to prove near - optimality
, we consider a static policy that admits file requests at rates @xmath216 and keeps the queues stable in a weaker sense ( since the arrival rate is now in the boundary @xmath56 ) . by comparing the drift - plus - penalty quantities and using telescopic sums and jensen s inequality on the time average utilities ,
we obtain the near - optimality of out proposed policy .
the full proof is in appendix [ appendix : performance_proof ] .
in this section , we compare our proposed delivery scheme with the following two other schemes , all building on decentralized cache placement in and . * * unicast opportunistic scheduling * : for any request , the server sends the remaining @xmath217 bits to the corresponding user without combining any files .
here we only exploit the local caching gain . in each slot
the serve sends with full power to user @xmath218 where @xmath219 is the empirical average rate for user @xmath3 up to slot @xmath29 . * * standard coded caching * : we use decentralized coded caching among all @xmath0 users . for the delivery ,
non - opportunistic tdma transmission is used .
the server sends sequentially codewords @xmath220 to the subset of users @xmath64 at the weakest user rate among @xmath64 : @xmath221once the server has sent codewords @xmath222 , every user is able to decode one file
. then the process is repeated for all the demands .
we consider the system with normalized memory of @xmath223 , power constraint @xmath224 , file size @xmath225 bits and number of channel uses per slot @xmath226 .
we divide users into two classes of @xmath227 users each : strong users with @xmath228 and weak users with @xmath229 .
we compare the three algorithms for the cases where the objective of the system is sum rate maximization ( @xmath12 ) and proportional fairness ( @xmath19 ) .
the results are depicted in fig .
[ subfig-1:sumrates ] and [ subfig-2:utilities ] , respectively .
regarding the sum rate objective , standard coded caching performs very poorly , indicative of the adverse effect of users with bad channel quality .
it is notable that our proposed scheme outperforms the unicast opportunistic scheme , which maximizes the sum rate if only private information packets are to be conveyed .
the relative merit of our scheme increases as the number of users grows .
this can be attributed to the fact that our scheme can exploit any available multicast opportunities . _
our result here implies that , in realistic wireless systems , coded caching can indeed provide a significant throughput increase when an appropriate joint design of routing and opportunistic transmission is used .
_ regarding the proportional fair objective , we can see that the average sum utility increases with a system dimension for three schemes although our proposed scheme provides a gain compared to the two others .
we provided an algorithm to solve the problem of ensuring fairness in the long term delivery rates in wireless systems employing decentralized coded caching .
our results imply that appropriately combining the opportunism arising from the fading channels with the multicasting opportunities that arise from coded caching can mitigate the harmful impact of users with bad channel conditions in standard coded caching schemes and provide significant increase in the performance of the system .
we provide the converse proof for @xmath81 and the general case @xmath230 follows readily .
notice that the channel output of user @xmath3 in for @xmath231 channel use can be equivalently written as @xmath232 where @xmath233 for @xmath234 and @xmath235 identity matrix of size @xmath231 . since @xmath236 , we set @xmath237 } m_{{{\cal k}}}$ ] the message that must be decoded by user @xmath3 ( user @xmath3 decodes all bits that user @xmath238 decodes ) at rate @xmath239 .
more explicitly , @xmath240 , @xmath241 , @xmath242 . by fano s inequality ,
we have @xmath243 consider @xmath244 since @xmath245 , there exist @xmath246 such that @xmath247 using and we obtain @xmath248 next consider @xmath249 using the conditional entropy power inequality in @xcite , we have @xmath250 and imply @xmath251 equivalent to @xmath252 since @xmath253 , there exists @xmath254 such that @xmath255 and @xmath256 using , and it follows @xmath257 superposition coding achieves the upper bound . for @xmath273 , generate random sequences @xmath274 , @xmath275 $ ] each i.i.d . @xmath276 . to transmit a triple message @xmath277 the encoder set @xmath278 . for decoding : * receiver @xmath88 recover @xmath279 from @xmath280 by considering @xmath281 as noise .
the probability of error tends to zero as @xmath282 if @xmath283 .
* receiver @xmath79 uses successive cancellation .
first , it decodes @xmath279 from @xmath284 by considering @xmath281 as noise .
the probability of error tends to zero as @xmath282 if @xmath285 .
since @xmath286 and @xmath287 , the later condition is satisfied .
second , it subtracts off @xmath288 and recover @xmath289 from @xmath290 by treating @xmath291 as noise .
the probability of error tends to zero as @xmath282 if @xmath292 . *
receiver @xmath87 uses successive cancellation twice .
first , it decodes @xmath279 from @xmath293 by considering @xmath281 as noise .
the probability of error tends to zero as @xmath282 if @xmath294 .
since @xmath295 and @xmath287 , the later condition is satisfied .
second , it subtracts off @xmath288 and decodes @xmath289 by treating @xmath291 as noise .
the probability of error tends to zero as @xmath282 if @xmath296 .
since @xmath297 and @xmath298 , the later condition is satisfied .
last , it subtracts off @xmath289 and recover @xmath291 .
the probability of error tends to zero as @xmath282 if @xmath299 . from lemma [ lem : equivalence ] and corollary [ cor : equivalentoptimization ]
, it suffices to prove that under the online policy the queues are strongly stable and the resulting time average admission rates maximize the desired utility function subject to minimum rate constraints .
we first look at policies that take random decisions based only on the channel realizations .
since the feasibility region @xmath56 is a convex set ( see theorem ) , any point in it can be achieved by properly time - sharing over the possible control decisions .
we focus on two such policies , one that achieves the optimal utility and another on that achieves ( i.e. admits and stabilizes the system for that ) a rate vector in th @xmath314 interior of @xmath56 .
we then have the following lemmas : [ lem : staticoptimalpolicy ] define a policy @xmath315 that in each slot where the channel states are @xmath182 works as follows : ( i ) it pulls random user demands with mean @xmath316 , and it gives the virtual queues arrivals with mean @xmath317 as well ( ii ) the number of combinations for subset @xmath64 is a random variable with mean @xmath318 and uniformly bounded by @xmath319 , ( iii ) selects one out of @xmath320 suitably defined rate vectors @xmath321 with probability @xmath322 .
the parameters above are selected such that they solve the following problem : @xmath323 then , @xmath324 results in the optimal delivery rate vector ( when all possible policies are restricted to set @xmath325 ) .
[ lem : stablepolicy ] define a policy @xmath326 that in each slot where the channel states are @xmath182 works as follows : ( i ) it pulls random user demands with mean @xmath327 such that @xmath328 , and gives the virtual queues
random arrivals with mean @xmath329 for some @xmath330 ( ii ) the number of combinations for subset @xmath64 is a random variable with mean @xmath331 and uniformly bounded by @xmath319 , ( iii ) selects one out of @xmath320 suitably defined rate vectors @xmath321 with probability @xmath332 .
the parameters above are selected such that : @xmath333 for some appropriate @xmath334 .
then , the system under @xmath335 has mean incoming rates of @xmath336 and is strongly stable .
the proof of the performance of our proposed policy is based on applying lyapunov optimization theory @xcite with the following as lyapunov function @xmath337 defining its drift as @xmath338 , using the queue evolution equations and the fact that @xmath339^+)^2\leq x^2 $ ] , we have @xmath340 where @xmath341 is a constant that depends only on the parameters of the system . adding the quantity @xmath342 to both hands of and rearranging the right hand side
, we have @xmath343 now observe that the control algorithm minimizes right hand side of given the channel state @xmath169 ( for any channel state ) .
therefore , taking expectations over the channel state distributions , for every vectors @xmath344^k , \bar{\mathbf{\gamma}}\in [ 1,\gamma_{max}]^k , \bar{\mathbf{\sigma}}\in conv(\{0, .. ,\sigma_{max}\}^m ) , \bar{\mu}\in \sum_{{\pmb{h}}\in\mathcal{h}}\pi_{{\pmb{h}}}\gamma({\pmb{h}})$ ] it holds that @xmath345 we will use to compare our policy with the static policies defined in lemmas [ lem : staticoptimalpolicy ] , [ lem : stablepolicy ] . more specifically , replacing the time averages we get from the static stabilizing policy @xmath335 of lemma [ lem : stablepolicy ] for some @xmath346
, we get that thre exist @xmath347 such that @xmath348 since @xmath349 , it follows that @xmath350 , therefore , we have from the foster - lyapunov criterion that the system @xmath351 has a unique stationary probability distribution , under which the mean queue lengths are finite . , then @xmath352 , i.e. their length is deterministically bounded ] therefore the queues are strongly stable under our proposed policy .
* proof of near optimal utility : * here we compare @xmath149 with the static optimal policy @xmath324 from lemma [ lem : staticoptimalpolicy ] . since @xmath324 takes decisions irrespectively of the queue lengths , we can replace quantities @xmath353 with the time averages corresponding to @xmath324 , i.e. @xmath354 .
we thus have : @xmath355 taking expectations over @xmath356 for both sides and summing the inequalities for @xmath357 we get @xmath358 assuming @xmath359 ( this assumption is standard in this line of work , for example it holds if the system starts empty ) , taking the limit as @xmath360 goes to infinity gives @xmath361 in addition , since @xmath362 are concave , jensen s inequality implies @xmath363 proving the near optimality of the online policy follows from the above and the fact that @xmath364 ( since the virtual queues @xmath188 are strongly stable ) .
g. caire , r. r. muller and r. knopp , hard fairness versus proportional fairness in wireless communications : the single - cell case , " in ieee trans .
theory , vol .
1366 - 1385 , april 2007 .
denote @xmath300 the number of files that have been admitted to the system for user @xmath3 up to slot @xmath29 . also , note that due to our restriction on the class of policies @xmath57 and our assumption about long enough blocklengths , there are no errors in decoding the files , therefore the number of files correctly decoded for user @xmath3 till slot @xmath29 is @xmath47 . since @xmath301 , if suffices to show that for every arrival rate vector @xmath183 , there exists a policy in @xmath57 for which the delivery rate vector is @xmath302 .
we shall deal only with the interior of @xmath56 ( arrival rates at the boundaries of stability region are exceptional cases ) .
take any arrival rate vector @xmath303 . from (
* theorem 4.5 ) it follows that for any there exists a randomized demand combination and transmission policy @xmath304 , the probabilities of which depending only on the channel state realization each slot , for which the system is strongly stable .
in addition , any arrival rate vector can be constructed via a randomized admission policy . since the channels are i.i.d .
random with a finite state space and queues are measured in files and bits , the system now evolves as a discrete time markov chain @xmath305 , which can be checked that is aperiodic , irreducible ad with a single communicating class . in that case
, strong stability means that the markov chain is ergodic with finite mean .
further , this means that the system reaches to the set of states where all queues are zero infinitely often .
let @xmath306 $ ] be the number of timeslots between the @xmath307th and @xmath308th visit to this set ( we make the convention that @xmath309 $ ] is the time slot that this state is reached for the first time ) . in addition , let @xmath310 , \hat{d}_k[n]$ ] be the number of demands that arrived and were delivered in this frame , respectively . then , since within this frame the queues start and end empty , we have @xmath311 = \hat{d}_k[n ] , \forall n , \forall k.\ ] ] | the performance of existing _ coded caching _
schemes is sensitive to worst channel quality , a problem which is exacerbated when communicating over fading channels . in this paper
we address this limitation in the following manner : _ in short - term _ , we allow transmissions to subsets of users with good channel quality , avoiding users with fades , while _ in long - term _ we ensure fairness across the different users . our online scheme combines ( i ) joint scheduling and power control for the broadcast channel with fading , and ( ii ) congestion control for ensuring the optimal long - term average performance .
we restrict the caching operations to the decentralized scheme of @xcite , and subject to this restriction we prove that our scheme has near - optimal overall performance with respect to the convex alpha - fairness coded caching optimization . by tuning the coefficient alpha ,
the operator can differentiate user performance with respect to video delivery rates achievable by coded caching .
we demonstrate via simulations our scheme s superiority over legacy coded caching and unicast opportunistic scheduling , which are identified as special cases of our general framework .
broadcast channel , coded caching , fairness , lyapunov optimization . |
@xcite discovered the first group of b - type hot subdwarf ( sdb ) pulsators , which they termed ec14026 after the prototype and are now designated as v361hya stars , and referred to as sdbv stars .
this discovery opened the possibility of constraining the interiors of sdb stars through asteroseismology , which , in turn , may help to better understand their still controversial evolution .
so far , about 40 class members have been found , with periods ranging from 60 to 600s and typical amplitudes of 10mmag , with some exceptions of higher amplitudes up to 60mmag .
theoretical studies by @xcite demonstrated that their oscillations could be explained by acoustic ( @xmath1- ) modes of low degree , @xmath6 , and low radial orders , @xmath11 , driven by an opacity bump caused by a local enhancement of iron - group elements . in 2002 another class of pulsating sdb stars was found @xcite .
the new class is characterised by lower amplitudes ( 12mmag ) and longer periods ( @xmath81hour ) .
the new variable stars are referred to as long - period sdbs , or pg1716 stars , after that prototype .
@xcite determined that their variability could be explained in terms of high radial order gravity ( @xmath0- ) modes .
the same @xmath12 mechanism that operates in v361 hya stars also works in pg1716 stars , although theoretical unstable modes were at slightly lower effective temperatures than observed .
@xcite , using op opacity tables and considering an enhancement of both iron and nickel in the driving zone , were able to obtain unstable modes at the same effective temperatures as the observed ones
. additionally , their non - adiabatic analysis led to simultaneous instability of @xmath1- and @xmath0-modes , which can account for the two known hybrid sdb pulsators .
the two classes ( v361 hya and pg1716 stars ) of pulsating sdbs have slightly different effective temperatures and surface gravities , the former being hotter and denser than the latter .
their pulsating behaviour resembles two other pairs of pulsating stars showing @xmath1- and @xmath0-modes : @xmath13 cephei and slowly pulsating b stars and @xmath14 scuti and @xmath15 doradus stars .
balloon090100001 ( hereafter bal09 ) is one of the brightest pulsating sdb stars discovered to date .
it was found to be an sdb star by @xcite , and a short - period pulsating sdb by @xcite .
it has relatively long periods and large amplitudes of oscillation , which makes this object an excellent candidate for follow - up photometry .
bal09 was observed in a long - time base campaign independently by two of us , in august and september 2004 using the 60-cm telescope at mt .
suhora observatory ( ab ) and the 80-cm iac80 telescope at tenerife ( ro ) . four wide - band filters , @xmath2 , were used in the former ( * ? ? ?
* hereafter bar05 ) , while only @xmath3-filter data were obtained in the latter ( * ? ? ?
* hereafter ore05 ) .
the use of a common filter allowed an analysis of the combined data , from which almost 50 frequencies were detected @xcite .
these campaigns revealed a rich spectrum of frequencies clustering in groups within the @xmath1-mode domain , and having lower amplitudes as the frequency increases .
a high - amplitude frequency ( @xmath860mmag ) dominates the spectrum , with an equally - spaced triplet nearby . assuming that the triplet splitting is caused by rotation , a rotational period near 7days was derived .
interestingly , variability near 4mmag was also detected in the low - frequency region , which is typical for pg1716 stars .
thus , bal09 turned out to be a hybrid object and became very interesting from the seismological point of view .
the star was the second hybrid sdb pulsator discovered , the first being hs0702 + 6043 @xcite , which shows a single frequency in the @xmath0-mode region .
despite the frequency resolution achieved with the 2004 observations , the spectrum of bal09 suffered from strong aliasing , hindering a clear frequency identification at lower amplitudes .
closely - spaced frequencies and/or amplitude variability made the situation even worse . [ cols="^,^,^,^,>,>,^",options="header " , ] [ models ] in fig .
[ esquemap ] a schematic spectrum of bal09 in the @xmath1-mode domain is presented . as can be seen , all @xmath1-modes except @xmath16 cluster in four narrow ( from 0.05 to 0.12 mhz wide ) frequency groups near 2.8 , 3.8 , 4.7 and 5.5 mhz , consisting of 9 , 22 , 24 , and 16 frequencies , respectively ( table [ a_fi_2005 ] ) .
-modes occur .
the dashed line corresponds to @xmath16.,width=313 ] using combined multicolour photometry and spectroscopy of bal09 , @xcite identified @xmath7 as a radial mode and constrained the @xmath6 value for the 2.825mhz triplet components and @xmath9 . for a complete identification of @xmath7 , its radial order ( @xmath11 ) needs to be determined . assuming that @xmath7 is the fundamental radial mode ( @xmath6 = 0 , @xmath11 = 1 ) and using the grid of sdb models of @xcite , bar05 came to a reasonably good agreement between the observed frequency and the location of bal09 in the @xmath17
@xmath18 diagram ( see fig . 13 in bar05 ) .
the first overtone was not regarded as an alternative for @xmath7 by bar05 , however .
let us consider such a case .
the period of this fundamental radial mode ( @xmath19 ) would be of the order of 356 s/0.74 @xmath20 480 s , which is not detected in our data sets . )
found from theoretical models ; see also fig .
[ ratios ] . ]
model periods can be matched within the spectroscopic @xmath17 and @xmath21 constraints only if slightly more massive models than those of @xcite are considered , e.g. some models on sequence a in fig . [ dhr ] .
we can also find an appropriate model if we assume that @xmath7 is the second radial overtone , though the match is worse . in this case
@xmath19 equals to 593 s. however , all models of a given @xmath19 have nearly the same value of @xmath17 .
this can be explained as the period of the fundamental radial mode depends primarily on average density , i.e. , stellar radius . for a set of models that do not differ very much in mass
this implies that models with the same pulsation period would have approximately the same @xmath17 .
this can be seen , for example , in the bottom panel of fig .
13 of bar05 . as the spectroscopic error box for @xmath17 and @xmath21 is still quite large ( fig .
[ dhr ] ) we can not constrain @xmath11 for @xmath7 since both the fundamental and first overtone are allowable , and even the second overtone can not be excluded .
while only higher overtones can be rejected , we can constrain @xmath17 for our three possibilities to 5.51 ( @xmath7 fundamental ) , 5.34 ( 1st overtone ) , or 5.22 ( 2nd overtone ) .
improved constraints on @xmath17 will distinguish between these three possibilities .
however , we prefer to assume that @xmath7 is the fundamental radial pulsation as this is the most likely possibility . if @xmath7 were the first overtone then the fundamental radial pulsation ( and also some non - radial @xmath1 modes with similar frequencies ) should occur near 2.1 mhz and these pulsations are not observed . .
two sets of models were used : ( i ) those described in section [ ev_models ] and listed in table [ models ] ( large symbols ) , ( ii ) the sequences 5 to 7 from @xcite ( small symbols ) .
three horizontal strips denote the observed ranges of frequency ratios . all modes from @xmath1-mode groups near 3.8
, 4.7 and 5.5 mhz are considered.,width=313 ] another argument in favour of @xmath7 as the fundamental radial mode is the overtone ratio .
if we assume that at least one mode in the 3.8mhz group is also radial then the overtone ratio is 2.807mhz/([email protected])mhz @xmath20 [email protected] which can be compared with theoretical ratios , as is done in fig .
[ ratios ] .
it can be seen that for most models the @xmath23 ratio is near to 0.74 and only for model c is also @xmath24 allowed in some of its evolutionary phases .
the latter case implies @xmath7 to be the first overtone . as such
, we can conclude that if any of the 3.8 mhz - region frequencies are radial , then @xmath7 is most likely the fundamental radial pulsation . additionally , fig .
[ ratios ] indicates that by choosing @xmath7 to be the fundamental radial mode , the 4.7 mhz and 5.5 mhz groups match overtone ratios for @xmath25 and @xmath26 .
it seems unlikely that such a good match is serendipitous . reversing this reasoning
we may therefore conclude that each of the groups of @xmath1-modes in bal09 may contain a radial mode ( if so , we know which overtone it is ) .
alternatively , the same could be true for consecutive overtones of non - radial modes as they have period ratios similar to the radial ones , unless when avoided crossing occurs .
modes : near 2.8 , 3.8 , 4.7 , and 5.5 mhz.,width=313 ] considering splittings , we have searched for additional multiplets using the same frequency spacing as observed for multiplets at 2.8 mhz .
because of the high frequency density in these regions , there are likely rotationally split multiplets
. however , the simple structure of frequencies in the 2.8mhz group contrasts with more complicated frequency patterns in the three remaining groups of @xmath1-mode frequencies ( fig . [ pattern ] ) .
another likely multiplet is the symmetric triplet of @xmath27 , @xmath28 , and @xmath29 ( table [ a_fi_2005 ] ) with splittings of 1.440 @xmath30 0.021 and 1.446 @xmath30 0.023 @xmath5hz .
bearing in mind the richness of the spectrum , observed amplitude variations , and uncertainties caused by aliasing , we prefer not to speculate further on additional multiplets until better data more fully reveal the frequency spectrum .
the high frequency density also indicates that all groups of @xmath1-modes must have @xmath31 2 modes , except those in the 2.8 mhz region .
bal09 shows a very rich spectrum in the low - frequency domain , which theoretically corresponds to @xmath0-modes . in fig .
[ schemodosg ] , the schematic amplitude spectrum of bal09 in this frequency range is displayed . for comparison ,
theoretical frequencies of a representative sdb model in sequence c are also shown .
we did not intend to find a model which matches the observed frequencies , but just to compare the density of theoretical and observed frequencies in this region . as can be seen in fig .
[ schemodosg ] , at least @xmath6 = 1 and 2 modes are needed to account for the observed frequencies .
however , as the density of theoretical frequencies increases with @xmath6 , the observed spectrum would be better reproduced if higher degree modes were considered .
-modes with @xmath6 = 1 , 2 and 3 .
the latter are shown for a representative sdb model from sequence c. numbers on the right indicate the value of @xmath11 for the first @xmath0-mode below 1 mhz.,width=313 ] we also notice a lack of frequencies between 0.4 and 0.6mhz .
either they have amplitudes below our detection threshold or they are not excited .
we investigate the possibility that they are not excited and show in fig .
[ trabajo ] the derivative of the work integral ( @xmath32 ) for @xmath0-modes with @xmath6 = 2 and radial orders @xmath33 = 1 , 7 and 30 .
these modes were chosen as representative of low , intermediate and high radial orders .
we used the graco non - adiabatic code @xcite to compute @xmath32 , which is positive in regions contributing to driving oscillations , while negative in damping regions .
for hot b subdwarfs the driving zone is related to the iron - peak elements opacity bump . in fig .
[ trabajo ] , the rosseland mean opacity for theoretical model c used is plotted as a solid line .
[ trabajo ] shows also that , while the region of the bump ( shown in gray ) contributes to the excitation of low- and high - radial order @xmath0-modes ( @xmath34 0 there ) , it has a negligible effect in driving intermediate - radial order modes .
this result could support the hypothesis that @xmath0-modes with intermediate radial orders are not excited in bal09 .
non - adiabatic computations by @xcite also result in exciting low- and high - order @xmath0-modes but damping intermediate ones for some models .
this seems a reasonable explanation for the lack of observed frequencies between 0.4 and 0.6mhz .
theoretical @xmath0-modes with low radial orders can have frequencies up to @xmath82.3mhz ( not shown in fig . [ schemodosg ] ) .
we detect three frequencies ( @xmath35 , @xmath36 and @xmath37 ) between 1.8 and 2.3 mhz which could then be attributed to low radial order @xmath0-modes .
= 2 , @xmath11 = @xmath381 , @xmath387 , and @xmath3830 @xmath0-modes as a function of a fractional stellar radius for a representative model from sequence c. regions contributing to driving ( damping ) produce positive ( negative ) @xmath32 values . the rosseland mean opacity is included as a solid line and plotted in arbitrary units .
the gray area delimits the driving region associated with an iron - group opacity bump.,width=313 ] most @xmath0-modes observed have periods in the range between 2500 and 7200s , although a few extend this range to 120010300s . theoretical @xmath0-modes with @xmath6 = 13 and periods in the range 25007200s ( fig . [ schemodosg ] )
have radial orders in the range @xmath33 @xmath81180 , which extends to @xmath39 = 5115 , if we include all of the observed periods .
with high - radial orders , we can expect asymptotic behaviour which results in equally - spaced periods @xcite .
however , @xcite and @xcite have shown that sdbs might display trapped @xmath0-modes , mainly caused by the chemical transition between the he and h radiative layers in the envelope .
mode trapping causes some periods to move away from asymptotic behaviour . in fig .
[ deltag ] , we show a schematic @xmath0-mode spectrum _ in periods _ and indicate the differences ( in seconds ) between adjacent periods
. however , the observed period spectrum does not appear equally spaced , which suggests that some @xmath11 orders were not detected or excited . fig .
[ schemodosg ] also shows that the density of theoretical periods is much higher than the observed density , even if only periods with @xmath40 3 are considered .
-mode spectrum _ in periods _ of bal09 .
period distances between adjacent modes are indicated with numbers expressed in seconds.,width=313 ] we searched for equally - spaced periods using different techniques . while we could produce some reasonable alignments in echelle diagrams ( e.g. for period spacing @xmath41 286s )
, the results are not convincing .
if mode trapping is effective , we should not expect good alignment from an echelle diagram .
a way around this problem would be to fine - tune models using the @xmath1-modes and then search the @xmath0-modes for periodicities , or even match their periods directly to the model ones .
if some @xmath0-modes could be identified , it could greatly increase our understanding of sdb interiors .
the 2005 campaign on bal09 was carried out because the object is one of the most interesting pulsating sdb stars .
it has high - amplitude oscillations , it is one of the brightest sdb pulsators ( @xmath3 = 11.8 mag ) , and it has the richest pulsation spectrum of both @xmath1- and @xmath0-modes .
the 2005 campaign led to the discovery of nearly a hundred independent frequencies ( table [ a_fi_2005 ] ) .
moreover , we showed that , like in many other sdbv stars @xcite , amplitudes of many modes in bal09 vary and the time scale of these changes is as short as several weeks .
the 2005 data also allowed the detection of all components of the quintuplet .
this was possible because of the low detection threshold of @xmath80.2mma in the @xmath42 data , which also resulted in the detection of a large number of frequencies , especially near 3.8 and 4.7mhz . even if some of them turn out to be non - axisymmetric multiplets , the frequency density is such that at least some of them must have @xmath31 2 .
the most surprising result of this campaign is the discovery of a considerable change of splittings in two multiplets ( a triplet and quintuplet ) which are believed to be rotationally split @xmath6 = 1 and 2 modes . as far as we are aware , this is the first clear detection of such a change of splitting in any pulsating star .
the presence of multiplets is usually attributed to rotation of the star , causing the @xmath43 degeneracy to be broken .
two clear multiplets are detected in bal09 : a @xmath6 = 1 triplet at 2.82mhz and a @xmath6 = 2 quintuplet at 2.85mhz .
the structure of the quintuplet is especially interesting .
its components , all of which are detected in 2005 , form a symmetric but not equidistant pattern .
specifically , the mean separation between @xmath44 = 2 and @xmath44 = 1 components ( 2.100 @xmath30 0.014@xmath5hz ) is considerably larger than the mean separation between @xmath44 = 1 components and the central peak of the quintuplet ( 1.577 @xmath30 0.007@xmath5hz ) .
the same was true in 2004 .
such a pattern can not be produced by a solid body rotation , nor even by a spherically symmetric rotation , @xmath45 .
indeed , in both these cases splitting within a multiplet is , in the limit of slow rotation , proportional to @xmath43 ( e.g. * ? ? ?
consequently , components of the multiplet have to be equidistant .
this is not what we observe .
thus , the observed structure of @xmath6 = 2 quintuplet implies that rotation rate of bal09 must depend on the stellar latitude , @xmath46 .
additional information can be inferred from comparison of rotational splitting for @xmath44 = 2 and for @xmath44 = 1 .
because of its spacial structure , modes of @xmath44 = 2 are most sensitive to equatorial rotation of the star , while modes of @xmath44 = 1 sense mostly rotation in the intermediate latitudes . the observed splitting is larger for @xmath44 = 2 , which implies that rotation of bal09 is fastest on the equator .
this has to be true at least in a large part of the star s envelope .
qualitatively , this dependence on @xmath46 is similar to that inferred for the sun ( e.g. * ? ? ?
the most puzzling result of 2005 campaign is the detection of changing rotational splittings .
both in the triplet and in the quintuplet the separation of components is much wider in 2005 than in 2004 .
the differences range from 12% for @xmath6 = 2 @xmath44 = 1 to 16% for @xmath6 = 2 @xmath44 = 2 .
similar behaviour of both multiplets puts strong constraint on any theoretical explanation of the observed effect .
for example , it excludes internal 1:1 resonances within the multiplets @xcite .
such a coupling could modulate the separation of multiplet components , but it would act in each multiplet independently , giving no explanation why they all change in the same way .
simultaneous widening of both multiplets must be caused by a common mechanism , which affects the structure of the star .
taken at the face value , widening of multiplets in bal09 would imply a considerable change of its internal rotation , occurring in just one year . at first glance
this seems to be a very far fetched hypothesis .
however , variations of internal stellar rotation on time scale of a few years are not unheard of .
such variations are actually observed in the sun @xcite .
these so - called torsional oscillations cause periodic changes of the solar rotation pattern with the 11-year solar cycle .
the cause of torsional oscillations in the sun is not clear , although several hypotheses have been put forward @xcite .
we must note that the internal structure of bal09 and the sun is very different .
in contrast to the sun , the envelope of an sdb star is almost entirely radiative , with the exception of an extremely thin convective layer associated with the heii ionisation zone @xcite .
however , the presence of convection is not necessary to drive torsional oscillations in the star , as they can be excited by interaction between differential rotation and magnetic field @xcite . in this mechanism ,
differential rotation and the toroidal component of the field exchange energy , causing periodic redistribution of angular momentum inside the star .
this in turn , leads to periodic changes of rotational splittings .
according to @xcite , strong differential rotation is what we expect in envelope of every sdb star .
the magnetic field in bal09 has not yet been measured , but observations of several other sdb stars reveal fields of @xmath81.5 kg @xcite .
the amplitude of torsional oscillations discovered in the sun is rather small , of the order of @xmath3011.5% of the local rotation velocity .
this is amount too small to account for changes in multiplet splittings as large as those observed in bal09 .
torsional oscillations in bal09 have to be at least an order of magnitude stronger than in the sun . whether torsional oscillations in bal09 can be driven to such amplitudes
remains an open question until detailed model calculations become available .
these results indicate that bal09 may be a key object in our understanding of pulsations in sdbv stars and its variability is definitely worth further study .
first , the changes of splittings should be monitored on a longer time scale as this may help to understand their origin .
the same is true for amplitude changes .
next , as the amplitudes change so distinctly , it may happen that modes undetectable in one season will become visible in the other , thus allowing to a complete spectrum of excited modes .
finally , the observed frequencies should be compared to a larger grid of models using identifications of the strongest frequencies as done in @xcite .
such constrained modes should result in constraining global stellar parameters for bal09 and reveal its internal structure .
this project was partially supported by grants no .
1p03d 013 29 and 1p03d 011 30 kindly provided by the polish mnisw .
ab acknowledges help of students of university of hawaii at hilo : a.hackmann , j.berghuis , h.butler , t.shimura , m.hyogo , b.dekoning , and j.slivkoff during his observations at mauna kea .
mdr and ayz were supported by the national science foundation grant ast007480 .
any opinions , findings , and conclusions or recommendations expressed in this material are those of the author(s ) and do not necessarily reflect the views of the national science foundation .
travel grants for mdr were supplied by the american astronomical association .
99 baran a. , pigulski a. , kozie d. et al . , 2005 , mnras , 360 , 737 baran a. , oreiro r. , pigulski a. et al . , 2006 ,
balta , 15 , 227 baran a. , oreiro r. , pigulski a. et al . , 2007 , a.s.p .
ser . , 372 , 607 baran a. , pigulski a. , otoole s.j . , 2008 , mnras , 385 , 255 bixler j.v . ,
bowyer s. , laget m. , 1991 , a&a , 250 , 370 buchler j.r . , goupil m .- j . , serre t. , 1995 , a&a , 296 , 405 charpinet s. , fontaine g. , brassard p. et al . , 1997 , apj , 483 , l123 charpinet s. , fontaine g. , brassard p. et al . , 2000a , apjs , 131 , 22 charpinet s. , fontaine g. , brassard p. , dorman b. , 2000b , apjs , 131 , 223 charpinet s. , fontaine g. , brassard p. , dorman b. , 2002 , apjs , 140 , 469 downes r.a . , 1986 , apjs , 61 , 569 fontaine g. , brassard p. , charpinet s. et al . , 2003 , apj , 597 , 518 goode p.r . & dziembowski w.a . , 1991 , nature , 349 , 223 green r.f . , schmidt m. , liebert j. , 1986 , apjs , 61 , 305 green e.m . , fontaine g. , reed m.d .
et al . , 2003 ,
apj , 583 , l31 howe r. , 2008 , advspres , 41 , 846 jeffery c.s . & saio h. , 2006 , mnras , 372 , l48 jimenez r. & macdonald j. , 1996 , mnras , 283 , 721 kawaler s.d . , 1999 , a.s.p . conf .
ser . , 169 , 158 kawaler s.d . & hostler s.r . , 2005 , apj , 621 , 432 kilkenny d. , koen c. , odonoghue d. et al .
, 1997 , mnras , 285 , 640 kilkenny d. , kotze j.p . , jurua e. , browstone m. , babiker h.a . , 2006 ,
balta , 15 , 255 moya a. & garrido r. , 2008 , ap&ss , in press .
moya a. , garrido r. & dupret m.a . , 2004 ,
a&a , 414 , 108 odonoghue d. , lynas - gray a.e . , kilkenny d. , stobie r.s . ,
koen c. , 1997 , mnras , 285 , 657 oreiro r. , ulla a. , prez hernndez f. et al . , 2004 , a&a , 418 , 243 oreiro r. , prez hernndez f. , ulla a. et al . , 2005 , a&a , 438 , 257 stensen r.h . , 2000 ,
ph.d.thesis , university of troms , norway otoole s.j .
, jordan s. , friedrich s. , heber u. , 2005 , a&a , 437 , 227 reed m.d . , kawaler s.d . , zoa s. , et al . , 2004 ,
mnras , 348 , 1164 reed m.d .
, otoole s.j . ,
terndrup d.m . , et al . ,
2007a , apj , 664 , 518 reed m.d .
, terndrup d.m .
, zhou a .- y . , unterborn c.t . , an d. , eggen j.r .
, 2007b , mnras , 378 , 1049 schuh s. , huber j. , dreizler s. et al . , 2006 , a&a , 445 , 31 shibahashi h. , 2004 , iau symp . 223 , 23 silvotti r. , schuh s. , janulis r. et al . , 2007 , nature , 449 , 189 stetson , p.b . , 1987 , pasp , 99 , 191 stetson , p.b . , 1990 , pasp , 102 , 932 stobie r.s . , kilkenny d. , odonoghue d. et al . , 1997 , mnras , 287 , 848 tassoul m. , 1980 , apjs , 43 , 469 telting j.h .
, stensen r.h .
, heber u. , augusteijn t. , 2006 , balta , 15 , 235 thompson m.j . , christensen - dalsgaard j. , miesch m.s . , toomre j. , 2003 , araa , 41 , 599 | we present the results of a multisite photometric campaign on the pulsating sdb star balloon090100001 .
the star is one of the two known hybrid hot subdwarfs with both long- and short - period oscillations , theoretically attributed to @xmath0- and @xmath1-modes .
the campaign involved eight telescopes with three obtaining @xmath2 data , four @xmath3-band data , and one strmgren @xmath4 photometry .
the campaign covered 48 nights , providing a temporal resolution of 0.36@xmath5hz with a detection threshold of about 0.2mmag in @xmath3-filter data .
balloon090100001 has the richest pulsation spectrum of any known pulsating subdwarf b star and our analysis detected 114 frequencies including 97 independent and 17 combination ones .
most of the 24 @xmath0-mode frequencies are between 0.1 and 0.4mhz . of the remaining 73 ,
presumably @xmath1-modes , 72 group into four distinct regions near 2.8 , 3.8 , 4.7 and 5.5mhz .
the density of frequencies requires that some modes must have degrees @xmath6 larger than 2 .
the modes in the 2.8 mhz region have the largest amplitudes . the strongest mode ( @xmath7 )
is most likely radial while the remaining ones in this region form two nearly symmetric multiplets : a triplet and quintuplet , attributed to rotationally split @xmath6 = 1 and 2 modes , respectively .
we find clear increases of splitting in both multiplets between the 2004 and 2005 observing campaigns , amounting to @xmath815% on average .
the observed splittings imply that the rotational rate in bal09 depends on stellar latitude and is the fastest on the equator .
we also speculate on the possible reasons for the changes of splitting .
the only plausible explanation we find are torsional oscillations .
this hypothesis , however , need to be verified in the future by detailed modelling . in this context
, it is very important to monitor the splittings on a longer time scale as their behaviour may help to explain this interesting phenomenon .
the amplitudes of almost all terms detected both in 2004 and 2005 were found to vary .
this is evident even during one season ; for example , amplitudes of modes @xmath9 and @xmath10 were found to change by a factor of 23 within about 50 days during 2005 .
we use a small grid of models to constrain the main mode ( @xmath7 ) , which most likely represents the radial fundamental pulsation .
the groups of @xmath1-mode frequencies appear to lie in the vicinity of consecutive radial overtones , up to the third one .
despite the large number of @xmath0-mode frequencies observed , we failed to identify them , most likely because of the disruption of asymptotic behaviour by mode trapping .
the observed frequencies were not , however , fully exploited in terms of seismic analysis which should be done in the future with a larger grid of reliable evolutionary models of hot subdwarfs .
[ firstpage ] oscillations subdwarf stars : individual : balloon090100001 . |
* cascades for @xmath2 . *
this figure shows the attracting set for @xmath3 for @xmath4 .
the attracting set is created at a saddle - node bifurcation at @xmath5 ( green dot ) .
the path of unstable fixed points ( red ) exists for all @xmath6 .
the stable fixed point undergoes infinitely many period - doubling bifurcations , limiting to the value @xmath7 .
this set of period doublings is called a period - doubling cascade .
this map also has infinitely many period - doubling cascades that begin with periodic orbits of period @xmath8 .
the red curve consists of unstable regular fixed points that exist for all @xmath9 .
, scaledwidth=100.0% ] map of the double - well duffing equation : @xmath10 .
numerical studies show regions of chaos interspersed with regions without chaos , as in the off - on - off chaos theorem ( theorem [ theorem : offonoff ] ) .
[ fig : vdp ] ] map of the forced - damped pendulum equation : @xmath11 , indicating the general areas with chaotic dynamics for this map . then the attracting sets were plotted in black , hiding some periodic points .
parameter ranges with and without chaos are interspersed .
[ fig : pend ] ] in figure [ fig : mu - xx ] , as @xmath12 increases from @xmath5 towards a value @xmath13 , a family of periodic orbits undergoes an infinite sequence of period doublings with the periods of these orbits tending to @xmath14 .
this infinite process is called a cascade .
we will later define it more precisely .
it has been repeatedly observed in a large variety of scientific contexts that the presence of infinitely many period - doubling cascades is a precursor to the onset of chaos .
for example , cascades occur in what numerically appears as the onset to chaos for both the double - well duffing equation , as shown figure [ fig : vdp ] , and the forced - damped pendulum , shown figure [ fig : pend ] .
cascades were first reported by myrberg in 1962 @xcite , and studied in more detail by feigenbaum @xcite .
this cascade is not the only cascade .
in fact , there are infinitely many distinct period - doubling cascades .
namely , there are infinitely many windows , that is , disjoint intervals in the parameter that begin with a saddle - node ( or source - sink ) bifurcation , and continue with the attractor undergoing an infinite sequence of period doublings within that interval of parameters .
quadratic maps as in the example in figure [ fig : mu - xx ] has the quite atypical property that as the parameter increases , there are no bifurcations that destroy periodic orbits .
such maps are called * monotonic*. ( this monotonicity was originally proved implicitly by douady and hubbard in the complex analytic setting .
see the milnor - thurston paper @xcite for a proof . )
once one knows that a map is monotonic , it is easy to show that as chaos develops there must be infinitely many cascades .
see figures [ fig : mu - xx ] , and [ fig : quadratic][fig : perdoubcascade ] .
the monotonicity property is a quite severe restriction , even in one - dimension .
no higher dimensional maps that develop chaos are monotonic ; yet numerical studies indicate that there are infinitely many cascades whenever there is one . see for example figures [ fig : vdp ] , [ fig : pend ] , and [ fig : henon1 ] . in this paper we summarize our progress and give new extensions to our theory that explains why there are infinitely many cascades in the onset to chaos .
our explanation is also valid for maps of arbitrary dimension . in the first result of this paper
, we consider the context in which virtually all periodic saddles have the same unstable dimension .
( by * virtually * we mean all except for a finite number . ) in this case , the onset of chaotic behavior always includes infinitely many cascades .
there is an extensive literature on routes to chaos ; that is , situations in which for some @xmath15 and @xmath0 , the trajectory of @xmath16 under @xmath1 is in the basin of a chaotic attractor , whereas under @xmath17 it is not .
whatever might have happened that caused this change between @xmath15 and @xmath0 is called a route to chaos .
we prefer to call these `` routes to a chaotic attractor '' to be more specific .
there are many different routes to a chaotic attractor .
see our discussion section for a partial enumeration .
for many maps there is competition between instability and stability .
for example , the appearance of an attracting periodic orbit as a parameter is varied may mask the chaotic dynamics , and when the orbit becomes unstable , a chaotic attractor is likely to appear .
hence a periodic orbit s loss of stability is one example of a route to a chaotic attractor .
this approach ignores the question we address here : how did the chaotic dynamics arise in the first place ?
* cascades for @xmath18 . *
the logistic map has infinitely many cascades of attracting periodic orbits , and all cascades start at the stable orbit of a saddle - node bifurcation .
the unstable orbits form what we call the * stems * of the cascades ( shown in color ) .
each stem continues to exist for all large @xmath12 values . by our terminology
, this means that all the cascades shown are solitary ( on any parameter interval @xmath19 $ ] , for @xmath20 and any @xmath21 ) since the stem does not connect its cascade to a second cascade .
the stems are shown here up to period six .
different colors are used for different periods . ] * paired cascades in the hnon map @xmath22 . *
the top bifurcation diagram shows a set of four period-7 cascades .
the bottom bifurcation diagram shows detail of the top part .
only one point of each of the period-7 orbits of the hnon map are shown so that it is clearer how the two pairs connect to each other .
the leftmost and rightmost cascades form a pair that is connected by a path of unstable regular periodic orbits ( shown in red ) . likewise
, the two middle cascades form a pair .
it is connected by a path of attracting period - seven orbits ( blue ) .
paired cascades are not robust to moderate changes in the map.,title="fig:",scaledwidth=90.0% ] * paired cascades in the hnon map @xmath22 . *
the top bifurcation diagram shows a set of four period-7 cascades .
the bottom bifurcation diagram shows detail of the top part .
only one point of each of the period-7 orbits of the hnon map are shown so that it is clearer how the two pairs connect to each other .
the leftmost and rightmost cascades form a pair that is connected by a path of unstable regular periodic orbits ( shown in red ) . likewise
, the two middle cascades form a pair .
it is connected by a path of attracting period - seven orbits ( blue ) .
paired cascades are not robust to moderate changes in the map.,title="fig:",scaledwidth=90.0% ] * two types of cascades .
* for maps with the monotonicity property , each cascade is * solitary * , in that it is not connected to another cascade by a path of * regular * periodic orbits .
these paths are the colored * stems * shown in figure [ fig : quadratic ] .
see section [ sec : definitions ] for full definitions of these terms .
furthermore , the chaos persists for all parameters larger than a certain value .
however , in many scientific contexts , it is common to see chaotic behavior appear and then disappear as the parameter @xmath12 increases .
thus as it increases there is both a route to chaos followed by a route away from chaos . in this situation
virtually all cascades are * paired * ; that is , two cascades are connected by a path of regular periodic orbits .
see figure [ fig : henon1 ] .
* solitary cascades are robust . * in our second set of results , we show that while paired cascades can be easily created and destroyed , solitary cascades are far more robust even in the presence of rather large perturbations of a map .
solitary cascades usually have stems with a constant period .
this * stem - period * can be thought of as the period that starts the cascade .
these ideas give quite striking results .
for example for each period @xmath23 the following two maps @xmath24 and @xmath25 have exactly the same number of solitary cascades of stem - period @xmath23 assuming the bifurcations of the second map are * generic*. namely , we call a map generic if its periodic orbit bifurcations are all generic .
see section [ sec : definitions ] for the full definitions of these terms .
we know that all the bifurcation orbits are generic for almost every smooth map , and that if a smooth map is not generic , then it has infinitely many generic maps arbitrarily close to it , but unfortunately with few exceptions we can not tell if a given map is generic .
the map @xmath26 has no paired cascades , but the @xmath27 may .
for example there is exactly one solitary cascade with stem - period @xmath28 and one with stem - period @xmath29 .
these results extend to @xmath30 where g is smooth ( ie .
infinitely continuously differentiable ) and @xmath31 and @xmath32 are uniformly bounded as would be the case if @xmath33 was smooth and periodic in each variable , again assuming the map @xmath3 has generic bifurcations
. * outline .
* the paper proceeds as follows : in section [ sec : definitions ] , we give some basic definitions , including what we mean by a cascade , the definition of chaos that we use here , and the class of generic maps with which we work .
section [ sec : firstresult ] contains a series of results relating chaos and cascades , with an explanation of the concrete relationship between periodic orbits within the chaos and the resulting cascades along the route towards this chaos . in section [ sec : secondresult ] , we discuss the fact that all cascades are either paired or solitary , and show that solitary cascades are robust under changes in the map . in section [ sec : paired ] , we show that if there is chaotic behavior interspersed with non - chaotic behavior , then virtually all cascades are paired .
it is common in scientific applications that chaos is interspersed with orderly behavior , in what we call * off - on - off chaos * ( defined formally in section [ sec : paired ] ) .
our numerical studies indicate that this occurs multiple times for both the forced - damped pendulum and double - well duffing examples .
we end with a discussion and present open questions in section [ sec : discussion ] .
* center manifold of saddle - node bifurcations and period - doubling bifurcations .
* this figure shows typical saddle - node and period - doubling ( halving ) bifurcations , along with the stability . for generic maps @xmath34 , it is sufficient to examine the @xmath35 center manifold .
we plot one - dimensional @xmath16 vertically and @xmath12 horizontally .
we use vertical arrows to show how the stability varies near a bifurcation periodic point indicated by a large dot . in each case , all the stability arrows can be reversed , thereby generating four more cases . ]
* the regular periodic orbits form a one - manifold near regular saddle - node and period - doubling bifurcation orbits . * in this schematic figure each point is an orbit and the horizontal axis is the parameter , usually @xmath12 in this paper .
( a & b ) : near a standard saddle - node bifurcation of a periodic orbit , the local invariant set consists of a curve of periodic orbits .
they are either all flip orbits or all regular periodic orbits .
therefore rpo is locally a curve .
( c & d ) : near a standard period - doubling ( or period - halving ) bifurcation of a periodic orbit , the local invariant set consists of two curves of periodic orbits , one of period @xmath23 shown as segment ab , and one of period @xmath36 shown as segment c. segment c always consists of regular periodic orbits , whereas exactly one of a and b consists of flip orbits and the other regular periodic orbits .
thus rpo is locally a curve consisting of c and either a or b , depending on which is regular . for the quadratic map @xmath37 , only ( a ) and ( d ) occur .
that is , periodic orbits are created but never destroyed as @xmath12 increases .
when @xmath16 is two - dimensional , such simplicity virtually never occurs @xcite . ]
* a depiction of a monotonic cascade . *
a cascade is a path of regular periodic orbits that has infinitely many period - doubling bifurcations with the periods going to infinity at the end of the path .
solid lines denote regular orbits , and dashed lines denote flip orbits .
this figure uses only the orbit - creation bifurcations , ( a ) and ( d ) in figure [ fig : snpd ] .
if only bifurcation types ( a ) and ( d ) are present and @xmath16 is scalar , then each saddle - node spawns an unstable branch and an attractor as @xmath12 increases . for @xmath38 , there are no attractors for @xmath39 for @xmath0 sufficiently large .
hence this branch of attractors can not continue forever and must bifurcate . only ( d ) is available , spawning a flip orbit branch and a doubled - period branch of attractors . again
this new branch of attractors can not continue forever as @xmath12 increases . in this way , an infinite set of period - doublings results before @xmath12 reaches @xmath0 .
much more complicated patterns are possible when all four types of bifurcations are allowed , including period - halving as well as period - doubling , and the path may contain many saddle - node bifurcations . ]
we investigate smooth maps @xmath40 where @xmath12 is in an interval @xmath41 , and
@xmath16 is in a smooth manifold @xmath42 of any finite dimension .
for example , for the forced damped pendulum , @xmath43 we take @xmath40 to be the time-@xmath44 map , where @xmath45 is the period of the forcing , and @xmath46 .
then the first coordinate of @xmath16 is on a circle , and the second is a real number . hence @xmath42 is a cylinder .
+ we say a point @xmath47 is a * period-@xmath23 point * if @xmath48 and @xmath23 is the smallest positive integer for which that is true .
its * orbit * , sometimes written @xmath49 $ ] , is the set @xmath50 by the * eigenvalues * of a period-@xmath23 point @xmath47 , we mean the eigenvalues of the jacobian matrix @xmath51 .
an orbit is called * hyperbolic * if none of its eigenvalues has absolute value 1 .
all other orbits are * bifurcation orbits*. figure [ fig : direction - reversing ] depicts two standard examples of bifurcation orbits and the resulting stability of nearby periodic points .
we call a periodic orbit a * flip * orbit if the orbit has an odd number of eigenvalues less than -1 , and -1 is not an eigenvalue .
( in one dimension , this condition is : derivative with respect to @xmath16 is @xmath52 . in dimension two ,
flip orbits are those with exactly one eigenvalue @xmath53 . )
all other periodic orbits are called * regular*. for example , the periodic orbits of constant period switch between flip and regular orbits at a period - doubling bifurcation orbit since an eigenvalue crosses @xmath54 .
see figure [ fig : snpd ] .
we write * rpo * for the set of regular periodic orbits .
for some @xmath55 and @xmath56 , let @xmath57 be a * path of regular periodic points * depending continuously on @xmath58 .
assume @xmath58 does not retrace any orbits .
that is , each @xmath59 is a periodic point on a regular periodic orbit , and distinct values of @xmath58 correspond to periodic points on distinct orbits .
we call a regular path @xmath59 a * cascade * if the path contains infinitely many period - doubling bifurcations , and for some period @xmath23 , the periods of the points in the path are precisely @xmath60 . as one traverses the cascade ,
the periods need not increase monotonically , but as @xmath61 , the period of @xmath59 goes to @xmath14 .
the orbits of a cascade with monotonic period increase are depicted schematically in figure [ fig : perdoubcascade ] .
write @xmath62 for the set of fixed points of @xmath63 and @xmath64 for the _ number _ of those fixed points .
we say that there is * exponential periodic orbit growth * at @xmath12 if there is a number @xmath65 for which the number of periodic orbits of period p satisfies @xmath66 for infinitely many @xmath23 .
for example , this inequality might hold for all even @xmath23 , but for odd @xmath23 there might be no periodic orbits .
this is equivalent to the statement that for some @xmath67 , we have @xmath68 * periodic orbit chaos . * chaotic behavior is a real world phenomenon , and trying to give it a single definition is like trying to define what a horse is .
definitions are imperfect
. a child s definition of a horse might be clear but would be unsatisfactory for a geneticist ( whose definition might be in terms of dna ) and neither would satisfy a breeder of horses who might give a recursive definition , `` an offspring of two horses . ''
definitions of real - life phenomena describe aspects of that phenomena .
they might agree in the great majority of cases on which animals are horses , though there may be rare atypical exceptions like clones where they might disagree . just as it is impossible currently to connect the shape and sound of a horse with its dna sequence , it is similarly impossible currently to identify in full generality positive lyapunov exponents with exponential growth of periodic orbits .
similarly `` chaos '' and `` chaotic '' should have definitions appropriate to the needs of the user . on the other hand , an experimenter might insist that to be chaotic , there must be a chaotic attractor , until he / she starts looking for chaos on basin boundaries and finds transient chaos .
that approach leaves no terms for the chaos that occurs outside an attractor , as on fractal basin boundaries .
we make no such restriction .
our results involve periodic orbits , and we make our definition accordingly .
we say a map @xmath69 has * periodic orbit ( po ) chaos * at a parameter @xmath12 if there is exponential periodic orbit growth .
this occurs whenever there is a horseshoe for some iterate of the map .
it is sufficiently general to include having one or multiple co - existing chaotic attractors , as well as the case of transient chaos .
as hinted at by equation [ eqn : limsup ] , in many cases po chaos is equivalent to positive topological entropy .
we discuss this relationship further in section [ sec : discussion ] .
the * unstable dimension * dim@xmath70 of a periodic point @xmath47 or periodic orbit is defined to be the number of its eigenvalues @xmath71 having @xmath72 , counting multiplicities .
we say there is * virtually uniform ( po ) chaos * at @xmath12 if there is po chaos , and all but a finite number of periodic orbits have the same unstable dimension , denoted dim@xmath73 . for the pendulum map discussed above ,
whenever there is po chaos at some parameter value @xmath12 , we expect the periodic orbits to be primarily saddles , and if it likewise had virtually uniform po chaos , then we would expect @xmath74 .
assuming there are infinitely many periodic orbits , roughly half would be regular saddles , with the rest being flip saddles .
furthermore , all attracting periodic points are regular .
our first goal is to describe the * route to ( po ) chaos*. that is , if at @xmath15 there is no chaos , while at @xmath0 there is virtually uniform chaos , we explain what must happen in this interval in order for chaos to arise .
we believe that generally there is one typical route to chaotic dynamics .
namely , there must be infinitely many period - doubling cascades when @xmath12 is between @xmath15 and @xmath0 .
( each of these cascades in turn has infinitely many period - doubling bifurcations . ) * a regular periodic hopf bifurcation point is locally contained in a unique curve of regular periodic orbits . * to understand generic hopf bifurcations
, it is sufficient to determine the dynamics on its ( invariant ) center manifold , which is locally like @xmath75 .
it contains all periodic orbits that are near the bifurcation orbit . near a generic hopf bifurcation with no eigenvalues which are roots of unity ,
the local invariant set consists of a curve of periodic points ( in solid black ) and a surrounding paraboloid ( at left ) .
although there are infinitely many periodic orbits on the paraboloid in each neighborhood of the hopf bifurcation point , the middle curve of periodic points is disconnected from all periodic points on the paraboloid .
specifically , the paraboloid contains infinitely many annuli converging to the bifurcation point , each with periodic orbits formed and destroyed at saddle - node bifurcations .
typical annuli are depicted here by colored regions .
the periodic orbits are depicted in blue ( saddles ) and red ( nodes ) . between every two annuli
, there is an invariant circle ( in black ) with irrational rotation number , as made clear by the projection ( at right ) of the paraboloid to the center manifold plane of the phase space .
, title="fig : " ] * a regular periodic hopf bifurcation point is locally contained in a unique curve of regular periodic orbits .
* to understand generic hopf bifurcations , it is sufficient to determine the dynamics on its ( invariant ) center manifold , which is locally like @xmath75
. it contains all periodic orbits that are near the bifurcation orbit . near a generic hopf bifurcation with no eigenvalues which are roots of unity ,
the local invariant set consists of a curve of periodic points ( in solid black ) and a surrounding paraboloid ( at left ) .
although there are infinitely many periodic orbits on the paraboloid in each neighborhood of the hopf bifurcation point , the middle curve of periodic points is disconnected from all periodic points on the paraboloid . specifically , the paraboloid contains infinitely many annuli converging to the bifurcation point , each with periodic orbits formed and destroyed at saddle - node bifurcations .
typical annuli are depicted here by colored regions .
the periodic orbits are depicted in blue ( saddles ) and red ( nodes ) . between every two annuli
, there is an invariant circle ( in black ) with irrational rotation number , as made clear by the projection ( at right ) of the paraboloid to the center manifold plane of the phase space .
, title="fig : " ] * generic maps . * our results are given for generic maps of a parameter .
specifically , we say that the map @xmath3 is * generic * if all of the bifurcation orbits are * generic * , meaning that each bifurcation orbit is one of the following three types : 1
. a standard saddle - node bifurcation .
( where `` standard '' means the form of the bifurcation stated in a standard textbook , such as robinson @xcite . )
in particular the orbit has only one eigenvalue @xmath71 for which @xmath76 , namely @xmath77 .
2 . a standard period - doubling bifurcation . in particular
the orbit has only one eigenvalue @xmath71 for which @xmath76 , namely @xmath78 .
3 . a standard hopf bifurcation . in particular
the orbit has only one complex pair of eigenvalues @xmath71 for which @xmath76 .
we require that these eigenvalues are not roots of unity ; that is , there is no integer @xmath79 for which @xmath80 .
these three bifurcations are depicted in figures [ fig : snpd ] and [ fig : hopf ] .
generic @xmath3 have at most a countable number of bifurcation orbits , so almost every @xmath12 has no bifurcation orbits .
see @xcite for the details showing that these maps are indeed generic in the class of smooth one - parameter families . for systems with symmetry such as the forced - damped pendulum ,
a fourth type of bifurcation occurs , such as a pitchfork or symmetry - breaking bifurcation .
this adds complications , though in fact with extra work our results remain true .
our motivation for considering generic maps is given in proposition [ prop : components ] in the next section , which states that each regular periodic orbit for a generic map is locally contained in a unique path of periodic orbits .
the connection to cascades can be summarized as follows : starting at each regular periodic orbit @xmath26 for @xmath81 $ ] , there is a local path of regular periodic orbits through @xmath26 . enlarge this path as far as possible .
either the path reaches @xmath15 or @xmath0 , or there is a cascade .
this idea is explained in more detail in the next section .
our first result is theorem [ theorem : version1 ] , which demonstrates that the route to virtually uniform po chaos contains infinitely many period - doubling cascades .
theorem [ theorem : version2 ] is a restatement of these results in a way that makes the relationship between chaos and cascades much more transparent .
write @xmath41 for the closed parameter interval @xmath82 $ ] .
our main hypotheses will be used for a variety of results so we state them here . *
list of assumptions . * 1 .
assume @xmath3 is a generic smooth map ; that is , @xmath3 is infinitely differentiable in @xmath12 and @xmath16 , and all of its bifurcation orbits are generic .
assume there is a bounded set @xmath83 that contains all periodic points @xmath84 for @xmath85 .
3 . assume all periodic orbits at @xmath15 and @xmath0 are hyperbolic .
4 . assume that the number @xmath86 of periodic orbits at @xmath15 is finite .
assume at @xmath0 there is virtually uniform po chaos .
write @xmath87 for the number of periodic orbits at @xmath0 having unstable dimension not equal to dim@xmath88 .
[ theorem : version1 ] assume ( @xmath89 ) .
then there are infinitely many distinct period - doubling cascades between @xmath15 and @xmath0 . based on numerical studies ,
a number of maps appear to satisfy the conditions of the above theorem .
note that numerical verification involves significantly more work than just plotting the attracting sets for each parameter , since we are concerned about both the stable and the unstable behavior to determine whether there is chaos .
examples include the time-@xmath90 maps for the double - well duffing ( fig .
[ fig : vdp ] ) , the triple - well duffing , the forced - damped pendulum ( fig . [ fig : pend ] ) , the ikeda map ( introduced to describe the field of a laser cavity ) , and the pulsed damped rotor map .
this formulation gives no idea how the behavior at @xmath0 is connected to the cascades that must exist .
thus before giving a proof , we reformulate the conclusions of the theorem in a more transparent way . specifically , we reformulate so that it is clear how infinitely many cascades in the strip @xmath91 \times \mathfrak m$ ] are connected to regular periodic orbits at @xmath0 by continuous paths in rpo . * paths of orbits .
* we now give a formal definition for paths of regular periodic orbits and show that they connect regular periodic orbits at @xmath0 with cascades between @xmath15 and @xmath0 .
we start with the following theorem , which guarantees that a path through each regular periodic orbit is unique .
assume hypotheses @xmath92 , and consider a local path through a regular periodic orbit as guaranteed by the above proposition .
specifically , let @xmath93 be a regular periodic point .
since all such points at @xmath0 in the above theorem are assumed to be hyperbolic , the orbit can be followed without a change of direction for nearby @xmath94 .
we can define @xmath95 continuously for @xmath58 in an interval @xmath96 $ ] as follows .
let @xmath97 .
then @xmath59 follows the branch of regular periodic points for decreasing @xmath12 until reaching @xmath98 , which is either a bifurcation orbit or @xmath99 .
if it is a bifurcation point , which would be a generic bifurcation point , then from the above proposition there is a unique branch of regular hyperbolic periodic points leading away from @xmath98 , and @xmath59 follows that branch
. if @xmath98 is a period - doubling bifurcation point , there are two branches of regular periodic points , but both are on the same orbits .
either branch can be followed since both represent the same orbits and both are regular
. each branch of hyperbolic periodic orbits can be parametrized by @xmath12 and is easy to find numerically by solving a differential equation .
write @xmath100 for the square matrix @xmath101 and @xmath102 for the vector @xmath103 . differentiating @xmath104 with respect to @xmath12 and manipulating yields @xmath105 @xmath106 where @xmath107 denotes the identity matrix .
the matrix inverse above exists since @xmath108 , being hyperbolic , can not have @xmath109 as an eigenvalue .
this construction suggests a definition .
we say @xmath59 is a * path * in @xmath110 for @xmath58 in some interval @xmath111 if it is continuous and * @xmath59 is a regular periodic point in @xmath110 for each @xmath112 ; * y does not retrace orbits ; that is , @xmath113 is never on the same orbit for different @xmath58 .
[ prop : components ] for a smooth generic map @xmath114 , each regular periodic point is locally contained in a path of regular periodic points .
the corresponding path of periodic orbits is unique .
this proposition appears in reference @xcite .
here we give an idea of the proof .
the implicit function theorem shows that there is a local curve of periodic points through a regular hyperbolic periodic point .
therefore , the proof consists of showing that there is also a local curve of orbits through each periodic bifurcation orbit .
this must be shown for each of the three types of generic bifurcations .
the curve is not necessarily monotonic : it may reverse direction with respect to @xmath12 , as depicted in figure [ fig : direction - reversing ] .
namely , a path @xmath57 * reverses direction * at @xmath115 if @xmath116 changes from increasing to decreasing or vice versa at that point . for n - dimensional
@xmath16 , the study of generic saddle - node and period - doubling bifurcations can be reduced to one - dimensional @xmath16 due to the center manifold theorem .
the bifurcation orbit has @xmath117 eigenvalues outside the unit circle , and these do not affect the dynamics on the two - dimensional invariant @xmath118-plane .
hence here we can consider @xmath16 as being one - dimensional . in the case of a saddle - node bifurcation ,
one branch of orbits is stable and one is unstable in @xmath119 .
since the other @xmath117 eigenvalues do not cross the unit circle near the point , it follows that the unstable dimension of the orbits in @xmath120dimensions is even for one branch and odd for the other , i.e. , the unstable dimension @xmath121 of @xmath59 changes parity as @xmath59 passes through the bifurcation point @xmath122 . for a period - doubling bifurcation ,
consider the notation depicted in figure [ fig : direction - reversing ] .
namely , let ( a ) denote the period-@xmath123 orbits with no coexisting period-@xmath124 orbits , let ( b ) denote the branch of period-@xmath123 orbits coexisting at the same parameter values with the branch ( c ) of period-@xmath124 orbits .
if branch ( a ) is regular and ( b ) is flip , then the path @xmath113 includes branches ( a ) and ( c ) , which are on different sides of the bifurcation point , and @xmath116 does not change direction and @xmath121 does not change parity . if however ( b ) is regular and ( a ) is flip , then @xmath116 does change direction and @xmath121 does change parity . for hopf bifurcations ( see figure [ fig : hopf ] )
, the path proceeds monotonically past the bifurcation point and the unstable dimension changes by @xmath125 , so the parity does not change .
hence in all cases , the path changes direction if and only if the parity changes .
the cases of saddle - node and period - doubling bifurcations are straightforward , as depicted in figure [ fig : snpd ] . to show that rpo is locally a curve at a hopf bifurcation is trickier than the other two cases , since there are in fact infinitely - many periodic orbits near a hopf bifurcation point , but they are not connected to the hopf bifurcation point by any path of periodic points , as shown in figure [ fig : hopf ] .
the proof follows from the steps listed .
* cascades .
* we call a regular path @xmath59 for @xmath126 a * cascade * if the path contains infinitely many period - doubling bifurcations , and for some period p , the periods of the points in the path are precisely @xmath127 . as one traverses the cascade ,
the periods need not increase monotonically , but as @xmath61 , the period of @xmath59 goes to @xmath14 .
we will say a path @xmath59 is * maximal * if the following additional condition holds : * @xmath113 can not be extended further to a larger interval , and it can not be redefined to include points of more regular orbits . figure [ fig : maximal - path ] shows an example of a middle portion of a maximal path
. figure [ fig : pathsfromchaos1 ] and [ fig : pathsfromchaos2 ] show different possibilities for how the maximal path ends .
* paths of regular periodic orbits .
* we assume here that all bifurcation orbits are generic .
it follows that a regular periodic point is in a path of periodic points .
since each periodic point is in a periodic orbit , we can think of the path as being a path of regular periodic orbits .
this is a unique path of regular orbits ( solid ) .
here we show orbits , not points .
flip orbits are depicted by dashed lines , showing that the path is no longer unique if we include all periodic orbits .
let @xmath59 be a path of regular ( non - flip ) periodic points parametrized by @xmath58 , such that @xmath59 does not pass through the same orbit twice .
write @xmath128 $ ] for the orbit that the point @xmath59 is in .
it is a useful fact that whenever the @xmath12 coordinate @xmath116 of @xmath59 changes direction , the unstable dimension of @xmath59 changes by one .
hence it changes from odd to even or vice versa .
the arrows along the path show which way the path is traveling as @xmath58 increases . ]
* paths of regular periodic orbits starting from @xmath0 . *
assume the bifurcations are generic .
we use the notation of the above figure .
let @xmath129 be a regular hyperbolic periodic point .
if its maximal path of orbits extends to a hyperbolic orbit @xmath130 at @xmath15 , then the path has changed directions an even number of times and the unstable dimensions of @xmath131 and @xmath130 have the same parity ; that is , both are even or both are odd .
if the path starting at a regular hyperbolic orbit @xmath132 returns to a point @xmath133 at @xmath0 , the path changes directions an odd number of times , and so the unstable dimensions of @xmath134 and @xmath133 have opposite parity and in particular the unstable dimensions of the two are different . ] * an interior path of regular orbits starting from @xmath0 yields cascades . *
continuing the assumptions and notations of the above figures , assume the hypotheses of theorem [ theorem : version1 ] . only a finite number of paths that start at @xmath0 can either reach @xmath15 or return to @xmath0 .
hence there are an infinite number of regular hyperbolic orbits at @xmath0 that yield paths that neither reach @xmath15 nor return to @xmath0 .
such * interior paths * must contain an infinite number of bifurcations .
since generically there are only a finite number of bifurcation orbits of each period between @xmath15 and @xmath0 , the period of the orbits must tend to infinity along the path and it must contain a cascade .
hence there are infinitely many cascades between @xmath15 and @xmath0 . ]
let orbits(y ) be the set of periodic points on orbits traversed by @xmath113 .
that is , if @xmath135 for some @xmath58 , then @xmath136 and so are the other points on the orbit , @xmath137 for all n. two integers are said to have the same * parity * if both are odd or both are even .
otherwise they have opposite parity .
figures [ fig : direction - reversing ] and [ fig : maximal - path ] demonstrate why this is a critical idea .
namely , the unstable dimension of a path @xmath59 * changes parity * as @xmath58 increases precisely when the path changes directions .
thus the parity of the unstable dimension corresponds to the orientation of the path .
[ theorem : version2 ] assume ( @xmath89 )
. then the following are true .
* there are infinitely many regular periodic points at @xmath0 .
* for each maximal path @xmath138 in @xmath110 starting from a regular periodic point @xmath139 , the set of orbits traversed , denoted by orbits@xmath140 , depends only on the initial orbit containing @xmath131 .
that is , different initial points on the same orbit yield paths that traverse the same set of orbits , so we can write orbits@xmath141 for orbits@xmath140 .
* let @xmath139 and @xmath142 be regular periodic points on different orbits . then orbits@xmath141 and orbits@xmath143 are disjoint . *
let @xmath111 denote the unstable dimension of a regular periodic point @xmath144 .
for a maximal path @xmath138 in @xmath110 starting from @xmath144 , let @xmath121 denote the unstable dimension of @xmath59 . at each direction - reversing bifurcation , @xmath121 changes parity ; that is it changes from odd to even or vice versa .
initially @xmath145 so initially @xmath116 is decreasing and @xmath146 , so initially @xmath147 is even .
hence in general @xmath116 is decreasing if @xmath148 is even and increasing if it is odd .
* let @xmath113 be a maximal path on @xmath149 $ ] in @xmath110 and @xmath150 .
if @xmath151 , then @xmath116 is increasing at that point so @xmath152 is even .
hence @xmath152 is odd , so @xmath153 .
* there are infinitely many distinct period - doubling cascades between @xmath15 and @xmath0 .
* there are at most @xmath154 regular periodic orbits at @xmath0 with unstable dimension dim@xmath88 that are not connected to cascades .
theorem [ theorem : version2 ] is a version of theorem [ theorem : version1 ] that allows us to show that most regular orbits at @xmath0 are connected to cascade by a path of regular orbits .
the conclusions of this theorem were shown to hold in @xcite under assumptions @xmath92 along with the additional hypothesis that at @xmath0 , there are infinitely many periodic regular orbits . however , we are now able to show that this last hypothesis is unnecessary because it is automatically true .
in fact , under assumptions @xmath92 , approximately half of the periodic orbits are regular .
this proof is quite technical in that it involves topological fixed point index theory . to make this treatment more readable
, we give the full details in @xcite .
assume phase space is two - dimensional .
if at @xmath0 there is a transverse homoclinic point then there will be infinitely many saddles and exponential orbit growth .
condition @xmath155 is often satisfied if the attractor(s ) are periodic orbits and there is a smale horseshoe in the dynamics . our numerical studies of the forced - damped pendulum indicate that it satisfies the hypotheses of this theorem on @xmath19 \approx [ 1.8,20 ] , [ 73,20 ] , [ 73,175],$ ] and @xmath156 $ ] .
thus there are infinitely - many cascades on each of these intervals .
* weakening the uniformity hypothesis .
* we end this section with the conjecture that @xmath157 can be weakened without effecting the conclusions of the theorem .
specifically , let @xmath158 be the fraction of all period - p orbits at @xmath12 that have unstable dimension @xmath111 .
we will say that * most periodic orbits have unstable dimension @xmath111 * if @xmath159 as @xmath160 .
we believe that for most generic smooth maps and most @xmath12 , there is some dimension @xmath161 for which this holds .
assume ( @xmath162 ) ( not including ( @xmath157 ) ) .
assume that most periodic orbits at @xmath0 have unstable dimension @xmath111 .
then there are infinitely many regular periodic orbits with unstable dimension @xmath111 , and infinitely many of these are connected to distinct period - doubling cascades between @xmath15 and @xmath0 .
let @xmath41 be an interval , and let @xmath59 for @xmath163 be a maximal path of regular periodic orbits containing a cascade .
for any point @xmath164 in the interior of @xmath41 , infinitely many period doublings of the cascade occur either for @xmath165 or @xmath166 . in the following definition , we distinguish two types of cascades based on what happens to the other end of @xmath59 .
let @xmath3 satisfy @xmath167 .
a cascade is * solitary * in @xmath19 $ ] if the maximal path of regular periodic orbits containing it contains no other cascades . in this case
, we call the non - cascade end of the maximal path the * stem * of the cascade .
a cascade is * paired * if its maximal path contains two cascades .
a cascade is * bounded * in @xmath19 $ ] if its maximal path is contained in the interior of the interval and never hits the boundary @xmath168 or @xmath169 .
we call a path which stays in the interior of the interval an * interior path*. on a sufficiently large parameter interval , all cascades for quadratic maps are solitary , as shown in figure [ fig : quadratic ] .
figure [ fig : henon1 ] shows an example of two sets of paired cascades .
the following theorem shows that the classification of solitary versus paired is equivalent to classifying cascades in an interval as being purely interior versus hitting the interval boundary .
[ theorem : solitarypaired ] let @xmath3 satisfy @xmath167 on @xmath19 $ ] .
a cascade is paired if and only if the maximal path containing the cascade is an interior path . in particular , * bounded cascades are always paired*. the maximal path of a solitary cascade contains a periodic point on the interval boundary ; i.e. there is a point @xmath118 in the maximal path of the cascade such that @xmath168 or @xmath169 .
namely , * solitary cascades are never bounded*. we give a sketch of the proof .
let @xmath41 be an interval , and let @xmath59 for @xmath163 be a maximal path of regular periodic points in @xmath19 $ ] containing a cascade as @xmath58 increases .
fix @xmath170 , and consider the one - sided maximal paths @xmath171 where @xmath172 respectively for @xmath173 , @xmath163 .
we have assumed that @xmath174 contains the cascade .
therefore @xmath175 is an interior path , since otherwise there would not be infinitely many period - doubling bifurcations .
key fact : by the methods described in section [ sec : firstresult ] , a one - sided maximal path of regular periodic points is an interior path if and only if it contains a cascade .
assume the cascade in @xmath59 is paired .
then @xmath176 contains a cascade , so by the key fact above , @xmath177 is an interior path , implying @xmath59 is as well .
conversely , if @xmath59 is an interior path , then @xmath177 is as well , and by the key fact @xmath177 contains a cascade , meaning that the cascades in @xmath178 are paired .
* conservation of solitary cascades . *
if @xmath3 satisfies @xmath92 , and @xmath179 , then theorem [ theorem : version1 ] implies that there is * conservation * of solitary cascades .
specifically , let @xmath180 be any generic map such that @xmath3 and @xmath180 agree at @xmath15 and @xmath0 , though they may have completely different behavior inside the interval .
then @xmath3 and @xmath180 have exactly the same solitary cascade structure .
since the number of cascades is infinite , this does not appear to be a meaningful statement .
however , we can classify solitary cascades by looking at the period of their stems , and it is in this sense that their cascades are the same . in figure
[ fig : quadratic ] we label five solitary cascades , one of period three , one of period four , two of period five , and one of period six . * quadratic , cubic , and quartic maps . * the three maps from eqn .
[ eqn : quad+g ] , are shown for @xmath181 and for @xmath182 and @xmath183 , respectively .
the squares depicted are @xmath184 ^ 2 $ ] .
the maxima and minima are far larger than the size of the boxes , resulting in purely uniform chaotic behavior within the boxes .
this is typical for large @xmath12 .
as @xmath12 increases , these three graphs would be stretched vertically .
the values of the map @xmath3 for large @xmath12 at the critical points are proportional to @xmath12 , @xmath185 , and @xmath186 , respectively , largely unaffected by @xmath33 , which is bounded and has bounded derivative .
, title="fig : " ] * quadratic , cubic , and quartic maps . * the three maps from eqn .
[ eqn : quad+g ] , are shown for @xmath181 and for @xmath182 and @xmath183 , respectively .
the squares depicted are @xmath184 ^ 2 $ ] .
the maxima and minima are far larger than the size of the boxes , resulting in purely uniform chaotic behavior within the boxes .
this is typical for large @xmath12 .
as @xmath12 increases , these three graphs would be stretched vertically .
the values of the map @xmath3 for large @xmath12 at the critical points are proportional to @xmath12 , @xmath185 , and @xmath186 , respectively , largely unaffected by @xmath33 , which is bounded and has bounded derivative .
, title="fig : " ] * quadratic , cubic , and quartic maps . * the three maps from eqn .
[ eqn : quad+g ] , are shown for @xmath181 and for @xmath182 and @xmath183 , respectively .
the squares depicted are @xmath184 ^ 2 $ ] .
the maxima and minima are far larger than the size of the boxes , resulting in purely uniform chaotic behavior within the boxes .
this is typical for large @xmath12 .
as @xmath12 increases , these three graphs would be stretched vertically .
the values of the map @xmath3 for large @xmath12 at the critical points are proportional to @xmath12 , @xmath185 , and @xmath186 , respectively , largely unaffected by @xmath33 , which is bounded and has bounded derivative . ,
title="fig : " ] three rigorous examples of conservation of solitary cascades are encapsulated in the following one - dimensional maps ( see figure [ fig : threemaps ] ) : @xmath187 where @xmath33 is smooth , and for some real positive @xmath188 , @xmath189 it is straightforward to show that for @xmath190 , each of these maps has a @xmath191 $ ] such that there are no regular periodic orbits for @xmath3 at @xmath192 , and @xmath3 has virtually uniform ( po ) chaos at @xmath193 .
this leads to the fact that for @xmath33 of the form given , each of the three maps has no regular periodic orbits for @xmath15 sufficiently negative , and for @xmath0 sufficiently large has a one - dimensional horseshoe map .
the conditions on @xmath33 guarantee that it does not significantly affect the periodic orbits when @xmath194 is sufficiently large ; in particular it does not affect their eigenvalues , so it does not affect the number of period-@xmath23 regular periodic orbits for large @xmath194 . hence all three maps have no regular periodic orbits for @xmath12 small and have no attracting periodic orbits for @xmath12 large , and for sufficiently large @xmath194 , all periodic orbits are contained in the set @xmath195 $ ] .
this leads to the following result : [ theorem : conservation ] for each of the functions @xmath3 in eqn .
[ eqn : quad+g ] , @xmath3 is generic for a residual set of @xmath33 chosen as in eqn .
[ eqn : g1 ] . for each of the three examples ,
the number of stem - period-@xmath196 solitary cascades for all generic @xmath3 is independent of the choice of @xmath33 . in other words ,
fix @xmath3 to be one of the three types in eqn .
[ eqn : quad+g ] .
fix any @xmath33 as in eqn .
[ eqn : g1 ] such that @xmath3 is generic
. then there exist @xmath197 and @xmath198 such that as long as @xmath199 and @xmath200 , @xmath3 has the same number of solitary cascades on @xmath19 $ ] as occur in the @xmath190 case for the interval @xmath191 $ ] . as an interpretation of this statement
take a set @xmath201 as large as you like in parameter cross phase space , and make @xmath202 in @xmath201 , so @xmath203 in @xmath201 .
hence the only periodic orbit in @xmath201 is the fixed point @xmath204 , so @xmath201 contains no cascades .
it might seem that we have annihilated all cascades by this process , but the theorem guarantees that every single one of these solitary cascades will appear .
they are just displaced from their original location , moved outside of @xmath201 .
the conservation principle works for higher - dimensional maps as well .
for example , we have shown in @xcite that even large - scale perturbations of the two - dimensional hnon map have conservation of solitary cascades .
that is , writing @xmath205 , @xmath206 where @xmath188 is fixed , and the added function @xmath207 is smooth and is very small for @xmath208 sufficiently large .
see @xcite for a precise ( and technical ) formulation . here
the added terms can not destroy the stems which are unchanged in the domain where @xmath209 is very large .
each stem must still lead to its own solitary cascade . if @xmath210 is a set of @xmath211 coupled quadratic maps such that @xmath212 where @xmath33 is bounded with bounded first derivatives , and @xmath213 , then there is conservation of solitary stem - period-@xmath196 cascades .
in the previous section , we concentrated on the implications of theorem [ theorem : solitarypaired ] on solitary cascades , but it also has implications for paired cascades .
it describes a situation which is quite common in physical systems , namely the case in which parameter regions with and without chaos are interspersed , such as show in figs .
[ fig : vdp ] and [ fig : pend ] . specifically , we give the following definition describing the situation where is no chaos at @xmath15 and @xmath214 , while at @xmath3 is chaotic at @xmath0 .
assume @xmath3 satisfies @xmath215@xmath157 on @xmath19 $ ] and on @xmath216 $ ] , where @xmath217 .
then @xmath3 is said to have * off - on - off chaos * for @xmath217 .
if @xmath3 has off - on - off chaos , we can apply theorem [ theorem : version2 ] to both @xmath19 $ ] and to @xmath218 $ ] and conclude that there are infinitely many cascades in each of the two intervals .
the following theorem is stronger , in that we also conclude that virtually all regular periodic points at @xmath0 are contained in paired cascades .
[ theorem : offonoff ] if @xmath3 has off - on - off chaos on @xmath217 , then @xmath3 has infinitely many ( bounded ) paired cascades and at most finitely many solitary cascades in @xmath219 .
the result follows directly from combining the results of theorems [ theorem : version2 ] and [ theorem : solitarypaired ] .
our numerical studies indicate that the time-@xmath90 maps of the forced double - well duffing ( fig .
[ fig : vdp ] ) and forced damped pendulum ( fig .
[ fig : pend ] ) have off - on - off chaos , and that it occurs on multiple non - overlapping parameter regions . here
@xmath90 is the forcing period .
we can not prove there are only finitely many periodic attractors , though we find very few .
these systems have no periodic repellers .
* our results in the context of routes to chaos .
* we now contrast our view with the traditional view of many different routes to chaos .
people write about routes to chaos - where chaos does indeed mean there is a chaotic attractor .
below we list classes of distinct routes to chaotic attractors .
our results say that for generic smooth maps depending on a parameter , there is a unique route from no chaos to chaos ( which we have defined to mean virtually uniform po chaos ) in the sense of having a chaotic set that need not be attracting , for example when there is a transverse homoclinic point . between a parameter value where there is no chaos and a parameter value where there is chaos , there must be infinitely many period - doubling cascades .
* routes to a chaotic attractor * 1 .
a chaotic attractor develops where there was no previous horseshoe dynamics ; this includes what we might call the feigenbaum cascade route .
2 . there is a transient chaos set and a simple non - chaotic attractor ( equilibrium point or periodic orbit ) and that attractor becomes unstable . for periodic orbits
there are three ways to become unstable .
3 . like above except
the attractor is a torus with quasiperiodic dynamics .
how many ways can a torus become unstable ?
we suspect in many ways .
crisis route : as a parameter decreases , a chaotic attractor collides with its basin boundary and the chaotic set is no longer attracting .
how many ways can this happen ? also unknown .
there is a chaotic attractor and a non - chaotic attractor and as a parameter is varied , the initial condition being used migrates into the basin of the chaotic attractor .
homoclinic explosions lead to chaotic dynamics .
however , using the viewpoint described in this paper , * there is only one route to chaos*. * relating po chaos and positive entropy . *
consider the following definition , related to exponential growth of periodic orbits , alluded to when the definition was given : if @xmath220 for large @xmath23 , then we say @xmath221 is the growth factor , or we call @xmath222 the * periodic orbit entropy*. taking logs of both sides , dividing by @xmath23 , and taking limits , we get @xmath223 we believe that many of the methods used here will lead to a fruitful study of periodic orbit entropy . note that po chaos occurs exactly when @xmath224 .
if there are only finitely many orbits for example , @xmath225 . if there is at most a fixed number @xmath196 of periodic orbits of each period @xmath23 , again @xmath226 , even if there may be infinitely many orbits . in many situations ,
there is a relationship between exponential periodic growth and positive topological entropy , and in fact in some cases periodic orbit entropy is equal to topological entropy .
we elaborate below .
bowen showed @xcite that for axiom a diffeomorphisms , you can find topological entropy exactly by looking at the growth rate for the number of period-@xmath23 orbits as @xmath23 goes to infinity .
for hnon - like maps , wang and young proved that the topological entropy coincides with the exponential growth rate of the number of periodic points of period @xmath23 @xcite .
chung and hirayama @xcite show that for any surface diffeomorphism with hlder continuous derivative , the topological entropy is equal to the exponential growth rate of the number of hyperbolic periodic points of saddle type .
that is , you can throw away the attractors and repellers . in terms of interval maps : misiurewicz and szlenk @xcite proved that if @xmath227 is a continuous and piecewise monotone map of the interval , then the topological entropy is bounded above by the exponential growth rate of the number of periodic orbits . building on @xcite , chung @xcite showed that if @xmath227 is @xmath228 , similar results can be obtained if one counts only the set of hyperbolic periodic orbits ( namely , periodic points @xmath16 for which @xmath229 ) or the set of transversal homoclinic points of a source .
* non - smooth maps .
* continuous maps @xmath3 which are piecewise smooth but not smooth violate our assumptions , but such maps can be thought of as the limits of generic smooth maps @xmath230 which differ from @xmath3 only very near the discontinuities of @xmath231 . in dimension one
, @xmath232 gives a rich collection of examples obtained by choosing constants @xmath233 and @xmath234 in interesting ways .
see the extensive literature on border collision bifurcations @xcite .
such maps deserve more discussion than we can give here but we mention two examples .
\(1 ) for the tent map @xmath235 , there are no periodic orbits when @xmath236 ; there is one periodic orbit , a fixed point at 0 , when @xmath237 , and orbits of all periods when @xmath238 .
all orbits are in families of straight line rays that bifurcate from and originate at @xmath239 , the map s only bifurcation periodic orbit . in this case , all bifurcations in all the cascades that would exist for the approximating generic families @xmath230 tend to @xmath239 as @xmath240 .
\(2 ) for the tent map with slope @xmath241 , namely latexmath:[$f(\mu , x ) = 1 -\mu at @xmath204 for countably many @xmath12 . for example
if @xmath214 denotes the smallest parameter with a period - three orbit , then the point @xmath204 is a period - three point , and only infinitely many periodic orbits whose periods are multiples of three bifurcate from it .
all of the cascades for those orbits that we would expect for a smooth map are collapsed into the single point @xmath243 .
we thank safa motesharrei for his corrections and detailed comments . e.s .
was partially supported by nsf grants dms-0639300 and dms-0907818 , as well as nih grant r01-mh79502 .
was partially supported by nsf grant dms-0616585 and nih grant r01-hg0294501 . | the appearance of infinitely - many period - doubling cascades is one of the most prominent features observed in the study of maps depending on a parameter .
they are associated with chaotic behavior , since bifurcation diagrams of a map with a parameter often reveal a complicated intermingling of period - doubling cascades and chaos .
period doubling can be studied at three levels of complexity .
the first is an individual period - doubling bifurcation .
the second is an infinite collection of period doublings that are connected together by periodic orbits in a pattern called a cascade .
it was first described by myrberg and later in more detail by feigenbaum .
the third involves infinitely many cascades and a parameter value @xmath0 of the map at which there is chaos .
we show that often virtually all ( i.e. , all but finitely many ) `` regular '' periodic orbits at @xmath0 are each connected to exactly one cascade by a path of regular periodic orbits ; and virtually all cascades are either paired connected to exactly one other cascade , or solitary connected to exactly one regular periodic orbit at @xmath0 .
the solitary cascades are robust to large perturbations .
hence the investigation of infinitely many cascades is essentially reduced to studying the regular periodic orbits of @xmath1 .
examples discussed include the forced - damped pendulum and the double - well duffing equation . |
the classic 1935 paper of erds and szekeres @xcite published in _
compositio mathematica _ was a starting point of a very rich discipline within combinatorics : ramsey theory ( see , e.g. , @xcite ) .
erds liked to call the main result of this paper the happy ending theorem , " as its discovery was triggered by a geometric observation of esther klein , and the authors collaboration with her eventually led to the marriage of klein and szekeres . at the early stages of its development , ramsey theory focused on the emergence of large monochromatic _ complete _ subgraphs in colored graphs and hypergraphs . in the 1960s and 70s ,
many initial results of ramsey theory were revisited in a more general setting .
for instance , gerencsr and gyrfs @xcite proved that for any 2-coloring of the edges of a complete graph with roughly @xmath19 vertices , there is a monochromatic path of length @xmath4 .
recently , figaj and uczak @xcite and , in a more precise form , gyrfs et al .
@xcite have settled an old conjecture of faudree and schelp , according to which any 3-colored complete graph with roughly @xmath20 vertices has a monochromatic path of length @xmath4 .
it follows from a theorem of erds and gallai @xcite that for every positive integer @xmath9 there exists a constant @xmath21 such that no matter how we color the edges of the complete graph @xmath22 with @xmath9 colors , we can always find a monochromatic path of length @xmath4 .
the smallest value of the constant @xmath23 is not known .
the problem becomes simpler if we consider _ ordered _ complete graphs and we want to find a long _ increasing _ path such that all of its edges are of the same color .
it follows from the results of erds and szekeres @xcite , and also from dilworth s theorem on partially ordered sets @xcite , that for any 2-coloring of the edges of a complete graph with more than @xmath24 vertices , one can find a monochromatic monotone path with @xmath4 vertices .
an easy construction shows that this statement does not remain true for graphs with @xmath24 vertices . the result and its proof
readily generalize to colorings with @xmath25 colors : in this case , we have to replace @xmath24 with @xmath26 .
the corresponding problems for hypergraphs are quite hard and their solution usually utilizes some variant or extension of the celebrated hypergraph regularity lemma .
the only asymptotically sharp result of this kind was proved by haxell , uczak et al .
let @xmath27 be a @xmath1-uniform hypergraph .
a _ tight path _ or , shortly , a _ path of length _
@xmath4 in @xmath28 is comprised of a set of @xmath4 distinct vertices @xmath29 and the set of @xmath30 hyperedges of the form @xmath31 where @xmath3 .
it was shown in @xcite that no matter how we 2-color all triples of a complete 3-uniform hypergraph on @xmath32 vertices , one of the color classes will contain a path of length roughly @xmath4 . this result is best possible .
it was generalized by nagle , olsen et al .
@xcite , cooley , fountoulakis et al .
@xcite , and conlon et al .
these results imply that every @xmath9-coloring of the hyperedges of a complete @xmath1-uniform hypergraph on @xmath33 vertices contains a monochromatic copy of any @xmath4-vertex hypergraph with maximum degree @xmath34 , where @xmath35 is a suitable constant depending only on @xmath1 , @xmath9 , and @xmath34 .
in the present paper , we consider the analogous problem for _ ordered _ hypergraphs .
let @xmath36 denote the _
complete @xmath1-uniform hypergraph _ ,
consisting of all @xmath1-element subsets ( @xmath1-tuples ) of the vertex set @xmath8 = \{1,2, ... ,n\}$ ] .
for any @xmath4 positive integers , @xmath0 , we say that the hyperedges @xmath37 @xmath3 , form a _
monotone ( tight ) path of length _ @xmath4 .
we will present several results on finding monochromatic monotone paths in edge - colored ordered hypergraphs . as a geometric application , in section 7
we will show that every family of at least @xmath38 plane convex bodies in general position , any pair of which share at most two boundary points , has @xmath4 members in convex position .
this substantially improves the previous double - exponential bound of hubard et al .
@xcite for this problem . here , and
throughout the paper , all logarithms unless otherwise stated are in base 2 .
let @xmath10 denote the smallest integer @xmath7 such that for every @xmath9-coloring of the hyperedges ( @xmath1-tuples ) of @xmath36 , there exists a monotone path of length @xmath4 such that all of its hyperedges are of the same color .
using this notation for graphs @xmath39 , the corollary of the erds - szekeres theorem or the dilworth theorem mentioned above can be rephrased as @xmath40 for 3-uniform hypergraphs , a straightforward generalization of the cup - cap argument of erds and szekeres @xcite gives [ erdos - szekeres ] let @xmath41 be the smallest integer such that for every @xmath42-coloring of all triples in @xmath8 $ ] , there exists a monochromatic monotone path of length @xmath4
. then we have @xmath43 for a larger ( but fixed ) number of colors , we prove the following theorem which provides lower and upper bounds that are tight apart from a logarithmic factor in the exponent . [ main1 ] for any integers @xmath44 and @xmath45 , we have @xmath46 define the tower function @xmath47 recursively as follows .
let @xmath48 and @xmath49 , so @xmath47 is a tower of @xmath50 @xmath4 s with an @xmath51 on top .
we let @xmath52 . using a stepping - up " approach , developed by erds and hajnal ( see @xcite ) and strengthened in @xcite , we show the following extension of theorem [ main1 ] to @xmath1-uniform hypergraphs , for any @xmath53 .
[ main2 ] for every @xmath54 and @xmath9 , there are positive constants @xmath55 depending only on @xmath1 and @xmath9 such that , for any @xmath56 , we have @xmath57 in section [ section2 ] , we prove the recursive upper bound @xmath58 .
this upper bound , together with equation ( [ dil ] ) , implies that @xmath59 for @xmath54 .
the upper bounds in theorems [ main1 ] and [ main2 ] follow from this inequality . in sections [ section3 ] and [ section4 ]
, we establish the lower bound in theorem [ main1 ] .
we also prove a general statement there ( theorem [ stepup4 ] ) , providing a lower bound on @xmath10 , which is exponential in @xmath60 , for every @xmath61 . putting these results together
, the lower bound in theorem [ main2 ] readily follows .
the case @xmath62 is crucial to understanding the growth of @xmath10 .
indeed , the stepping - up lower bound mentioned above ( see theorem [ stepup4 ] ) together with the recursive upper bound in theorem [ alternativeupperbound ] and inequality [ trivial123 ] , described in the next subsection , show that closing the gap between the upper and lower bounds on @xmath10 in the special case @xmath62 would also close the gap for all larger values of @xmath1 .
consider the following game played by two players , _ builder _ and _
painter_. for @xmath63 , at the beginning of stage @xmath64 , a new vertex @xmath65 is added ( the vertices @xmath66 for @xmath67 already are present ) , and for each @xmath68-tuple of vertices @xmath69 with @xmath70 , builder decides ( in any order ) whether to draw the edge @xmath71 . if builder draws the edge , then painter has to immediately color it in one of @xmath9 colors @xmath72 .
the _ ( vertex ) online ramsey number _
@xmath73 is the minimum number of edges builder has to draw to guarantee a monochromatic monotone path of length @xmath4 .
an ordered @xmath1-uniform hypergraph @xmath74 is said to be _
@xmath75-path ramsey _ if for every @xmath9-edge - coloring of @xmath74 one can find a monochromatic monotone path of length @xmath4 .
the _ size ramsey number _
@xmath76 is the minimum number of edges of an ordered @xmath1-uniform hypergraph @xmath74 which is @xmath75-path ramsey .
it follows from the definitions that @xmath77 indeed , the first inequality comes from the following painter strategy .
let @xmath78 be the @xmath79th vertex that builder adds in which there is an edge whose largest vertex is @xmath78 ( there may be vertices not given labels ) .
that is , @xmath80 where @xmath81 is the @xmath79th stage for which at least one edge is added .
painter colors each edge @xmath82 with @xmath83 the same color as the color of @xmath84 in a @xmath9-edge - coloring of the ordered complete @xmath1-uniform hypergraph on @xmath85 vertices with no monochromatic path of length @xmath30 .
all other edges , i.e. , those containing a vertex which is not the largest vertex in some edge , can be colored arbitrarily .
note that every monotone path of length @xmath4 contains a path of length @xmath30 such that each vertex is the largest vertex in an edge of the ordered hypergraph .
therefore , this coloring has no monochromatic monotone path of length @xmath4 , and painter guarantees that there are at least @xmath86 edges when the first monochromatic monotone path of length @xmath4 appears .
on the other hand , if builder selects all edges of a @xmath75-path ramsey ordered @xmath1-uniform hypergraph , she will definitely win ; which proves the second inequality .
the third inequality is trivial , taking into account that the ordered complete @xmath1-uniform hypergraph @xmath36 with @xmath87 is @xmath75-path ramsey .
conlon , fox , and sudakov @xcite used online ramsey numbers to give an upper bound for the ramsey number of a _ complete _ @xmath1-uniform hypergraph with @xmath4 vertices . here
we apply this technique to establish an upper bound for the ramsey number of a _ monotone path _ of length @xmath4 .
more precisely , we show [ alternativeupperbound ] for every @xmath88 and for every @xmath89 , we have @xmath90 in the special case @xmath62 , this bound is worse than the upper bound in theorem [ main1 ] . however , theorem [ alternativeupperbound ] is already useful for @xmath91 .
indeed , any improvement on the upper bound in theorem [ main2 ] for @xmath62 , together with the trivial upper bound on @xmath92 in ( [ trivial123 ] ) , will lead to analogous improvements for all larger @xmath1 .
for example , if , for fixed @xmath9 , we would know that @xmath93 , then using theorem [ alternativeupperbound ] we would also know that @xmath94 .
note that the recursive bound in theorem [ upineq ] does not give such an improvement even if we knew that @xmath95 is exponential in @xmath17 .
similarly , the stepping - up lower bound in theorem [ stepup4 ] shows that any improvement on the lower bound in theorem [ main2 ] for @xmath62 will lead to analogous improvements for all larger @xmath1 .
in addition to being useful for bounding classical ramsey numbers as discussed above , online and size ramsey numbers are now well - studied topics in their own right .
the _ size ramsey number _ @xmath96 of a graph @xmath28 is the minimum number of edges of a graph @xmath74 in which every @xmath42-edge - coloring of @xmath74 contains a monochromatic copy of @xmath28 .
the study of these numbers was initiated by erds et al .
a fundamental problem of erds in this area is to determine the growth of the size ramsey number of paths .
beck @xcite solved this problem , proving that these numbers grow linearly .
that is , @xmath97 , where @xmath98 is an absolute constant .
in contrast , rdl and szemerdi @xcite disproved a conjecture of beck by showing that there are graphs of maximum degree @xmath99 whose size ramsey number grows superlinear in the number of vertices . in the other direction ,
kohayakawa et al .
@xcite recently showed that graphs on @xmath4 vertices with fixed maximum degree @xmath34 have size ramsey number at most @xmath100 .
another online ramsey game which is quite close to ours was introduced independently by beck @xcite and kurek and ruciski @xcite . in this game , there are two players , builder and painter , who move on the originally empty graph with an unbounded number of vertices . at each step
, builder draws a new edge and painter has to color it either red or blue immediately . the _ edge
online ramsey number _
@xmath101 is the minimum number of edges that builder has to draw in order to force painter to create a monochromatic @xmath28 . as builder
can simply draw the edges of a given graph , we have the inequality @xmath102 .
a basic conjecture in this area due to rdl is to show that @xmath103 .
conlon @xcite made substantial progress on this conjecture , showing that @xmath104 holds infinitely often , where @xmath105 is an absolute constant .
randomized variants of the edge online ramsey number were studied in @xcite , @xcite , @xcite .
note that the problem of estimating @xmath73 and @xmath76 is most interesting in the case of graphs ( @xmath106 ) .
indeed , for larger @xmath1 , @xmath10 grows roughly @xmath107-fold exponential in @xmath17 , so the lower and upper bounds on these numbers are roughly determined by @xmath10 .
we will therefore focus our attention on the interesting case @xmath106 . in view of ( [ trivial123 ] )
, one may think that the functions @xmath73 and @xmath76 can not differ too much . rather surprisingly , already for graphs ( @xmath106 ) this is not the case . in section [ sectionsizeandvertex ] , we show that the size ramsey number satisfies @xmath108 . on the other hand ,
also in that section we prove [ keyonlinethm ] we have @xmath109 and for every fixed @xmath14 , there are constants @xmath110 and @xmath111 such that @xmath112 moreover , the proof of theorem [ keyonlinethm ] shows that builder has a winning strategy which uses @xmath113 vertices and each vertex @xmath114 belongs to at most @xmath115 edges @xmath116 with @xmath117 .
this theorem shows that , by using the information on the initial portion of the coloring , builder can substantially reduce the number of edges that guarantee a monochromatic monotone path . in the last section ,
we apply theorem [ main1 ] and lemma [ clique ] to a problem for families of convex bodies . to formulate our question ,
we need some definitions .
a family @xmath118 of convex bodies ( compact convex sets with nonempty interior ) in the plane is said to be in _ convex position _ if none of its members is contained in the convex hull of the union of the others .
we say that @xmath118 is in _ general position _ if 1 .
every three members of @xmath118 are in convex position ; 2 .
no two members @xmath119 have a common tangent that meets @xmath120 ; and 3 .
no three members of @xmath118 share a common tangent .
bisztriczky and fejes - tth @xcite generalized the happy ending theorem of erds and szekeres @xcite , @xcite as follows .
they proved that for every @xmath4 , there exists a function @xmath121 such that any family of @xmath121 pairwise disjoint convex bodies in general position in the plane contains @xmath4 members in convex position .
it was proved in @xcite that @xmath122 . in @xcite ,
the bisztriczky - fejes tth theorem has been extended to families of _ noncrossing _ convex bodies , that is , to convex bodies , no pair of which share more than _ two _ boundary points .
it was proved that there exists a function @xmath123 such that from every family of @xmath7 noncrossing convex bodies in general position in the plane one can select @xmath4 members in convex position .
more recently , hubard et al .
@xcite showed that the function @xmath124 grows at most double exponentially in @xmath4 .
we use theorem [ main1 ] to obtain a much better bound .
[ convex ] let @xmath125 denote the smallest integer such that every family of @xmath7 noncrossing convex bodies in general position in the plane has @xmath4 members in convex position .
then we have @xmath126 note that no analogue of the last theorem is true if we drop the assumption that the bodies are noncrossing .
one can construct a family of arbitrarily many pairwise crossing rectangles that which are in general position , but no four of them are in convex position ( see @xcite ) .
theorem [ convex ] can be established , as follows .
first , we choose a coordinate system , in which the _ left endpoint _ ( i.e. , the leftmost point ) of every member of the family is unique , and the @xmath51-coordinates of the left endpoints are different .
let @xmath127 denote the members of our family , listed in the increasing order of the @xmath51-coordinates of their left endpoints .
suppose that @xmath128 . in the last section ,
we define a 3-coloring of all ordered triples @xmath129 with @xmath130 .
we will show that this coloring has the property that if a sequence of sets @xmath131 @xmath132 induces a monochromatic monotone path , then they are in convex position . according to theorem [ main1 ] , @xmath133 therefore , our colored triple system contains a monochromatic monotone path of length @xmath4 .
in other words , there exists a sequence @xmath134 such that all ordered triples induced by the sets @xmath135 are of the same color .
it follows from the special property of our coloring mentioned above that the sets @xmath136 are in convex position .
this will complete the proof of theorem [ convex ] .
* organization : * the rest of the paper is organized as follows . in the next section
we prove a recursive upper bound on @xmath10 . in section [ section3 ] , we prove results on monochromatic walks in digraphs and the minimum chromatic number of a @xmath75-path ramsey graph which will be used in the subsequent two sections . in section [ section4 ] , we prove lower bounds on @xmath10 . in section [ sectionsizeandvertex ] ,
we prove bounds on size and online ramsey numbers of paths , as well as an alternative upper bound on @xmath10 in terms of online ramsey numbers . in section [ transitivesect ]
, we prove a lemma relating monochromatic cliques in `` transitive '' colorings to monochromatic monotone paths in general colorings .
we use this lemma in section [ geomsect ] in the proof of theorem [ convex ] .
we finish with some concluding remarks and open problems in section [ concluding ] .
given any two positive integers @xmath137 , let @xmath138 $ ] denote the set of positive integer @xmath139 with @xmath140 , and @xmath141=[1,i]$ ] .
we begin by proving a recursive upper bound on @xmath10 .
[ upineq ] for any integers @xmath5 and @xmath6 , we have @xmath142 * proof . *
suppose for contradiction that there is a @xmath9-coloring @xmath143 \choose k } \rightarrow [ q]\ ] ] of the edges of the complete @xmath1-uniform ordered hypergraph @xmath36 on @xmath144 vertices with no monochromatic path of length @xmath4 .
define the auxiliary coloring @xmath145 \choose k-1 } \rightarrow [ k-1,n-1]^{q-1},\ ] ] as follows . for any @xmath146 , we let @xmath147 , where @xmath148 is the length of the longest monochromatic path in color @xmath79 ending with @xmath149 . by convention ,
any @xmath16 vertices @xmath149 form a monochromatic monotone path of length @xmath16 with respect to coloring @xmath150 .
thus , we have @xmath151 , for every @xmath152 . by our assumption ,
@xmath153 , so that in coloring @xmath154 , there is a monochromatic monotone path of length @xmath4 .
let @xmath155 denote the vertices of this path . to complete the proof of the theorem , it is sufficient to show that these vertices form a monochromatic monotone path in color @xmath9 , with respect to the coloring @xmath150 .
suppose that this is not the case .
then , for some @xmath156 and @xmath157 , we have @xmath158 . since the vertices @xmath159 form a monochromatic monotone path with respect to the coloring @xmath154
, it follows that in coloring @xmath150 , the length of the longest monochromatic monotone path of color @xmath79 ending with @xmath160 is the same as the length of the longest monochromatic monotone path of color @xmath79 ending with @xmath161 .
however , any longest monochromatic monotone path of color @xmath79 ending with @xmath160 can be extended to a longer monochromatic monotone path of color @xmath79 , by adding the vertex @xmath162 .
this contradiction completes the proof of the theorem .
now we are in a position to prove the upper bound in theorem [ main1 ] .
* proof of the upper bounds in theorems [ main1 ] and [ main2 ] .
* we prove the stronger inequality @xmath163 for every @xmath54 .
the proof is by induction on @xmath1 . by ( [ dil ] )
, we have @xmath164 . according to theorem [ upineq ] ,
@xmath165 now suppose that the desired inequality holds for @xmath1 .
then it also holds for @xmath166 , because , again by theorem [ upineq ] , we have @xmath167
let @xmath168 denote the _ complete digraph _ on @xmath7 vertices , that is , the directed graph in which each pair of distinct vertices is connected by two edges with opposite orientations .
a _ walk of length @xmath4 _ in a digraph is a sequence of @xmath4 vertices @xmath169 with possible repetitions such that for every @xmath170 , the directed edge @xmath171 belongs to the digraph .
if @xmath172 , then the walk is called _ closed_. note that closed walks can be used to construct walks which are arbitrarily long . a digraph with no closed walk is _ acyclic_.
let @xmath173 be the smallest number @xmath7 such that for every @xmath9-coloring of the edges of @xmath168 , there is a monochromatic walk of length @xmath4 .
first , we show that for a fixed @xmath9 , the order of magnitude of @xmath173 is @xmath17 . in the next section
, this fact is used to establish the lower bound on @xmath95 , stated in theorem [ main1 ] .
[ ftheorem ] for any integers @xmath174 , we have @xmath175 * proof . *
we first establish the upper bound .
consider a @xmath9-coloring of the edges of @xmath168 with @xmath176 .
we have to show that there is a monochromatic walk with @xmath4 vertices .
suppose that the set of edges of color @xmath9 does not determine a walk of length @xmath4 .
then these edges form an acyclic digraph . hence
, there is an ordering of the vertices of @xmath168 so that the edges of color @xmath9 go backwards .
thus , removing the backwards edges , we have an ordered complete graph on @xmath177 vertices , whose edges are colored with @xmath178 colors . in one of the color classes , we can find a monotone path with @xmath4 vertices , which is a monochromatic directed path in @xmath168 .
the lower bound follows from lemma [ lowf ] below .
indeed , letting @xmath179 , it is easy to check that @xmath180 and @xmath181 , so @xmath182 [ lowf ] for any integers @xmath183 , we have @xmath184 . * proof .
* define the coloring @xmath154 of the edges of the complete digraph on @xmath185^{q-1}$ ] , as follows . for any pair of distinct vertices , @xmath186 , define the color of the directed edge @xmath187 as the smallest coordinate @xmath79 for which @xmath188 . if there is no such coordinate , that is , if @xmath189 for all @xmath190 ) , then color the directed edge @xmath187 with color @xmath9 . for each color @xmath191 ,
the length of a longest monochromatic walk of color @xmath79 is @xmath4 , because traversing any edge of color @xmath79 , the @xmath79th coordinate must increase .
on the other hand , along any walk in color class @xmath9 , the coordinates are never allowed to increase , and in each step at least one of them must strictly decrease , so that the sum of the coordinates strictly decreases .
the sum of the coordinates of a point in @xmath185^{q-1}$ ] is at least @xmath178 and is at most @xmath192 .
therefore , the length of such a monochromatic walk does not exceed @xmath193 .
recall from subsection [ online ] that an ordered graph @xmath74 is called _ @xmath75-path ramsey _ if for every @xmath9-edge - coloring of @xmath74 there exists a monochromatic monotone path of length @xmath4 .
let @xmath194 denote the minimum chromatic number of an ordered graph @xmath74 which is @xmath75-path ramsey .
we close this section by showing that the functions @xmath194 and @xmath173 are actually identical .
therefore , theorem [ ftheorem ] implies that , for a fixed @xmath9 , @xmath194 also grows on the order of @xmath17 .
the ( classical ) _ ramsey number _
@xmath195 is the minimum @xmath7 such that every @xmath9-edge - coloring of the complete graph on @xmath7 vertices contains a monochromatic clique on @xmath4 vertices .
[ chif ] for any integers @xmath196 , we have @xmath197 * proof .
* we first show that @xmath198 . consider a @xmath9-edge - coloring @xmath154 of the edges of the complete digraph with vertex set @xmath199 $ ] without a monochromatic walk of length @xmath4 .
assume for contradiction that there exists an ordered graph @xmath74 with @xmath200 , which is @xmath75-path ramsey , and consider a partition @xmath201 into independent sets with @xmath202 .
define a @xmath9-coloring @xmath203 of the edges of @xmath74 , as follows .
if @xmath204 are adjacent with @xmath205 and @xmath206 and @xmath207 , let @xmath208 . if the vertices @xmath209 form a monochromatic monotone path with respect to the edge - coloring @xmath203 of @xmath74 , then denoting by @xmath210 the integer for which @xmath211 , we have that @xmath212 form a monochromatic walk with respect to the edge - coloring @xmath154 .
this is a contradiction , showing that our assumption @xmath200 was wrong . to show that @xmath213 , we define an ordered graph @xmath28 , which is @xmath173-colorable and @xmath75-path ramsey .
let @xmath214 $ ] , where @xmath215 and @xmath216 is the @xmath217-color ramsey number for the complete graph on @xmath4 vertices with @xmath218 .
connect two vertices @xmath219 by an edge in @xmath28 if @xmath220 is not a multiple of @xmath64 .
the partition of @xmath221 modulo @xmath64 defines a proper @xmath64-coloring of @xmath28 . since the first @xmath64 vertices of @xmath28 form a clique
, we have @xmath222 .
it remains to prove that @xmath28 is @xmath75-path ramsey .
consider an edge - coloring @xmath223 of @xmath28 with @xmath9 colors .
define an auxiliary @xmath217-coloring @xmath224 of the complete graph on @xmath225 $ ] vertices , where the color of the edge between @xmath226 is given as a @xmath227 matrix @xmath228 with @xmath229 s in the diagonal and @xmath230 for every @xmath231 . obviously , the number of colors used in this coloring is at most the number of possible matrices @xmath232 , which is equal to @xmath233 . by the definition of the ramsey number @xmath216 ,
@xmath225 $ ] contains a monochromatic clique on @xmath4 vertices , with respect to the coloring @xmath224 .
denote the vertices of such a clique by @xmath234 .
define a @xmath9-edge - coloring @xmath235 of the complete digraph on @xmath236 $ ] , where @xmath237 .
since @xmath215 , there is a monochromatic walk @xmath212 of length @xmath4 with respect to the coloring @xmath235 .
then the vertices @xmath238 form a monochromatic monotone path in the @xmath9-edge - coloring @xmath223 of @xmath28 , which completes the proof .
in this section we adapt the stepping - up approach of erds and hajnal on hypergraph ramsey numbers to provide lower bounds for @xmath10 , stated in theorems [ main1 ] and [ main2 ] .
first we address the case @xmath62 .
[ stepupto3 ] for any integers @xmath196 , we have @xmath239 * proof .
* let @xmath154 be a @xmath9-coloring of the edges of the complete digraph @xmath240 on vertex set @xmath241 $ ] without a monochromatic walk on @xmath242 vertices .
we use @xmath154 to define a @xmath9-coloring @xmath150 of the hyperedges ( triples ) of the complete ordered @xmath99-uniform hypergraph @xmath243 on the vertex set @xmath244 $ ] with @xmath245 , as follows . for
any @xmath246 , write @xmath247 with @xmath248 for each @xmath79 . for @xmath249 ,
let @xmath250 denote the largest @xmath79 for which @xmath251 .
obviously , we have @xmath252 for every triple @xmath253 .
given any triple @xmath253 , define @xmath254 . to complete the proof of the theorem ,
it is enough to show that , with respect to this coloring , there is no monochromatic monotone path of length @xmath4 .
suppose for contradiction that there is such a path , and denote its vertices by @xmath255 .
letting @xmath256 for every @xmath257 , it follows from the definition of the coloring @xmath150 that the integers @xmath258 $ ] induce a monochromatic walk in the digraph @xmath240 , with respect to the coloring @xmath154 .
this contradiction completes the proof . from lemma [ lowf ]
, we have the estimate @xmath259 with @xmath260 .
it is easy to check that @xmath181 if @xmath15 . together with theorem [ stepupto3 ] , we immediately obtain the following corollary , which is the same as the lower bound in theorem [ main1 ] .
[ corclear ] for any integers @xmath44 and @xmath45 , we have @xmath261 we next give a recursive lower bound on @xmath10 for @xmath91 .
the proof is an adaptation of an improved version of the stepping - up technique , due to conlon , fox , and sudakov @xcite .
it easy to see that , together with corollary [ corclear ] , this gives the lower bound in theorem [ main2 ] .
[ stepup4 ] for any integers @xmath262 and @xmath6 , we have @xmath263 * proof . *
we start the proof in a way similar to theorem [ stepupto3 ] .
let @xmath154 be a @xmath9-coloring with colors @xmath264 of the edges of the complete ordered @xmath68-uniform hypergraph on @xmath265 vertices without a monochromatic monotone path on @xmath4 vertices .
we use @xmath154 to define a @xmath9-coloring @xmath150 of the edges of the complete ordered @xmath1-uniform hypergraph @xmath36 on the vertex set @xmath244 $ ] with @xmath266 , as follows . for
any @xmath246 , write @xmath267 with @xmath248 for each @xmath79 . for @xmath249 ,
let @xmath250 denote the largest @xmath79 for which @xmath251 .
as in the proof of theorem [ stepupto3 ] , we have @xmath252 for every triple @xmath253 . given any @xmath1-tuple @xmath268 of @xmath269 , consider the integers @xmath270 .
if @xmath271 form a monotone sequence , then let @xmath272 .
now we have to color the @xmath1-tuple @xmath273 in the case when @xmath271 is not monotone .
we say that @xmath79 is a _ local minimum _ if @xmath274 , a _ local maximum _ if @xmath275 , and a _ local extremum _ if it is either a local minimum or a local maximum . since @xmath276 for every @xmath79 , every nonmonotone sequence @xmath271 has a local extremum .
let @xmath277 denote the first local extremum .
if @xmath271 is not monotone , we define @xmath278 if @xmath277 is even and a local maximum , or if @xmath277 is odd and a local minimum .
otherwise , let @xmath279 .
suppose for contradiction that the vertices @xmath280 induce a monochromatic monotone path with respect to the @xmath9-coloring @xmath150 .
for @xmath281 , let @xmath282 . since there is no monochromatic monotone path on @xmath4 vertices with respect to the coloring @xmath154 , the sequence @xmath283 $ ] must have a local extremum . moreover ,
if @xmath277 denotes the location of the first local extremum , we have @xmath284 .
case 1 : the first local extremum @xmath277 satisfies @xmath285 .
we claim that , if @xmath286 , then the first two edges of the path , @xmath287 and @xmath288 , receive different colors , contradicting our assumption that the path is monochromatic . if @xmath289 , then the edges @xmath290 and @xmath291 receive different colors , which is again a contradiction . indeed , in either case the _ type _ ( maximum or minimum ) of the first local extremum is the same for these two edges , but their locations differ by one and hence have different parity , which implies that @xmath292 and @xmath293 receive different colors .
case 2 : the first local extremum is @xmath294 . case 2(a ) : @xmath99 is a local extremum .
note that consecutive extrema have different types , so that the types of @xmath42 and @xmath99 are different
. therefore , the first two edges of the path , @xmath295 and @xmath288 , must have different colors , contradicting our assumption that the path is monochromatic .
indeed , for each of these edges , the first local extremum is the second @xmath296 , but these extrema are of different types , and hence @xmath292 and @xmath293 receive different colors . case 2(b ) : @xmath99 is not a local extremum . as the sequence @xmath297 can not be monotone , the sequence of @xmath296 s has a second local extremum @xmath298 satisfying @xmath299 . if @xmath300 , then the edges @xmath301 and @xmath302 have different colors . if @xmath303 , then the edges @xmath304 and @xmath305 have different colors . indeed , in either case , the first local extremum for the pair of edges are of the same type , but their locations differ by one and hence have different parity , which implies they have different colors .
this completes the proof of the theorem .
we begin this section by estimating vertex online ramsey numbers , introduced in subsection [ online ] of the introduction .
we then prove an alternative upper bound on @xmath10 , in terms of the online ramsey number @xmath306 .
we end this section by estimating the size ramsey number @xmath307 . in this subsection
we study vertex online ramsey numbers .
we first relate the online ramsey game to another game , which we call the _ @xmath75-lattice game _ ( or _ lattice game _ for short ) . in this game
there are two players , builder and coordinator . during the game ,
a sequence of ( not necessarily distinct ) points in the @xmath9-dimensional grid @xmath308 with positive coordinates is built .
after stage @xmath79 , the sequence has length @xmath79 . in stage @xmath309 , a new point @xmath310 is added to the sequence as described below . in each step of stage @xmath309
, builder picks a point @xmath311 with @xmath312 already in the sequence .
coordinator then decides a coordinate @xmath313 $ ] .
the point @xmath310 must satisfy that its @xmath1th coordinate is greater than the @xmath1th coordinate of @xmath314 , i.e. , @xmath315 .
after some steps , builder decides it is time to end this stage , and coordinator picks a point @xmath310 satisfying the conditions provided by the steps .
note that in stage @xmath316 , as there are no points yet in the sequence , no steps are taken and the only thing that happens is coordinator picks a point @xmath317 in @xmath318 to begin the sequence .
let @xmath319 be the minimum ( total ) number of steps needed for builder to guarantee that there is a point in the sequence with a coordinate at least @xmath4 .
the following lemma relates these numbers with online ramsey numbers .
[ equivlattice ] for all @xmath9 and @xmath4 , we have @xmath320 we first prove the bound @xmath321 . as builder and painter
are playing the ramsey game , builder and coordinator play a corresponding lattice game .
we will show that as long as painter guarantees in the ramsey game there is no monochromatic monotone path of length @xmath4 , coordinator can guarantee in the lattice game that no point in the sequence will have a coordinate at least @xmath4 .
each vertex @xmath66 in the ramsey game corresponds to a point @xmath322 in the lattice game . to prove the desired result
, coordinator will guarantee that the @xmath1th coordinate of the point @xmath322 in the lattice game is the length of the longest monotone monochromatic path in color @xmath1 ending at @xmath66 in the ramsey game . at each step of stage @xmath79 in the lattice game
, builder picks the point @xmath314 .
then , in the ramsey game builder picks the edge @xmath323 to add .
the color @xmath1 , @xmath324 , that painter picks to color the edge @xmath323 is the coordinate which coordinator picks so that the @xmath1th coordinate of @xmath322 is greater than the @xmath1th coordinate of @xmath314 . at the end of this stage
, coordinator picks the point @xmath322 such that the @xmath1th coordinate of @xmath322 is the length of the longest monotone monochromatic path in color @xmath1 in the ramsey game ending at @xmath66 .
this shows that , no matter what builder s strategy is , coordinator can mimic painter s strategy and continue the lattice game for as long as the ramsey game .
next we show that @xmath325 , which would complete the proof . as builder and coordinator
are playing the lattice game , builder and painter play a corresponding ramsey game .
we will show that as long as coordinator guarantees in the lattice game that no point in the sequence will have a coordinate at least @xmath4 , painter can guarantee in the ramsey game there is no monochromatic monotone path of length @xmath4 .
each point @xmath322 in the lattice game will have a corresponding vertex @xmath66 in the ramsey game . to prove the desired result , painter will guarantee that for each vertex @xmath66 and each color @xmath1 , the length of the longest monochromatic path in color @xmath1 ending at @xmath66 is at most the @xmath1th coordinate of @xmath322 .
at each step of stage @xmath79 in the ramsey game , the builder picks an edge @xmath323 to add .
then , in the corresponding lattice game , building picks the point @xmath314 to compare @xmath322 with .
the coordinate @xmath1 that coordinator picks so that the @xmath1th coordinate of @xmath322 is greater than the @xmath1th coordinate of @xmath314 will be the color @xmath1 that painter uses to color the edge @xmath323 .
the length of the longest monochromatic monotone path in color @xmath1 with last edge @xmath323 is one more than the length of the longest monochromatic monotone path in color @xmath1 ending at @xmath326 , and this length is at most the @xmath1th coordinate of @xmath314 by construction .
since the @xmath1th coordinate of @xmath322 is greater than the @xmath1th coordinate of @xmath314 , this guarantees that the length of the longest monochromatic monotone path in color @xmath1 ending at @xmath66 is at most the @xmath1th coordinate of @xmath322 .
this shows that , no matter what builder s strategy is , painter can mimic coordinator s strategy and continue the ramsey game for as long as the lattice game .
the next lemma tells us the minimum number of steps required for builder to win the @xmath75-lattice game . the lattice @xmath327 naturally comes with a partial order @xmath328 , where @xmath329 if @xmath330 and the @xmath1th coordinate of @xmath9 is at least the @xmath1th coordinate of @xmath331 for @xmath324 .
we will use this order further on .
[ stepslg ] in the @xmath75-lattice game , the minimum number of stages ( i.e. , the number of points in the constructed sequence ) builder needs to guarantee a point in the sequence with a coordinate at least @xmath4 is @xmath332 . indeed ,
in each stage , builder , by picking all previous points at steps , can ensure that the points in the constructed sequence are distinct .
hence , as there are only @xmath26 points in @xmath318 with coordinates at most @xmath242 , by the pigeonhole principle , the minimum number of stages required for builder is at most @xmath332 . picking a linear extension of the partial order @xmath328 on @xmath333^q$ ] described above
, coordinator can guarantee that @xmath322 is the @xmath79th smallest point in the linear extension . in this way , the minimum number of stages required for builder is at least @xmath332 , which completes the proof .
for a finite subset @xmath334 and a point @xmath335 , define the _ position of @xmath331 with respect to @xmath336 _ to be the maximum @xmath64 such that for each @xmath313 $ ] , there are at least @xmath64 points @xmath337 such that the @xmath1th coordinate of @xmath331 is at least the @xmath1th coordinate of @xmath338 . [ median ]
let @xmath336 be a finite nonempty subset of @xmath327 . 1 .
there is a point @xmath339 whose position with respect to @xmath336 is at least @xmath340 .
if @xmath335 is such that there is no @xmath337 satisfying @xmath341 , then the position of @xmath331 with respect to @xmath336 is at most @xmath342 .
we first show the first part . for each @xmath64 ,
the number of elements of position less than @xmath64 is less than @xmath343 .
indeed , for each of the @xmath9 coordinates , less than @xmath64 elements of @xmath336 have position less than @xmath64 in that coordinate .
all other elements have position at least @xmath64 .
setting @xmath344 , there is a point of @xmath336 of position at least @xmath340 .
we now show the second part .
suppose the position of @xmath331 with respect to @xmath336 is @xmath64 .
delete from @xmath336 all elements @xmath338 for which there is @xmath313 $ ] such that the @xmath1th coordinate of @xmath338 is larger than the @xmath1th coordinate of @xmath331 .
since @xmath331 has position @xmath64 , the number of deleted elements is at most @xmath345 . if @xmath346 , then @xmath331 is at least as large as an element of @xmath336 , a contradiction .
hence , @xmath347 , and we have @xmath348 , which completes the proof .
we now prove an upper bound on @xmath319 .
[ uplattice ] let @xmath349 and @xmath350 .
in the @xmath75-lattice game , builder has a winning strategy which uses the minimum possible number of stages , which is @xmath351 , such that each stage uses at most @xmath352 steps .
in particular , we have @xmath353 by lemma [ stepslg ] , the minimum possible number of stages builder needs to win the @xmath75-lattice game is indeed @xmath351 .
as there are at most @xmath26 points in @xmath327 with positive coordinates at most @xmath242 , it suffices for builder to guarantee at each stage @xmath79 , using at most @xmath352 steps , that coordinator picks a point @xmath322 not already in the sequence . to accomplish this
, it suffices for builder to guarantee that for each point @xmath314 in the sequence with @xmath354 , there is @xmath313 $ ] , such that the @xmath1th coordinate of @xmath322 must be greater than the @xmath1th coordinate of @xmath314 .
suppose we have finished @xmath50 stages , and we are now on stage @xmath79 . let @xmath355 denote the set of maximal elements in the already constructed sequence of @xmath50 points .
we have @xmath356 as @xmath355 can not contain two points with the same @xmath178 first coordinates . in the first step of stage @xmath79
, builder picks a point @xmath357 of position in @xmath355 at least @xmath358 .
such a point exists by the first part of lemma [ median ] .
coordinator then picks a coordinate @xmath359 . by the definition of position
, there are at least @xmath358 points in @xmath355 whose @xmath359th coordinate is at most the @xmath359th coordinate of @xmath360 .
therefore , point @xmath322 must have @xmath359th coordinate larger than the @xmath359th coordinate of these at least @xmath358 points in @xmath355 .
let @xmath361 denote those points in @xmath355 whose @xmath359th coordinate is larger than the @xmath359th coordinate of @xmath360 , so @xmath362 .
after step @xmath139 , we have a set @xmath363 with @xmath364 such that for every point @xmath365 , there is @xmath313 $ ] such that the @xmath1th coordinate of @xmath322 is guaranteed to be larger than the @xmath1th coordinate of @xmath314 .
if @xmath363 is nonempty , then builder moves on to step @xmath366 and picks a point @xmath367 of position in @xmath363 at least @xmath368 .
such a point exists by the first part of lemma [ median ] .
coordinator then picks a coordinate @xmath369 . by the definition of position
, there are at least @xmath368 points in @xmath363 whose @xmath369th coordinate is at most the @xmath369th coordinate of @xmath370 .
therefore , point @xmath322 must have @xmath369th coordinate larger than the @xmath369th coordinate of these at least @xmath368 points in @xmath363 .
let @xmath371 denote those points in @xmath363 whose @xmath369th coordinate is larger than the @xmath369th coordinate of @xmath370 , so @xmath372 .
if @xmath363 is empty , then builder decides to end the stage and coordinator must pick a point @xmath322 satisfying the required conditions . note that we eventually will have an empty @xmath363 as @xmath355 is a finite set , and @xmath373 is a proper subset of @xmath374 .
since in the previous step @xmath374 is nonempty , we have @xmath375 and so the number @xmath139 of steps satisfies @xmath376 since @xmath355 are the maximal elements of the already constructed sequence , and for every point @xmath377 , there is @xmath313 $ ] such that the @xmath1th coordinate of @xmath322 is greater than the @xmath1th coordinate of @xmath331 , then @xmath322 can not be an element of the already constructed sequence .
this completes the proof .
we remark that one can do a little better in the above proof by improving the bound on @xmath378 by using a result of de bruijn , van ebbenhorst tengbergen , and kruyswijk @xcite which extends sperner s theorem .
this result says that the set of elements in @xmath333^q$ ] whose coordinates sum to @xmath379 forms a maximum antichain in the poset on @xmath333^q$ ] .
since @xmath355 is an antichain , the size of this set is an upper bound on @xmath378 .
we next present a strategy for coordinator which gives a lower bound for @xmath319 .
[ lowlattice ] for each fixed @xmath14 , we have @xmath380
. we present a strategy for coordinator which satisfies that the number of steps in the stage the @xmath79th point not already in the sequence is added is at least a certain number depending on @xmath79 which will be defined later in the proof .
level @xmath381 of the grid @xmath333^q$ ] consists of all points in this grid whose coordinates sum to @xmath381 .
the new points will be added in levels , so no points of larger level are added until all points of lower level are already in the sequence . at stage @xmath79 , for some @xmath381 , all points in levels @xmath382 with @xmath383 are already in the sequence , and not all of level @xmath381 is in the sequence .
let @xmath384 denote the set of points in level @xmath381 not already in the sequence .
as @xmath384 is disjoint from the already constructed sequence , and no element in the already constructed sequence is larger than any point in @xmath384 , then no matter what point @xmath360 in the already constructed sequence builder chooses for the first step to compare @xmath322 with , by the second part of lemma [ median ] , the position of @xmath360 with respect to @xmath384 is at most @xmath385 . hence , there is @xmath313 $ ] such that at least @xmath386 points in @xmath384 have @xmath1th coordinate more than the @xmath1th coordinate of @xmath360 .
coordinator picks this coordinate @xmath1 to compare with , and lets @xmath387 denote the points in @xmath384 whose @xmath1th coordinate is larger than the @xmath1th coordinate of @xmath360 , so @xmath388 .
after @xmath139 steps of stage @xmath79 , we have a subset @xmath389 of level @xmath381 with @xmath390 and every point in @xmath389 satisfies the conditions that must be satisfied by @xmath322 .
notice this is satisfied for @xmath391 .
builder has two options , either continue on to step @xmath366 , or declare the turn over .
in the first case , builder picks a point @xmath370 in the already constructed sequence to compare @xmath322 with . by the second part of lemma [ median ] ,
the position of @xmath370 with respect to @xmath389 is at most @xmath392 .
hence , there is @xmath313 $ ] such that at least @xmath393 points in @xmath389 have @xmath1th coordinate more than the @xmath1th coordinate of @xmath370 .
coordinator picks this coordinate @xmath1 to compare with , and lets @xmath394 denote the points in @xmath389 whose @xmath1th coordinate is larger than the @xmath1th coordinate of @xmath370 , so @xmath395 .
this finishes step @xmath366 . in the second case
, builder declares the turn is over after the @xmath139 steps .
there are two possibilities : @xmath396 or @xmath397 .
if @xmath396 , coordinator picks @xmath322 to be the point in @xmath389 .
the number @xmath139 of steps in this case satisfies @xmath398 , so @xmath399 .
otherwise , @xmath400 , and we claim that there is still a point @xmath401^q$ ] in a level less than @xmath381 which can be chosen for @xmath322 .
indeed , if @xmath402 and @xmath403 are distinct points in @xmath389 , then the point @xmath404 with @xmath405 satisfies the desired properties .
indeed , as @xmath51 and @xmath406 are in level @xmath381 , the level of @xmath407 is @xmath408 .
further , the set of possibilities for @xmath322 is of the form that the @xmath1th coordinate is at least some @xmath409 for each @xmath1 , and , as @xmath51 and @xmath406 are in this set , then @xmath407 is also in this set .
the point @xmath407 is already in the sequence , and coordinator lets @xmath322 be this point . since @xmath322 is already in the sequence , after stage @xmath79 builder is exactly in the same situation as before stage @xmath79 in regards to winning the @xmath75-lattice game , and
so the steps in this stage did not help get closer to ending the game . in each stage
for which a new point appears in the sequence at least @xmath410 steps were used , where @xmath411 is the number of points not yet in the sequence in the level currently being used to fill the new points .
let @xmath412 denote the number of points in level @xmath381 .
we have @xmath413 if and only if @xmath414 or @xmath415 .
we thus have @xmath416 where in the second inequality we bounded a sum from below by an integral using the fact @xmath417 is an increasing function , and the last inequality follows from jensen s inequality and convexity of the function @xmath418 , where we substitute @xmath419 , and @xmath420 is the average of @xmath421 with @xmath381 ranging from @xmath422 to @xmath423 . for @xmath9
fixed , this bound can be simplified to @xmath424 which completes the proof . in this subsection
, we present the proof of theorem [ alternativeupperbound ] , which gives an upper bound on @xmath10 in terms of the online ramsey number @xmath92 .
define @xmath425 exactly as @xmath73 , except in each stage @xmath426 , builder must add at least one edge ( whose largest vertex is @xmath114 ) .
we refer to the online ramsey game in which builder must add at least one edge in each stage @xmath426 as the _ modified online ramsey game_. we have @xmath427 the lower bound being trivial .
note that in the usual online ramsey graph , those vertices @xmath114 which are not the largest vertex in an edge can only be among the first @xmath16 vertices of a monotone path .
hence , after a monochromatic monotone path of length @xmath428 is created , the last @xmath4 vertices of the path each are the largest vertex of some edge , which gives the upper bound .
we will use this upper bound in the proof below .
* proof of theorem [ alternativeupperbound ] .
* let @xmath429\choose k}\rightarrow[q]$ ] be a @xmath9-coloring of the edges of @xmath36 , where @xmath430 .
we have to show that @xmath36 contains a monochromatic monotone path of length @xmath4 .
we construct a set of vertices @xmath431 and a @xmath68-uniform hypergraph @xmath28 on these vertices with at most @xmath92 edges such that for any @xmath68-edge @xmath432 , @xmath433 , the color of any @xmath1-edge @xmath434 in @xmath36 with @xmath435 is the same , say @xmath436 .
moreover , this @xmath68-uniform hypergraph will contain a monochromatic path of length @xmath4 , which one can easily see will define a monochromatic path of length @xmath4 in @xmath36 . as described below , the way the vertices of the set are constructed
will be determined by playing the modified online ramsey game .
we begin the construction of this set of vertices by setting @xmath437 and setting @xmath438\setminus \{v_1\}$ ] . after stage @xmath64 of the process
, we have a set of vertices @xmath439 with @xmath440 and a set @xmath441 such that @xmath442 for all @xmath443 and for each @xmath68-edge @xmath444 in @xmath28 with @xmath445 , the color of the @xmath1-edge @xmath446 in @xmath36 is the same for every @xmath447 in @xmath441 and for every @xmath448 with @xmath449 . in the beginning of stage @xmath450 , we let @xmath451 be the smallest element in @xmath441 .
we play the modified online ramsey game , so that builder chooses the edges , which are @xmath68-tuples @xmath452 with @xmath453 , to be drawn according to his strategy .
painter then colors these edges . for the first edge @xmath292 chosen , painter looks at all @xmath1-tuples containing this edge and a vertex from @xmath454 .
the @xmath68-edge is colored @xmath455 $ ] in @xmath456 if there are at least @xmath457 such @xmath1-tuples that have color @xmath458 , breaking ties between colors arbitrarily .
such a color @xmath458 exists by the pigeonhole principle .
this defines a new subset @xmath459 , which are all vertices in @xmath460 such that together with edge @xmath292 form a @xmath1-tuple of color @xmath461 .
after @xmath382 edges have been drawn in this stage , we have a subset @xmath462 , and we color the next drawn edge @xmath463 color @xmath464 if the number of @xmath1-tuples of color @xmath464 containing it and a vertex from @xmath462 is at least @xmath465 , breaking ties between colors arbitrarily .
then we define @xmath466 to be the set of at least @xmath465 vertices in @xmath462 which together with @xmath463 form an edge of color @xmath464 .
after builder has added all @xmath68-edges with largest vertex @xmath451 , the remaining set will be @xmath467 .
let @xmath468 be the number of @xmath68-edges @xmath469 with @xmath470 in @xmath28 .
since we are playing the modified online ramsey game , we have @xmath471 for @xmath472 and @xmath473 for @xmath474 .
we now show by induction that @xmath475 for the base case @xmath476 , we have @xmath477 suppose we have proved the desired inequality for @xmath64 . when we draw a vertex @xmath451 , the size of our set decreases by 1 .
each time we draw an edge from @xmath451 , the size of our set @xmath336 goes down by a factor of @xmath9 .
therefore @xmath478 the number of edges drawn in the @xmath68-uniform hypergraph after @xmath64 stages is @xmath479 .
if this is at most @xmath480 , as @xmath14 and @xmath481 for all @xmath482 , we have @xmath483 since @xmath484 is an integer , @xmath485 and we can continue to the next stage .
thus , at the time we stop , @xmath486 and we have constructed a @xmath68-uniform hypergraph @xmath28 that contains a monochromatic monotone path in coloring @xmath456 on vertices @xmath487 . on the other hand ,
it follows from the properties of our construction that the vertices @xmath487 form a monochromatic monotone path of length @xmath4 in @xmath36 as required .
we close this section by proving a lower bound on the size ramsey number @xmath307 . recall that @xmath194 denotes the minimum chromatic number of an ordered graph @xmath74 which is @xmath75-path ramsey .
a graph is _ @xmath64-degenerate _ if every subgraph of it has a vertex of degree at most @xmath64 .
every @xmath64-degenerate graph has an ordering of its vertices such that each vertex is adjacent to at most @xmath64 earlier vertices .
indeed , let a vertex of degree at most @xmath64 be the last vertex in the ordering , and continue picking vertices out of degree at most @xmath64 from the remaining induced subgraph , adding them to the end of the ordering , until all vertices are picked out . from this ordering
, we get that every @xmath64-degenerate graph has chromatic number at most @xmath450 as we can color the vertices in order , picking the color of a vertex distinct the colors of its at most @xmath64 neighbors that come before it in the ordering .
let @xmath488 be integers , and let @xmath489 .
we have @xmath490 * proof .
* let @xmath74 be an ordered graph such that for every @xmath9-edge - coloring of @xmath74 there is a monochromatic monotone path of length @xmath4 .
suppose for contradiction that @xmath74 has fewer than @xmath491 edges .
partition @xmath492 , where @xmath493 is the @xmath64-core of @xmath74 with @xmath494 , which is formed by deleting , one - by - one , vertices from @xmath74 of smallest degree until the remaining induced subgraph has minimum degree at least @xmath64 .
since @xmath74 has fewer than @xmath495 edges , and the minimum degree in the subgraph of @xmath74 induced by the vertex set @xmath493 is at least @xmath64 , we have the inequality @xmath496 , and hence there is a @xmath9-coloring with color set @xmath497 $ ] of the edges inside @xmath493 with no monochromatic monotone path of length @xmath498 .
the subgraph of @xmath74 induced by the vertices in @xmath499 is @xmath500-degenerate , so it can be properly colored with @xmath494 colors .
hence , by definition of @xmath501 , there is a @xmath9-edge - coloring with color set @xmath497 $ ] of the edges inside @xmath499 without a monochromatic monotone path of length @xmath502 .
we still have to color the edges between @xmath499 and @xmath493 .
for any two vertices @xmath503 and @xmath504 , if @xmath505 , color the edge @xmath506 color @xmath316 , and if @xmath507 , color the edge @xmath506 color @xmath42 . for each @xmath508 , a monochromatic monotone path can not start with a vertex in @xmath509 ,
later have a vertex in @xmath510 , and then , even later have a vertex in @xmath509 , as otherwise it would contain an edge of color @xmath316 and an edge of color @xmath42 .
therefore , the length of the longest monochromatic monotone path is at most the sum of the lengths of the longest monochromatic monotone paths within @xmath499 and within @xmath493 , which is at most @xmath511 , contradicting our assumption that every @xmath9-edge - coloring of @xmath74 has a monochromatic monotone path of length @xmath4 .
the proof is complete .
taking @xmath512 in the previous theorem , and using @xmath11 and theorems [ ftheorem ] and [ chif ] , we have the following corollary . for any integers @xmath174
, we have @xmath108 for some constant @xmath513 only depending on @xmath9 .
this should be compared with the trivial upper bound @xmath514 thus , for fixed @xmath9 , the lower and upper bound on @xmath307 are roughly a factor @xmath4 apart .
in this short section we prove a simple lemma which is used in the proof of theorem [ convex ] .
a _ family _ @xmath515 of @xmath1-element subsets of @xmath8 $ ] ( i.e. , a @xmath1-uniform hypergraph on the vertex set @xmath8 $ ] ) is said to be _ transitive _ if for any @xmath516 such that @xmath517 we can conclude that _ all _ @xmath1-element subsets of @xmath518 belong to @xmath515 .
a _ @xmath9-coloring _ of all @xmath1-element subsets of @xmath8 $ ] is called _ transitive _ if each of its color classes is transitive . for the geometric application theorem [ convex ] , we need the following lemma relating monochromatic cliques in transitive colorings to monochromatic monotone paths in general colorings .
[ clique ] let @xmath87 be the smallest integer such that for every @xmath9-coloring of all hyperedges of @xmath36 , the complete @xmath1-uniform hypergraph on the vertex set @xmath8 $ ] , there exists a monochromatic monotone path of length @xmath4 .
then , for every transitive @xmath9-coloring of all hyperedges of @xmath36 , there exists a monochromatic complete subhypergraph of size @xmath4 .
lemma [ clique ] directly follows from the definition of @xmath10 and the following statement .
[ complete ] let @xmath519 , and let @xmath28 be a @xmath1-uniform ordered hypergraph on the vertex set @xmath185 $ ] , which contains a monotone path of length @xmath4 , that is , @xmath520 for all @xmath521 . if @xmath522 is transitive , then @xmath28 is the complete @xmath1-uniform hypergraph on @xmath185 $ ] .
we proceed by induction on @xmath4 .
the base case @xmath523 follows from the definition of transitivity .
for the inductive step , suppose that @xmath524 and that the statement has already been established for all ordered hypergraphs with fewer than @xmath4 vertices .
let @xmath28 be a hypergraph with @xmath4 vertices meeting the conditions of the lemma . by the induction hypothesis ,
the induced subhypergraph @xmath525\subseteq h$ ] , consisting of all edges of @xmath28 contained in @xmath526 is complete , and so is the induced subhypergraph @xmath527\subseteq h$ ] .
observe that this implies that for any @xmath528 , the induced hypergraph @xmath529 $ ] contains a monotone path of length @xmath242 .
as the edge set of @xmath529 $ ] is clearly transitive , we can apply the induction hypothesis to conclude that @xmath529 $ ] is a complete @xmath1-uniform hypergraph , for all @xmath79 .
therefore , @xmath28 is also a complete @xmath1-uniform hypergraph .
let @xmath530 be a family of @xmath531 noncrossing convex bodies in general position in the plane , numbered from left to right , according to the @xmath51-coordinates of their left endpoints ( leftmost points ) .
we have to show that @xmath118 contains @xmath4 members in convex position . in order to complete the proof outlined at the end of subsection [ noncrossing ] in the introduction
, we have to define a @xmath99-coloring of all triples @xmath129 with @xmath130 , which satisfies some special properties .
we need some definitions .
we say that the triple @xmath532 has a _ clockwise ( counterclockwise ) orientation _ if there exist distinct points @xmath533 such that they lie on the boundary of @xmath534 and appear there in clockwise ( counterclockwise ) order ( cf .
@xcite , @xcite ) . clearly , the orientation of a triple can be clockwise and counterclockwise at the same time
. see figure 1 .
as we numbered the members of @xmath118 according to the order of their left endpoints , for any @xmath535 , the left endpoint of @xmath536 must lie on the boundary of @xmath534 .
the triple @xmath129 is said to have a _ strong - clockwise ( strong - counterclockwise )
orientation _ if there exist points @xmath537 such that , starting at the left endpoint @xmath538 of @xmath536 , the triple @xmath539 appears in clockwise ( counterclockwise ) order along the boundary of @xmath534 .
we say that @xmath129 has _ both strong orientations _ if it has both a strong - clockwise and a strong - counterclockwise orientation .
notice that it is possible that @xmath129 has both orientations , but only one _ strong _ orientation . on the other hand ,
every triple has at least one strong orientation .
obviously , if @xmath129 has a strong - clockwise orientation , say , then it also has a clockwise orientation .
let @xmath530 be a family of noncrossing convex bodies in general position , numbered from left to right , according to the @xmath51-coordinates of their left endpoints ( leftmost points ) .
a subfamily of convex bodies @xmath540 is called _ separable _ if it has a member @xmath541 such that @xmath542 is disconnected .
we say that @xmath541 _ separates _
@xmath540 if @xmath542 is disconnected . clearly if a triple is not separable , then it only has one strong orientation . since @xmath118 is a family of noncrossing convex sets , every triple has at most one member that separates it
. we will make several observations on the triples of @xmath118 using the following notation .
for @xmath543 , let @xmath544 denote the boundary of @xmath534 , @xmath538 denote the left endpoint of @xmath536 , and let @xmath545 .
[ cknotseparate ] for @xmath543 , @xmath546 can not separate @xmath129 .
* proof . * for sake of contradiction
, suppose that @xmath546 separates the triple @xmath129 .
then @xmath547 is a simple continuous curve ( i.e. , homeomorphic to the unit interval ) that contains the left most point of @xmath544 , @xmath548 is also a simple continuous curves , and @xmath549 consists of two disjoint simple continuous curves @xmath550 and @xmath551 . since no two convex bodies have a common tangent that meets at their intersection , @xmath552 are pairwise disjoint .
moreover , @xmath553 appear in clockwise order along @xmath544 .
however since @xmath554 and since the left endpoint of @xmath555 lies inside @xmath556 to the left of the left endpoint of @xmath546 , this implies that @xmath555 and @xmath546 cross , that is , they share more than two boundary points which is a contradiction .
@xmath557 [ bothstrongcj ] for @xmath543 , the triple @xmath129 has both strong orientations if and only if @xmath555 separates it .
* proof . *
if @xmath555 separates the triple @xmath129 , then @xmath547 and @xmath549 are simple continuous curves such that @xmath554 , while @xmath548 consists of two disjoint simple continuous curves @xmath558 .
moreover since @xmath118 is in general position , @xmath559 are pairwise disjoint and appear in clockwise order along @xmath544
. therefore we can find points @xmath560 , and @xmath561 such that @xmath562 appear on @xmath544 in clockwise order .
hence @xmath129 has both strong orientations .
for the other direction , suppose that @xmath129 has both strong orientations . since the triple @xmath129 is separable , and we know that @xmath546 does not separate it , it suffices to show that @xmath536 does not separate @xmath129 .
suppose @xmath536 separates @xmath129 .
then @xmath548 and @xmath549 are simple continuous curves , and @xmath547 consists of two disjoint simple continuous curves @xmath563 and @xmath564 .
we can assume that @xmath565 .
since @xmath536 separates @xmath129 , @xmath566 are pairwise disjoint and must appear in clockwise or counterclockwise order along @xmath544 . if @xmath566 appear in clockwise order , then @xmath129 has only a strong - clockwise orientation since there does not exists points @xmath567 and @xmath568 such that @xmath569 appear in counterclockwise order along @xmath544 .
likewise if @xmath566 appear in counterclockwise order along @xmath544 , then @xmath129 has only a strong - counterclockwise orientation . in either case
, @xmath129 does not have both strong orientations and we have a contradiction .
@xmath557 suppose that @xmath536 separates @xmath129 . by the argument above @xmath129
has only one strong orientation since @xmath570 must appear in clockwise or counterclockwise order along @xmath544 .
let @xmath139 be a fixed line that goes through @xmath571 and intersects @xmath564 .
we can assume that @xmath139 is not vertical . then by the noncrossing property and by convexity of @xmath534 , regions
@xmath572 and @xmath573 must be separated by @xmath139 . therefore we make the following observation .
[ ciseparates ] for @xmath543 , suppose that @xmath536 separates the triple @xmath129 and let @xmath139 be the line described as above .
if @xmath572 lies in the upper half - plane generated by @xmath139 ( and @xmath573 lies in the lower half - plane ) , then @xmath129 has only a strong - clockwise orientation .
likewise if @xmath572 lies in the lower half - plane generated by @xmath139 ( and @xmath573 lies in the upper half - plane ) , then @xmath129 has only a strong - counterclockwise orientation . *
if @xmath572 lies in the upper half - plane generated by @xmath139 , then @xmath574 must appear in clockwise order along @xmath544 .
hence there does not exists points @xmath575 and @xmath568 such that @xmath576 appear in counterclockwise order along @xmath544 .
therefore @xmath129 does not have a strong - counterclockwise orientation , which implies that it only has a strong - clockwise orientation since every triple has at least one strong orientation . by a similar argument ,
if @xmath572 lies in the lower half - plane generated by @xmath139 , this implies that @xmath129 has only a strong - counterclockwise orientation .
@xmath557 for each pair of indices @xmath577 , let @xmath578 ( and @xmath579 ) denote the _ directed _ common tangent line of @xmath536 and @xmath555 which meets @xmath536 before @xmath555 , such that @xmath580 lies completely to the right ( left ) of @xmath578 ( @xmath579 ) .
let @xmath118 be a family of noncrossing convex sets in general position , and let @xmath544 denote the boundary of @xmath581 .
we say that the pair @xmath582 appears in _ consecutive clockwise ( counterclockwise ) order _ on @xmath544 if there exists points @xmath583 and @xmath584 such that if @xmath585 is the arc generated by moving from @xmath586 to @xmath587 along @xmath544 in the clockwise ( counterclockwise ) direction , then the interior of @xmath585 does not intersect any member of @xmath118 .
if @xmath582 appears in consecutive clockwise ( counterclockwise ) order on @xmath544 , then @xmath578 ( @xmath579 ) is the directed line going through points @xmath586 and @xmath587 ( where @xmath588 is defined as above ) such that @xmath581 lies to the right ( left ) of @xmath578 ( @xmath579 ) .
therefore we have the following . [ bothstrongvec ] for @xmath535 , if @xmath129 has both strong orientations , then @xmath589 lies to the right of @xmath578 and @xmath590 , and lies to the left of @xmath579 and @xmath591 . in other words , @xmath582 ( and @xmath592 ) appears in consecutive clockwise and counterclockwise order on the boundary of @xmath556
. * proof .
* by observation [ bothstrongcj ] , @xmath555 separates @xmath129 .
therefore @xmath547 and @xmath549 are simple continuous curves while @xmath548 consists of two disjoint simple continuous curves @xmath593 and @xmath594 .
moreover , @xmath595 are pairwise disjoint and appear in clockwise order along the boundary of @xmath534 . hence @xmath582 ( and @xmath592 ) appears in consecutive clockwise and counterclockwise order on the boundary of @xmath556 .
@xmath557 [ clockwisevec ] for @xmath535 , if @xmath129 has only a strong - clockwise orientation , then @xmath589 lies to the right of @xmath578 and to the left of @xmath596 .
in other words , @xmath582 appears in consecutive clockwise order and @xmath597 appears in consecutive counterclockwise order on the boundary of @xmath534 .
* by observation [ cknotseparate ] and [ bothstrongcj ] , either @xmath129 is not separable or @xmath536 separates it .
if @xmath129 is not separable , then @xmath598 are pairwise disjoint simple continuous curves that appear on the boundary of @xmath534 in clockwise order .
hence @xmath582 appears in consecutive clockwise order and @xmath597 appears in consecutive counterclockwise order on the boundary of @xmath534 and we are done . if @xmath536 separates @xmath129 , then @xmath548 and @xmath549 are simple continuous curves while @xmath547 consists of two disjoint simple continuous curves @xmath563 and @xmath564 .
moreover @xmath566 are pairwise disjoint and appear in clockwise order .
hence @xmath582 appears in consecutive clockwise order and @xmath597 appears in consecutive counterclockwise order on the boundary of @xmath534 @xmath557 by a symmetric argument , we have the following .
[ counterclockwisevec ] for @xmath535 , if @xmath129 has only a strong - counterclockwise orientation , then @xmath589 lies to the left of @xmath579 and to the right of @xmath599 .
[ bothorientations ] let @xmath600 .
if the triples @xmath129 and @xmath601 have both strong orientations , then the same is true for @xmath602 and @xmath603 .
* by observations [ bothstrongcj ] and [ bothstrongvec ] , both @xmath536 and @xmath604 must lie to the right of @xmath590 and to the left of @xmath591 as shown in figure 2 . .
] hence @xmath555 separates the triple @xmath602 and @xmath546 separates the triple @xmath603 , which implies @xmath602 and @xmath603 have both strong orientations .
@xmath557 [ oneorientation ] [ case ] let @xmath600 . if the triples @xmath129 and @xmath601 have only strong - clockwise orientations ( or only strong - counterclockwise orientations ) , then the same is true for @xmath602 and @xmath603
. * proof .
* by symmetry , it is sufficient to verify the statement in the case when the triples @xmath129 and @xmath601 have only strong - clockwise orientations .
[ ikl ] @xmath603 has only a strong - clockwise orientation .
notice that we can make the following assumption .
[ cols= " > , < " , ] indeed , by observation [ counterclockwisevec ] , we can assume that @xmath604 lies to the left of @xmath579 since otherwise @xmath602 would not have a strong - counterclockwise orientation and we would be done , that is , @xmath602 has only a strong - clockwise orientation since every triple has at least one strong orientation .
moreover by observation [ clockwisevec ] , @xmath604 must lie to the right of @xmath590 since @xmath601 has only a strong - clockwise orientation .
since @xmath129 has only a strong - clockwise orientation , then either @xmath129 is not separable , or @xmath536 separates it .
now the proof falls into two cases .
case 1 : assume that @xmath129 is not separable . then we define @xmath605 to be the vertical line through the left endpoint of @xmath555 , @xmath606 to be the left endpoint of @xmath555 , @xmath607 to be the first point on @xmath590 from @xmath555 in the direction of @xmath590 , @xmath608 to be the first point on @xmath579 from @xmath555 in the direction of @xmath579 , and @xmath609 to be the first point on @xmath591 from @xmath555 in the direction of @xmath591 . for simplicity
, we will assume that @xmath610 are distinct , since otherwise it would just make the remaining analysis easier .
now we make the following observation .
[ p5678v1 ] @xmath610 appear in clockwise order along the boundary of @xmath555 .
* let @xmath611 denote the boundary of @xmath612 .
then @xmath613 is the arc generated by moving along the boundary of @xmath555 from @xmath609 to @xmath607 in the clockwise direction . since @xmath606 is the leftmost point of @xmath612 , @xmath614 .
hence @xmath615 appear in clockwise order along the boundary of @xmath555 .
therefore it suffices to show that @xmath616 appear in clockwise order .
since @xmath129 is not separable and has a strong - clockwise orientation , @xmath546 does not lie completely to the left of @xmath579 .
since @xmath616 are all distinct , @xmath608 lies in the interior of @xmath612 .
thus @xmath616 appear in clockwise order along the boundary of @xmath555 and this completes the proof .
@xmath557 by @xmath617 and the ordering on @xmath118 , @xmath604 must lie to the right of @xmath605 , to the left of @xmath579 , and to the right of @xmath590 .
therefore @xmath618 must lie in one of the three regions defined as follows .
let _ region iv _ be the region enclosed by @xmath590 , @xmath605 , and the arc generated by moving from @xmath606 to @xmath607 along the boundary of @xmath555 in the clockwise direction , whose interior is disjoint from @xmath555 . let _
region v _ be the region enclosed by @xmath590 , @xmath579 , and the arc generated by moving from @xmath607 to @xmath608 along the boundary of @xmath555 in the clockwise direction , whose interior is disjoint from @xmath555 .
region vi _ be the region enclosed by @xmath605 , @xmath579 , and the arc generated by moving from @xmath608 to @xmath606 along the boundary of @xmath555 in the clockwise direction , whose interior is disjoint from @xmath555 .
see figures [ r4 t ] and [ r5 t ] .
@xmath618 can not lie in region iv * proof .
* for sake of contradiction , assume that @xmath618 does lie in region iv .
since @xmath615 appear in clockwise order along the boundary of @xmath555 , @xmath555 separates the triple @xmath601 .
moreover , @xmath618 lies in the upper half - plane generated by the line that goes through @xmath606 and @xmath607 . by observation [ ciseparates ]
, @xmath601 has only a strong - counterclockwise orientation and we have a contradiction
. see figure [ r4 ] .
@xmath557 @xmath618 can not lie in region v. * proof . * for sake of contradiction ,
assume that @xmath618 does lie in region v. since @xmath604 lies to the left of @xmath579 and since @xmath619 appear in clockwise order along the boundary of @xmath555 , we can conclude that @xmath604 lies to the left of @xmath591 .
since @xmath620 , @xmath546 separates the triple @xmath601 . by observation [ bothstrongcj ]
, @xmath601 has both strong orientations and we have a contradiction .
see figure [ r5 ] .
@xmath557 therefore @xmath618 must lie in region vi .
suppose for contradiction that @xmath536 does not intersect region vi .
then the tangent point of @xmath536 on @xmath579 lies to the left of @xmath605 .
therefore , the triangle @xmath621 whose vertices is this tangent point , @xmath606 , and @xmath608 contains region vi .
since the vertices of @xmath621 are in @xmath622 , the triangle @xmath621 is a subset of the convex hull @xmath623 .
hence , @xmath624 which is a contradiction .
therefore @xmath536 must intersect region vi and @xmath625 must lie in the lower half - plane generated by the line that goes through the left endpoint of @xmath536 and a point on @xmath626
. therefore @xmath536 separates @xmath602 and by observation [ ciseparates ] , @xmath602 has only a strong - clockwise orientation .
see figure [ r6 ] .
case 2 : assume that @xmath536 separates @xmath129 .
[ p5678v2 ] @xmath627 appear in clockwise order along the boundary of @xmath555 .
* let @xmath544 denote the boundary of @xmath628 .
since @xmath536 separates @xmath129 , the arc generated by moving along the boundary of @xmath555 from @xmath606 to @xmath607 in the clockwise direction must contain @xmath629 . since @xmath630 , @xmath627
appear in clockwise order along the boundary of @xmath555 .
@xmath557 by @xmath617 and the ordering on @xmath118 , @xmath604 must lie to the right of @xmath605 and to the right of @xmath590 .
hence @xmath618 must lie in one of the following two regions .
let _ region vii _ be the region enclosed by @xmath590 , @xmath605 , and the arc generated by moving from @xmath606 to @xmath607 along the boundary of @xmath555 in the clockwise direction , whose interior is disjoint from @xmath555 . finally let _
region viii _ be the region enclosed by @xmath605 , @xmath590 , and the arc generated by moving from @xmath607 to @xmath606 along the boundary of @xmath555 in the clockwise direction , whose interior is disjoint from @xmath555 .
see figures [ ijk2 ] and [ r789 ] .
@xmath618 can not lie in regions vii .
* proof . * for sake of contradiction ,
suppose @xmath618 lies in region vii .
then @xmath555 separates the triple @xmath601 , and @xmath618 lies in the upper half - plane generated by the line that goes through the points @xmath606 and @xmath607 . by observation [ ciseparates ] , ( @xmath631 has only a strong - counterclockwise orientation and we have a contradiction
. see figure [ r8 ] .
@xmath557 therefore @xmath618 must lie in region viii . since @xmath536 separates the triple @xmath129 , which has only a strong - clockwise orientation , @xmath536 must intersect region viii and @xmath590 . since @xmath632
, @xmath625 must lie in the lower half plane generated by the line that goes through the left endpoint of @xmath536 and a point on @xmath626 .
thus @xmath536 separates @xmath602 and by observation [ ciseparates ] , @xmath602 has only a strong - clockwise orientation .
see figure [ r9 ] .
this completes the proof of claim [ ijl ] and lemma [ oneorientation ] .
@xmath557 now we are ready to define a 3-coloring of the complete @xmath99-uniform ordered hypergraph @xmath633 formed by all triples in @xmath634 .
a triple @xmath129 , @xmath130 is colored 1 . _ red _ if it has only a strong - clockwise orientation ; 2 .
_ blue _ if it has only a strong - counterclockwise orientation ; 3 .
_ green _ if it has both strong orientations .
lemmas [ bothorientations ] and [ oneorientation ] imply that the above coloring is transitive .
since @xmath635 , the edges of @xmath633 , i.e. , the triples of @xmath118 , determine a monochromatic monotone path of length @xmath4 .
as this coloring is transitive , lemma [ complete ] implies that the members of this path induce a monochromatic _ complete _ @xmath99-uniform subhypergraph @xmath636 .
it follows from the definition of the coloring that all hyperedges of @xmath637 are triples which have the same strong orientations .
the proof of theorem [ convex ] is now a direct consequence of the following statement from @xcite and theorem [ main1 ] , which gives an upper bound on @xmath128 .
[ order ] if a family of noncrossing convex bodies in general position in the plane can be ordered in such a way that every triple has a clockwise orientation ( or every triple has a counterclockwise orientation ) , then all members of the family are in convex position .
@xmath316 . we proved estimates on the minimum number of vertices @xmath10 of a complete ordered @xmath1-uniform hypergraph for which every @xmath9-coloring of its edges has a monochromatic monotone path of length @xmath4 .
in particular , for @xmath638 and @xmath15 , we have @xmath639 it would be interesting to close the gap between the lower and upper bound . @xmath42 .
we proved that the smallest integer @xmath124 such that every family of @xmath124 noncrossing convex bodies in general position in the plane has @xmath4 members in convex position satisfies @xmath640 .
it would be interesting to improve the bound on @xmath124 further and , in particular , decide whether @xmath124 grow exponential in @xmath4 .
we proved that triples of noncrossing convex bodies with _ only _ a strong - clockwise orientation have the transitive property .
however it is easy to see that triples with a strong - clockwise orientation do not . in particular , it is possible to have four convex bodies ordered from left to right according to their left endpoints , such that @xmath641 and @xmath642 have a strong - clockwise orientation , but @xmath643 does not as shown in figure [ 134 ] . @xmath644 .
we proved that the vertex online ramsey number @xmath645 and the lattice game number @xmath319 are equal , asymptotically determined these numbers for @xmath646 , and determined the order of growth of these numbers for fixed @xmath9 .
we leave as an open problem to asymptotically determine these numbers for fixed @xmath647 . @xmath648 .
we proved that the size ramsey number @xmath307 , which is the minimum number of edges of a graph which is @xmath75-path ramsey , satisfies @xmath649 for some constants @xmath650 depending on @xmath4 .
it would be interesting to determine the growth of @xmath307 for fixed @xmath9 .
pach , j. , tth , g. : families of convex sets not representable by points , in : _ indian statistical institute platinum jubilee commemorative volume architecture and algorithms _ , world scientific , singapore , 2009 , 4353 . | for any sequence of positive integers @xmath0 , the @xmath1-tuples @xmath2 @xmath3 , are said to form a _ monotone path _ of length @xmath4 . given any integers @xmath5 and @xmath6 , what is the smallest integer @xmath7 with the property that no matter how we color all @xmath1-element subsets of @xmath8=\{1,2,\ldots , n\}$ ] with @xmath9 colors
, we can always find a monochromatic monotone path of length @xmath4 ? denoting this minimum by @xmath10 , it follows from the seminal 1935 paper of erds and szekeres that @xmath11 and @xmath12 determining the other values of these functions appears to be a difficult task .
here we show that @xmath13 for @xmath14 and @xmath15 . using a stepping - up " approach that goes back to erds and hajnal
, we prove analogous bounds on @xmath10 for larger values of @xmath1 , which are towers of height @xmath16 in @xmath17 . as a geometric application , we prove the following extension of the happy ending theorem .
every family of at least @xmath18 plane convex bodies in general position , any pair of which share at most two boundary points , has @xmath4 members in convex position , that is , it has @xmath4 members such that each of them contributes a point to the boundary of the convex hull of their union . _
dedicated to the 75th anniversary of the publication of the happy ending theorem _ |
the fluid dynamics of dense accretion discs is described by the navier - stokes and magneto - hydrodynamic equations .
within such a description , the microscopic properties of the fluid are parameterized ( in part ) via the kinematic viscosity @xmath7 , which describes the rate at which inter - particle collisions damp fluid motions , and the resistivity @xmath8 , which quantifies the dissipation of currents via ohmic losses .
timescales can be associated with each of these quantities .
the viscous timescale of a disc @xmath9 , for example , is the timescale on which the microscopic viscosity would lead to angular momentum transport over a scale @xmath10 .
as is well known @xcite , for astrophysical discs the timescales for evolution driven by the microscopic viscosity are many orders of magnitude larger than the observed evolution timescales .
it is therefore generally assumed that evolution is driven by an anomalous " or turbulent " viscosity @xmath11 @xcite , possibly augmented in some cases by angular momentum loss in a wind @xcite . in most astrophysical discs @xmath12 is likely generated from turbulence driven by the magneto - rotational instability ( mri ; see @xcite and references therein ) . since @xmath11 most studies of discs make the implicit assumption that the actual values of @xmath7 and @xmath8 are unimportant , and consider instead evolution under the action of the effective transport coefficients @xmath12 and @xmath13 .
provided that the reynolds ( @xmath14 ) and magnetic reynolds ( @xmath15 ) numbers are large enough , the absolute value of either one may indeed be of little import for accretion discs .
their ratio , however , which we define as the magnetic prandtl number , @xmath16 is almost certainly _ not _ ignorable @xcite .
an accretion disc serves to convert the large - scale energy present in the shear flow into other energy forms turbulent kinetic energy , magnetic field , and heat and the dissipation required to generate heat may vary strongly depending upon the microphysical transport coefficients . in particular , if @xmath17 the plasma is highly viscous on the scale at which resistivity operates .
in this regime one may expect that dissipation of the magnetic field will be suppressed , with a resulting modification in the saturation level of dynamo generated fields and an inverse cascade of magnetic energy to large scales @xcite .
the importance of the magnetic prandtl number for the resulting magnetic field structure has been illustarted analytically @xcite and numerically , both in idealised simulations of dynamo action within forced turbulence @xcite and in local shearing - box simulations of the mri within accretion discs @xcite dynamos is similar .
disc simulations are required , however , to assess the impact of the small - scale physics on mri saturation and the large scale flow properties .
it is unclear whether the shearing - box approach is adequate for study of these effects . ] . for discs
the current limited set of simulations indicates that the efficiency of angular momentum transport , parameterized via the shakura - sunyaev @xmath18 parameter , increases with @xmath19 , and can approach @xmath20 given the combination of modest @xmath21 and a weak net vertical magnetic field .
motivated by these results , @xcite calculated the expected radial variation of @xmath19 within standard shakura - sunyaev models of geometrically thin , radiatively efficient accretion discs .
they found that @xmath19 exceeds unity within @xmath22 schwarzschild radii of the central object , and suggested that the qualitatively different behaviour of the mri in the high @xmath19 limit might be responsible for the time - dependent outbursts and state changes that are observed in x - ray binaries . in this paper
, we compute the magnitude and radial dependence of the magnetic prandtl number in hyper - accreting discs . such discs , which can support accretion rates of the order of @xmath23 , form from compact object mergers ( e.g. * ? ? ?
* ) and when the cores of rapidly rotating massive stars collapse @xcite .
the disc structure is shaped by neutrino emission in the regions close to the accreting object , while in the outer parts neither neutrinos ( the temperature is too low ) nor photons ( which are trapped within the flow ) can provide cooling .
hyper - accreting discs are invoked to provide the initial energy injection in many models for @xmath0-ray bursts ( grbs ) .
our ultimate motivation for studying the magnetic prandtl number in these discs is to understand the complex phenomenology ( jets , highly variable prompt emission , x - ray flares , plateau phases etc ) that is observed in grbs ( e.g. * ? ? ? * ) and which may well derive from the evolution of the disc .
we anticipate different results from those found by @xcite for photon - cooled discs . in the neutrino - cooled region
the differences arise from two reasons .
first , since neutrino opacities are vastly smaller than photon opacities , hyper - accreting discs are much cooler and denser than an extrapolation of shakura - sunyaev discs would imply , and this on its own would alter the magnetic prandtl number .
second , the temperatures and densities in hyper - accreting discs are such that the electrons are mildly relativistic and mildly degenerate , while the elastic collision cross - section for the nuclei exceeds the coulomb cross - section . as a consequence ,
the classical @xcite expressions for @xmath7 and @xmath8 , which are adequate for ordinary accretion flows , no longer apply .
they do apply , instead , in the outer regions of hyper - accreting discs , where a different @xmath4 behaviour occurs since the flow is radiatively inefficient and photon - pressure dominated .
the organization of this paper is as follows . in 2 through 5
we describe the formalism for calculating the structure of hyper - accreting discs .
this is required here in order to determine consistently the plasma conditions within these discs including the temperature , density , nuclear composition and degree of degeneracy .
readers familiar with vertically averaged hyper - accreting disc models will find that our approach generally follows standard practice .
the structure of the resulting disc models is presented in 6 . given these conditions ,
we then calculate in 7 and 8 the resisitivity and viscosity in the neutrino cooled inner region of the disc .
these sections contain the principle new results of this paper . combining the viscosity and resistivity ,
we show in 10 that @xmath24 across the entire radial range relevant for grb models . in 11 we estimate the strength of other non - ideal mhd effects , and in 12 we study the structure of the outer non - radiative region of the disc . 13 and 14 discuss and
summarize our results , and what they may imply for disc evolution and grb observables .
we calculate the temperature and density at the midplane of a hyper - accreting disc from vertically averaged thin disc solutions . in hyper - accreting discs
the diffusion time for photons to leak out from the disc is much larger than the accretion time scale ( @xmath25 ) and photons remain trapped .
however , for sufficiently high accretion rates m@xmath26 sec@xmath27 for a non - rotating black hole and @xmath28 . ]
the temperature in the innermost region exceeds @xmath29 1 mev and neutrino production switches on ( popham , woosley & fryer 1999 ) .
this happens for radii r @xmath30 , where @xmath31 is the schwarzschild radius for a black hole of mass @xmath32 .
the neutrino flux liberates the energy deposited locally by viscous ( turbulent ) stresses and consequently the local sound speed @xmath33 remains smaller than the rotational velocity ( @xmath34 ) of the disc .
such discs are geometrically thin ( the vertical scale height is small compared to the radius ) , and can be described by a modified version of the standard
@xmath35disc " theory ( shakura & sunyaev 1973 ) .
many previous studies of hyper - accreting discs have adopted this formalism ( e.g. * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and references therein ) . in what follows
we consider only the innermost region ( @xmath36 ) where neutrino cooling is generally efficient . for simplicity the accretion rate
is assumed to be time - independent , but our basic results for the magnetic prandtl number would carry over locally to real discs in which the accretion rate varies with radius and time . in a steady state , conservation of mass and angular momentum for a differentially rotating fluid yield a relation between the surface density @xmath37 and the radial mass flow @xmath38 , @xmath39 , \label{eq : ang_con}\ ] ] where the fluid equations have been averaged in the vertical ( @xmath40 ) direction over the pressure scale height @xmath41 .
we assume that no viscous torque is acting at the location of the innermost stable orbit @xmath42 .
consistent with the thin disc approximation radial pressure gradients have been neglected and the orbital velocity is simply @xmath43 ( the keplerian velocity " ) .
the surface density is related to the midplane ( @xmath44 ) density via @xmath45 . the radial mass flow @xmath38 [ gr s@xmath27 ] is defined by the continuity equation = - 2 r v_r = constant , where the radial drift velocity @xmath47 is subsonic .
the turbulent kinematic viscosity is parameterized as @xcite , = c_sh .
we emphasise here that the kinematic viscosity @xmath48 is quite different in nature from the microscopic " ( also referred to as molecular " ) viscosity @xmath7 , that enters into the definition of the magnetic prandtl number .
the first is associated with turbulent flow and magnetic stresses in the plasma , while the second is given by deflection of ions and electrons by particle - particle collisions in the plasma . for a thin disk both radial advection of energy and flux in the radial direction
can typically be neglected ( we justify this for our case later ) . from energy conservation
the heating dissipated locally by viscous stresses is emitted per unit area through the surface of the disc as neutrino @xmath49 and anti - neutrino flux @xmath50 : @xmath51 the calculation of the flux in the optically thin and optically thick regions of the disc is detailed in the next section .
[ eq : ang_con ] and eq .
[ eq : ene_con ] allow us to solve for the disc structure once the equation of state is specified . in the inner regions that we are considering , densities ( @xmath52 gr @xmath53 )
are high for the temperatures of interest ( @xmath54 a few @xmath55 k ) .
this has a number of consequences .
first , electrons are mildly degenerate @xmath56 , where @xmath57 is the chemical potential in units of the electron rest mass energy @xmath58 and @xmath59 , where @xmath60 is the boltzmann constant .
second , the plasma is neutron rich . if @xmath61 is the number fraction of protons over baryons , @xmath62 .
finally , the baryon pressure @xmath63 dominates the total pressure @xmath64 ( e.g. * ? ? ?
* ) , and the isothermal sound speed @xmath65 is , c_s = 1.6 10 ^ 9 ( ) ^1/2 cm / s . at larger radii ,
on the other hand , the density is smaller , @xmath66 , and electron and photon pressure are more important ( e.g. * ? ? ? * ) .
note that in eq .
[ eq : state_b ] , and henceforth , we ignore the mass difference between protons and neutrons .
we also assume that all the baryons are free , since @xmath18 particles present at larger radii have been photo - dissociated .
this is a valid approximation since , for the parameters we use in our disc structure models , @xmath18 particles do not constitute more than 10% of the baryons ( see e.g. * ? ? ?
* thereafter cb07 ) . the neglect of advection in eq .
[ eq : ene_con ] is generally a good approximation .
the ratio of neutrino diffusion time to accretion time is @xmath67 and hence advection is important only if the optical depth to neutrinos is of the order of @xmath68 or higher . in our models
optical depths are , at most , @xmath69 a few tens , unless the accretion rate is very large , ( e.g. @xmath70sec@xmath27 , for @xmath71 ) . for a maximally rotating kerr black hole ( not treated here ) , advection can become important for lower accretion rates : @xmath72sec@xmath27 , for @xmath71 ( cb07 ) .
advection of energy in _
photons _ instead can play a role in the region we consider for @xmath73 , since the outer radiatively inefficient part of the disc can extend inward , within @xmath74 .
we will comment on this in [ results_trho ] .
at the temperatures and densities of interest for the inner regions of hyper - accreting discs neutrinos are produced primarily by electron capture onto protons @xmath75 while anti - neutrinos are produced by positron capture onto neutrons @xmath76 the inverse of these processes namely neutrino absorption by nuclei
constitute one source of neutrino opacity , others ( discussed in [ sec : opacity ] ) are neutrino scattering by both nuclei and electrons .
we note that at threshold reaction [ eq : pe ] can proceed if the electron has an energy equal to the mass difference between a neutron and a proton . in this case
the emitted neutrino has zero energy ) . ] .
the mild degeneracy of electrons suppresses the pair annihilation channel @xmath77 , since positrons are exponentially reduced with respect to electrons :
[ eq : suppress ] at equilibrium and neglecting advection , the capture rates [ eq : pe ] and [ eq : ne ] should be equal ( see also [ sec : nuc_comp ] ) .
we therefore set @xmath78 and derive only the neutrino flux .
the calculation basically follows the cb07 prescription . for @xmath79 capture onto protons ( [ eq : pe ] ) , the emissivity in units of @xmath58 is , @xmath80 ( e.g. shapiro & teukolsky 1983 ) , where @xmath81 is the electron energy in units of @xmath58 , @xmath82 , the matter density of protons is defined as @xmath83 , and @xmath84 is a constant .
the function @xmath85 is the fermi - dirac distribution , @xmath86 + 1}. \label{eq : fermi}\ ] ] the energy density in units of @xmath58 for an optically thin disc can be expressed in terms of the emissivity as @xmath87 if , conversely , the optical depth @xmath88 , neutrinos thermalize to a fermi - dirac distribution ( eq . [ eq : fermi ] ) and the energy density in units of @xmath58 is given by , @xmath89 for @xmath90 , @xmath91 . to deal with regions of the disc that have @xmath92 we interpolate between the optically thin and thick limits .
we define x = [ eq : x ] and express the neutrino flux for arbitrary optical thickness as = \ { ll u_,n c(1+_)^-1 & , + u_,c c(1+_)^-1 & . .
[ eq : flux ]
in the mid - plane of the disc pairs , baryons ( free protons and neutrons only in our case ) and photons are in thermodynamic equilibrium . for given ( @xmath93 , @xmath94 ) , and specified optical depth
, this fact suffices to determine the nuclear composition of the plasma . in practice
we compute the composition in the optically thick and optically thin limits separately , and deal with the intermediate regime by making an interpolation that is consistent with charge conservation .
let us first note some general considerations .
in addition to the processes of electron / positron capture onto nuclei ( equations [ eq : pe ] and [ eq : ne ] ) , pairs are created and absorbed via e^++e^- + .
this implies that = _ - = -_+ .
baryons , instead , are created and destroyed only according to the processes given by equations [ eq : pe ] and [ eq : ne ] .
therefore , at equilibrium ( @xmath95 ) , and neglecting advection , the emissivity for neutrinos and anti - neutrinos must everywhere be equal . in regions of the disc that are very optically _ thick _ to neutrinos , they will have relaxed into a fermi - dirac distribution and the above condition can be expressed as , n_=n _ , [ eq : eqthick ] where the number densities are calculated by integrating eq .
[ eq : fermi ] over momenta with the statistical weight set to unity .
the equality of the neutrino and anti - neutrino number densities ( eq . [ eq : eqthick ] ) implies a vanishing neutrino chemical potential , _ = _
[ eq : munu ] the chemical balance of reactions [ eq : pe ] and [ eq : ne ] ( and their inverses ) @xmath96 gives = + q , [ eq : muthick ] @xcite where protons and neutrons ( with number density @xmath97 , @xmath98 respectively ) have a maxwellian distribution . for regions of the disk that are optically _
, we use @xmath99 where the number of neutrinos @xmath7 and anti - neutrinos @xmath100 emitted per unit volume per unit time is @xmath101 and @xmath102 ( e.g. shapiro & teukolsky 1983 ) . in both regimes
, we impose the conditions of charge neutrality , @xmath103 where @xmath104 and baryon conservation , @xmath105 the electron @xmath106 and positron @xmath107 number densities are obtained by directly integrating eq .
[ eq : fermi ] over momenta with the statistical weight set to two .
to summarise , we use two different sets of equations in the two extreme regimes : when @xmath108 , we find @xmath109 and @xmath110 from eqs .
[ eq : muthick ] , [ eq : cc ] and [ eq : bc ] ; when @xmath111 , we find @xmath112 and @xmath113 from eqs .
[ eq : ndotnu_ndotbnu ] , [ eq : cc ] and [ eq : bc ] .
matter in the disc plane , of course , can have an arbitrary optically thickness . to compute the nuclear composition for specified @xmath93 and @xmath94 we first interpolate to find the chemical potential using , = ( 1-x ) + x , [ eq : mutot ] where @xmath114 is given by eq .
[ eq : x ] .
we then find the proton fraction @xmath115 by imposing charge conservation ( eq . [ eq : cc ] ) . the resulting @xmath115 is @xmath116 .
physically , the most important trends in the nuclear composition can be understood in terms of the ratio of the temperature to the ( relativistic ) degeneracy temperature t_deg 0.71 ( ) ^1/3 k , [ eq : tdeg ] where we set @xmath117 .
if @xmath118 ( at low temperature , or at high density ) then @xmath57 increases , positrons are suppressed ( eq . [ eq : suppress ] ) , and the resulting plasma is neutron - rich ( low @xmath115 ) . in the opposite limit of @xmath119 , @xmath120 but @xmath121
because , as noted before , reaction [ eq : ne ] is energetically favoured .
the vertical optical thickness of the disk to @xmath7 and @xmath100 in eq .
[ eq : flux ] _ = h , [ eq : tau ] is calculated considering three sources of opacity : @xmath122 [ @xmath123 gr@xmath27 ] .
first , absorption onto nuclei , _ a = ( _ , n(1-)+_,p ) where @xmath124 is the cross section for @xmath7 ( @xmath100 ) absorption by neutrons ( protons ) .
second , elastic scattering by electrons , _ s , e = 0.5 ( _ e,+_e , ) , where @xmath125 and @xmath126 is the electron - neutrino ( anti - neutrino ) cross section .
we neglect scattering onto positrons since @xmath127 .
finally , elastic scattering by nuclei , _ s = ( 0.3 + 0.36 ( 1- ) ) |e_^2 , where @xmath128 @xmath129 . for the explicit forms of the cross sections , we refer the reader to the appendix b of cb07 .
the cross sections for these processes depend on the neutrino energy , and the total cross section @xmath130 should be derived by integrating over the neutrino spectrum .
this requires knowledge of the neutrino distribution function at arbitrary optical depth .
we , instead , use approximate mean cross sections , which are adequate for our purposes .
we first evaluate @xmath130 at a mean neutrino energy ( in unit of @xmath131 ) : @xmath132 then , the usual interpolation allows us to link the optically thin and thick regimes : ( cb07 ) where corresponding to @xmath111 and corresponding to @xmath108 is the mean energy calculated from a fermi - dirac distribution function with zero chemical potential .
we use the same treatment for anti - neutrino cross sections . in practice
, we find that for conditions appropriate to our disc models absorption by nuclei is always the dominant opacity by one or two orders of magnitude .
electron scattering is least important .
to give an idea of the midplane temperatures and densities encountered in hyper - accreting discs , we show in figure [ fig : t_r ] and figure [ fig : rho_r ] illustrative examples of the radial profiles of these quantities .
the disc models plotted were computed for a black hole of @xmath133m@xmath26 , with @xmath28 and accretion rates of @xmath134 , @xmath135 and @xmath136m@xmath26sec@xmath27 . for these parameters
the disc is optically thin across the entire radial range shown for the lowest value of the accretion rate .
for the two high accretion rate models the disc is optically thick within about @xmath137 of the black hole .
the most important feature of the neutrino cooled solutions for our purposes is the relatively weak radial dependence of the temperature .
as noted elsewhere ( e.g. cb07 ) , the disc self - adjusts to maintain electrons in a state of mild degeneracy ( i.e. @xmath138 ) : higher degeneracy would cause the cooling to decrease drastically and the temperature to rise , lifting the degeneracy , and vice versa for lower degeneracy .
this thermostatic aspect of neutrino cooled discs has as a consequence that temperature does not vary strongly with radius , since the degeneracy temperature depends only weakly on density ( and in addition @xmath115 decreases with @xmath93 ) .
the proton fraction @xmath115 is everywhere less than @xmath139 , falling to around @xmath140 in the innermost regions .
we have verified that the physical conditions in our disc models reproduce those of prior authors .
comparing our results to those of cb07 , we find excellent agreement especially for @xmath141 between our models with a given @xmath18 and their model with @xmath142 ( the factor two arises primarily due to different definitions of viscosity and different inner boundary conditions , both of which are largely arbitrary ) .
for @xmath143 , their densities are lower and the advection dominated outer region ( which is not cooled efficiently by neutrinos ) extends to smaller radii .
since we ignore the advective term in the energy equation , our profiles have a different slope but they never differ by more than a factor of 2 in density and by 50% in temperature .
the computation of the resistivity of a plasma in which electrons are both relativistic and degenerate is quite subtle , and ultimately requires numerical methods .
we first review some general considerations , including the classical spitzer ( 1962 ) result , which we emphasise is _ not _ applicable to hyper - accreting discs .
these results will prove useful later when we discuss the relative importance of different non - ideal mhd effects .
the electrical resistivity of a plasma is a measure of the electron mobility in the presence of an applied electric field .
acceleration due to the electric field ( @xmath144 , where @xmath81 is now the electric charge ) is opposed by drag from ( primarily ) electron - proton collisions .
balancing these competing forces on a electron of velocity @xmath145 and momentum @xmath146 , @xmath147 where we neglect pressure gradients and magnetic fields and assume stationary protons .
ohm s law then gives , = , [ eq : eta ] ( see e.g. jackson 1963 p. 460
, for the non - relativistic derivation ) . in this expression
@xmath148 is strictly equal to @xmath149 : the number density of electrons with momentum @xmath150 , @xmath151 is the electron lorentz factor , @xmath152 is the classical electron radius and @xmath153 is the proton - electron deflection time .
the associated electrical conductivity is @xmath154 . extending the non - relativistic definition ( that can be found for example in longair 1992 , p. 304 )
, @xmath153 can be defined so that @xmath155 , where @xmath156 is the mean square impulse received by an electron per scattering perpendicular to its direction of motion .
we obtain t_ep = , [ eq : tep ] where @xmath157 is the coulomb logarithm , defined as the logarithm of the maximum to the minimum impact parameter for proton - electron collisions .
if we take the non - relativistic limit ( @xmath158 ) of eqs . [ eq : eta ] and [ eq : tep ] and insert the mean thermal velocity of non - degenerate electrons @xmath159 , we recover the scalings between the spitzer resistivity and our derivation . ]
derived by spitzer ( 1962 ) for a completely ionized plasma with @xmath160 _ s = 1.9 10 ^ 12 _ ep , st^-3/2 , [ eq : etas ] where the coulomb logarithm for electron - proton collision is : _ ep , s = 1.2410 ^ 4 .
[ eq : cou_log_ep_s ] in this non - degenerate regime , the ( inverse ) dependency of @xmath161 on density is weak . to compute the plasma resistivity under our conditions of temperature and density , we use the code by @xcite and @xcite , which is able to treat arbitrary degeneracy of fully relativistic electrons ( but not positrons ) .
the calculation involves an integration of eq .
[ eq : eta ] over energy , weighted by the fermi - dirac distribution ( eq . [ eq : fermi ] ) .
the coulomb logarithm is calculated via an effective scattering potential that takes into account debye charge screening ( see reference above for details ) .
we use as input parameters to the code the temperature @xmath94 and the proton density @xmath162 , which , in turn , depends ( via @xmath115 ) on @xmath94 and @xmath93 . from @xmath162
the code derives the electron density and thus the electron chemical potential which uniquely specifies the electron fermi - dirac distribution . in principle
the calculation of the total resistivity should include the contribution of positrons both as charge carriers ( they contribute to the current density and have their own resistivity ) and as deflectors for electrons .
however , we have already noted that positrons are exponentially suppressed with respect to electrons and from charge conservation it follows that they can also be neglected in the drag force which remains dominated by protons .
the code allow us to explore the plasma resistivity under different condition of electron degeneracy .
in the non - degenerate limit the resistivity _ decreases _ with temperature , because scattering becomes less frequent , while the density dependence is weak and confined to the coulomb logarithm ( e.g. spitzer resistivity , eq .
[ eq : etas ] , for @xmath163 ) . in the highly - degenerate regime , instead , the resistivity _ increases _ with temperature , because more energy states become available for the electrons to be scattered into .
the density dependence here is strong .
higher density ( for fixed temperature ) results in higher degeneracy and suppression of scattering : the plasma resistivity thus decreases .
the intermediate mildly degenerate regime , where the resistivity changes very slowly with temperature and density , is the one occupied by our plasma .
the calculated resistivity is shown as a function of temperature in figure [ fig : eta_nu_t ] . for a density of @xmath164 the degeneracy temperature @xmath165 ( evaluated at @xmath166 )
is approximately @xmath167 k. for temperatures above this value , the proton fraction remains approximately constant at a value of @xmath168 , thus @xmath162 does not change appreciably and the resistivity decreases with temperature since electron - proton collisions becomes less frequent . for lower temperatures ,
one s naive expectation is that the resistivity ought to flatten out and eventually decrease as electrons become more and more degenerate .
this behaviour is seen _ if we artificially impose _ a temperature independent electron density ( dashed line ) , but it does not reflect the physical behaviour of the disc plasma .
in fact , as the temperature drops toward and below @xmath165 the plasma becomes more and more neutron - rich ( i.e. @xmath115 decreases ) , with an attendant decrease in the electron density .
the effective degeneracy temperature decreases , and the electrons never become highly degenerate .
the overall result is that at low " temperatures the electrons self - adjust toward a mildly degenerate regime in which @xmath8 rises slowly with decreasing temperature .
the dependence on density is weak . within a hyper - accreting disc ,
the midplane temperature is a fairly weak function of radius ( figure [ fig : t_r ] ) .
this , when coupled with the weak dependence of the resistivity on density , results in a radial profile of @xmath8 , shown in figure [ fig : eta_nu_r ] , that increases slowly with distance from the black hole . between 5 and @xmath74
the variation in the resistivity is less than an order of magnitude .
there is a modest _ decrease _ with increasing accretion rate .
contributions to the viscosity of the plasma come from three sources , proton - proton coulomb collisions , elastic baryonic scattering due to the strong force , and electron viscosity .
it is not immediately obvious which of these processes dominates for a mildly relativistic and degenerate neutron - rich plasma .
we first consider plasma viscosity due to baryon collisions . typically ( but not in our case ) only ion - ion coulomb collisions need to be taken into account .
protons are non - degenerate and non - relativistic , so we use the non - relativistic limit of eq .
[ eq : tep ] , with @xmath169 where the mean proton relative velocity @xmath170 is written in terms of the mean thermal velocity @xmath171 .
multiplying by @xmath172 we obtain the proton - proton ( p - p ) deflection timescale , t_pp , s = 18.03 , [ eq : tep_s ] where the p - p coulomb logarithm is given by _ pp , s = 1.2410 ^ 4 [ eq : cou_log ] @xcite from the classic spitzer result . ] . scattering in this regime
is dominated by small - angle deflections .
once plasma temperatures exceed @xmath29 mev , however , as in hyper - accreting discs , the elastic nuclear cross section becomes larger than the coulomb one .
the shape of the nuclear two - body cross section can be derived theoretically with a partial - wave analysis ( considering only s - waves ) within the frame - work of the effective - range theory @xcite . for the particle energies of interest to us , the _ partial _ cross section ( for a _ single _ total spin state )
depends only on two parameters : the scattering length @xmath173 and the effective range of the nuclear potential @xmath174 , _
0,i = , where the wave number is @xmath175 , where @xmath176 is the centre - of - mass total kinetic energy . for the p - n collision
we need to consider contributions from both singlet @xmath177 ( antiparallel spins ) and triplet @xmath178 ( parallel spins ) states , and the total cross section is the weighted sum of partial siglet and triplet cross sections , _ pn = _ 0,t+ _ 0,s , with @xmath179 cm , @xmath180 cm , @xmath181 cm and @xmath182 cm @xcite . for identical particles ,
however , the triplet state is excluded .
the p - p and n - n cross section are thus , _
pp = _ 0,s , with @xmath183 cm and @xmath184 cm ; @xmath185 , with @xmath186 cm and @xmath187 cm @xcite .
the result is that @xmath188 .
once we allow for the possibility of strong interactions the total collision timescale for protons becomes = + + , where the nuclear p - p and p - n timescales are @xmath189 and @xmath190 respectively .
the collision timescale for a neutron is instead = + , where @xmath191 and @xmath192 .
note that for both protons and neutrons the collision timescale is dominated by encounters with _
neutrons _ , since the flow is neutron rich and the high temperature inhibits electromagnetic collisions . the corresponding mean free paths are @xmath193 and @xmath194 . beside baryons , we should consider the contribution to viscosity from electrons , since degeneracy results in a longer mean free path @xmath195 ( @xmath196 is the ratio of electron velocity to @xmath197 ) and a higher momentum density than non - degenerate electrons . according to the classical kinetic theory the dynamical viscosity @xmath198}$ ] for a single species
i " is proportional to @xmath199 .
therefore , we can write _ = c_1 ( _ p|v_p + _ n|v_n+p _ e ) where @xmath200 is a constant of the order of unity is the same for the collision processes considered here . ] .
the typical electron kinetic energy is @xmath201 , where @xmath57 is the chemical potential , thus its momentum is @xmath202 .
the total kinematic viscosity @xmath203 [ @xmath129s@xmath27 ] can be written as = _ s ( + ( 1- ) + ) [ eq : nu ] where @xmath204 and @xmath205 ( where @xmath206 is calculated for @xmath207 ) is the dynamical viscosity for a fluid of completely ionized hydrogen , _ s = 2.2110 ^ -15 , [ eq : nuspitzer ] ( braginskii 1957 , 1958 ; spitzer 1962 ) . in eq .
[ eq : nu ] , the first term on the right is the proton viscosity @xmath208 , the second is the neutron viscosity @xmath208 and the third is the electron viscosity @xmath209 .
this last reduces formally to the formula for strongly degenerate electrons @xmath210 @xcite ( with @xmath211 ) . in fig .
[ fig : nu_t_comp ] , we plot @xmath7 and its components as a fucntion of temperature . in the non - degenerate limit
( @xmath212 ) , protons and neutrons contribute equally to the total viscosity : they carry the same momentum density , transported in both cases by n - p collisions . in this regime ,
the electron contribution @xmath213 has been calculated as @xmath214 , where @xmath215 is the electron fraction and @xmath216 ( eq .
[ eq : tep ] with @xmath217 and @xmath218 ) .
as the temperature decreases , the flow becomes neutron rich and the electrons degenerate .
as this happens the proton momentum density drops faster than the neutron s one , strong collisions with neutrons dominate the baryon collision frequency and the electron mean free path increases ( while the baryon one keeps decreasing ) . as a consequence ,
electron viscosity becomes comparable or greater than the baryon viscosity .
since the electron density decreases with temperature , @xmath209 eventually decreases ( even if @xmath219 keeps increasing ) and it does not attain the limiting viscosity value for strongly degenerate electrons ( shown in the figure as the horizontal dashed line ) . therefore the total viscosity @xmath7 _ decreases _ as the plasma cools even when electron viscosity dominates .
the behaviour of @xmath7 as a function of density is shown in figure [ fig : eta_nu_t ] .
it decreases as the plasma becomes denser , since p - n collisions become more frequent and the electron ( and thus their momentum ) density decreases when below the nominal degeneracy temperature . in these figures
, the @xmath220 regime represents the typical situation in hyper - accreting discs .
figure [ fig : eta_nu_r ] shows the predicted radial profile of the viscosity within the hyper - accreting disc models . in the optically thin regions of the disc
the viscosity increases with radius , since the temperature in this region decreases much more slowly with radius than the density .
electron viscosity is higher than baryon viscosity but the two contributions tend to converge ( and they can become comparable for low accretion rates ) in the less dense regions away from the black hole where the electrons are less degenerate .
the trend is the opposite in the optically thick innermost region , which is present only at high accretion rates . in this case
the viscosity decreases moving outward . here
, the neutron contribution is highly suppressed ( @xmath29 one - two orders of magnitude ) and electrons completely dominate the flow viscosity .
photons and neutrinos can also in principle contribute to viscosity .
in the case of neutrinos , the mean free path is of course enormous compared to that of the baryons , so momentum transport by neutrinos will dominate _ if it acts as a viscosity_. this will depend upon the neutrino optical depth . in regions of the disc that are very optically thick neutrino viscosity will act as a ( very large ) navier - stokes viscosity .
indeed , @xcite have suggested that neutrinos might be able to shut off the mri completely , by reducing the reynolds number below that which admits linear growth of mri modes . in optically thin regions , on the other hand , momentum transport by neutrinos is non - local and acts as a radiation drag rather than a viscosity @xcite . in this paper
we are principally concerned either with optically thin regions of the disc ( where neutrinos do not contribute viscosity ) , or with regions of modest optical depth ( @xmath221 a few ) for which the correct treatment is unclear .
the reader should , however , bear in mind that the presence of neutrinos may have dynamical consequences in the optically thick regions of the flow .
since we derive high values of the prandtl number while ignoring the neutrino contribution , to the extent that neutrinos contribute to viscosity , they will only reinforce our conclusions on the large pm regime of hyper - accreting discs .
our magnetic prandtl numbers are conservative in that sense .
the discussion on the different regimes according to the optical thickness also applies to photon viscosity .
our discs are everywhere in the mid - plane extremely optically thick to photons .
this implies a much smaller mean free path for photons than for neutrinos and their contribution to viscosity is relatively suppressed by a factor equal to the ratio of mean free paths for the two species .
photon viscosity can also be neglected with respect to baryon viscosity , even if they have similar mean free paths , since the photon momentum density is small compared to the baryons one .
identical methods can be used to compute the resistivity and viscosity for photon - cooled discs , which are present in x - ray binaries at similar radii around black holes of a few solar masses . in figure
[ fig : eta_nu_rb ] we plot the radial profiles of these quantities for a model presented by balbus & henri ( 2007 ) as representative of a x - ray binary disc .
the model has a central black hole of mass @xmath222 , surrounded by a disc accreting at one tenth the eddington limit with @xmath223 .
we treat the photon flux in the diffusion approximation , adopt an opacity that is of kramers form plus an electron scattering correction ( @xmath224 ) , and write the total pressure as the sum of the gas pressure ( baryons and electrons ) and photon pressure .
the abundances are taken as 10% @xmath18 particles and 90% hydrogen by number .
the gas is completely ionised . in the x - ray binary regime
the spitzer resistivity ( eq . [ eq : etas ] ) is valid ( we neglect @xmath225 collisions ) , while the viscosity is given by p - p coulomb collisions and @xmath18-p coulomb collisions ( see appendix a in * ? ? ?
* ) . in the x - ray binary disc model
the plasma is cooler and less dense , and electrons are non - degenerate . these characteristics
enhance the e - p collision probability , resulting in a resistivity that is several orders of magnitude higher than in neutrino cooled discs .
the viscosity is also much higher , and moreover , it decreases towards larger radii .
this is the opposite trend to that seen for the viscosity in neutrino cooled discs .
with the results for the resistivity and viscosity in hand we can compute the magnetic prandtl number , defined as the ratio of the microscopic viscosity to the resistivity , = .
[ eq : pr ] results for the three disc models that we have discussed in detail are shown in figure [ fig : pr_r ] . the radial behaviour of @xmath4 is mainly determined by the viscosity , since the resistivity is a rather weak function of radius . in optically thin regions of the disc ( everywhere for @xmath226 , and at large radii , @xmath227 @xmath228 , for the higher accretion rate models )
the @xmath4 number is relatively flat but eventually increases with radius because the rapidly decreasing density increases the microscopic viscosity in the disc . in the optically thick regions , the temperature increases more rapidly toward the black hole than does the density , and the increased viscosity leads to larger @xmath4 in the central parts .
we note that , although the values of both the resistivity and the viscosity are far larger in the photon cooled x - ray binary disc model , the value of the magnetic prandtl number is not all that different in these inner parts .
more significant is the different radial behaviour . for hyper - accreting discs @xmath4 increases with radius once
the disc becomes optically thin , and this trend continues throughout the neutrino cooled zone . for x - ray
binary discs @xmath4 falls off with radius and eventually becomes smaller than one ( at @xmath229 r@xmath230 ) .
our results for this model are qualitatively similar to those of @xcite ( see their fig . 1 , left panel ) with the quantitative differences arising due to different boundary conditions and value of @xmath161 .
figures [ fig : pr_0.01 ] , [ fig : pr_0.1 ] and [ fig : pr_1 ] show the dependence of @xmath4 on radius and accretion rate for three different values of @xmath18 , @xmath231 .
we find that @xmath232 throughout the neutrino cooled region for all accretion rates in the range @xmath233 .
typically @xmath234 ( except in the @xmath235 regions for @xmath236 ) , with values as high as a few @xmath68 if @xmath237 .
as @xmath18 decreases , the disc density is higher and the viscosity drops : the lower viscosity results in a smaller magnetic prandtl number .
this is unlike x - ray binary discs , where the @xmath18 dependence of the microscopic viscosity is largely offset by a similar change in the resistivity @xcite .
all the results discussed here are computed assuming midplane disc conditions . moving toward the upper layers of the disc the density invariably drops more rapidly than the temperature , and as a consequence
the electrons become non - degenerate . however , verticle mixing is efficient and the nuclear composition ( @xmath115 ) at higher latitude is close to that at the midplane @xcite . under these conditions
the resistivity increases more slowly than the viscosity ( whose main contribution will be eventually come from neutrons ) and @xmath4 increases further .
our motivation for computing the magnetic prandtl number derives from various suggestions that the saturation level of the mri varies strongly with @xmath4 .
almost all numerical simulations addressing this issue include ohmic dissipation as the only non - ideal term in the mhd equations .
if other non - ideal terms are dominant in hyper - accreting discs there will be ( at the very least ) additional uncertainties in interpreting the results of simulations for astrophysical discs regime in the absence of the hall effect , while simulations that have included the hall term have focused on the regime where the magnetic reynolds number is low ( as is appropriate for protoplanetary discs ) .
although the hall term should not directly alter the dissipative properties of the fluid , we can not rule out that the behavior of a high prandl number disc with a strong hall effect might differ from the pure @xmath234 case . ] . here
, we estimate the magnitude of the different non - ideal terms for fiducial values of temperature , @xmath238 k , and density , @xmath239 gr @xmath53 . under these conditions
the plasma composition is @xmath240 , the chemical potential is @xmath241 and the sound speed @xmath242 cm s@xmath27 .
the equipartition magnetic field , b_eq = is @xmath243 gauss . the low resistivity of plasma in neutrino cooled discs results in a very large magnetic reynolds number . if we take @xmath244 cm ( @xmath245 ) , then r_em= 1.5 10 ^ 19 , [ eq : reynum ] where we have used the consistent value of the resistivity @xmath246
@xmath129 s@xmath27 .
_ all _ non - ideal mhd effects are therefore small . in relative terms
, however , the high density means that the plasma is sufficiently collisional that , for very weak magnetic fields , ohmic dissipation is stronger than both the hall effect ( the separation of charge carriers in the presence of a magnetic field perpendicular to the current ) and ambipolar diffusion ( the differential motion of neutrons and charged particles ) .
the dominance of ohmic dissipation persists until the magnetic field becomes strong enough that the cyclotron frequencies of the charged particles exceed the collision frequencies .
we can write the relative strength of the ohmic ( o ) and hall ( ha ) effects as ( e.g. * ? ? ?
[ eq : ho ] where @xmath247 is the electron gyro - frequency and @xmath153 the e - p deflection timescale . for our conditions
t_ep = 2.87 10 ^ -18 sec , [ eq : tep_2 ] where the electron density has been evaluated as @xmath248 , and _
ce = hz , where the mean lorentz factor is @xmath249 .
the field b here is the mean local field .
. [ eq : ho ] then gives , numerically , = 5.2 10 ^ -12 .
likewise , the ratio of ambipolar ( a ) to hall ( h ) strength can be written as ( e.g. * ? ? ?
* ) , = _ cp t_pn(2 - 2 ) = 9.4 10 ^ -17 , [ eq : ah ] where @xmath250 is the proton gyro - frequency and the p - n collision time - scale is @xmath251 sec .
we note again that the coulomb coupling ( @xmath153 ) is weaker than the strong collision coupling for our plasma conditions .
eq . [ eq : ho ] and eq .
[ eq : ah ] show that ohmic dissipation is the dominant non - ideal mhd term only in weakly magnetised plasmas which have a ratio of thermal to magnetic energy @xmath252 .
this means that , for any reasonable magnetic field in the disc , the hall effect is the dominant non - ideal mhd effect . for our fiducial parameters
the hall effect becomes the most important non - ideal mhd effect for @xmath253 g , @xmath153 and @xmath8 are formally the relaxation timescale and the resistivity along the magnetic field lines .
the transverse currents have a collision timescale reduced by @xmath254 . ] . at equipartition
ambipolar diffusion is also more important than resistive diffusion but we remain in the hall regime .
in the outer disc beyond @xmath29 one hundred @xmath228 the physical conditions are quite different . in this region
the disc is photon - pressure dominated .
the temperature ( @xmath255 ) is below the neutrino cooling threshold and the flow is effectively non - radiative .
the free baryons are in the form of @xmath256 nuclei @xcite and the density ( @xmath257 ) is low enough that the electrons are non - degenerate ( and non relativistic ) . the disc is thus both more resistive and more viscous , and the resistivity and viscosity have the classical spitzer dependencies on temperature and density ( eq . [ eq : etas ] and eq . [ eq : nuspitzer ] ) but with numerical coefficients ( also in the `` spitzer '' coulomb logarithms eqs .
[ eq : cou_log_ep_s ] and [ eq : cou_log ] ) adjusted to be appropriate for a completely ionised he flow .
it is worth emphasizing that unlike the radiatively inefficient flows discussed in the x - ray binary or agn contexts ( e.g * ? ? ?
* ; * ? ? ?
* ) , those in hyper - accreting scenarios would remain fully collisional .
this constitutes a rare situation in which the collisionality guarantees a unique temperature for electrons and nuclei
. moreover mhd dissipation scale arguments based on classical spitzer values are adequate .
the magnetic prandtl number in this region is , = 7.3 10 ^ -28 , [ eq : pr_adaf ] where @xmath258 is the product of the coulomb logarithm for @xmath259 and for @xmath260 collisions . if we assume that the temperature is set by a fraction of the virial energy density then @xmath261 .
numerically we find that @xmath262 and ( ) , [ eq : pr_adaf ] where we take @xmath263 ; however , the dependence of @xmath4 on @xmath264 is small . ] .
@xmath265 varies very slowly with radius and eq .
[ eq : pr_adaf ] with @xmath266 is actually a very good approximation for @xmath4 in the whole outer disc .
we note that @xmath234 and it falls off slowly as @xmath267 : for @xmath268 , @xmath269 .
this result indicates that in the outer regions of hyper - accreting discs @xmath4 has a very different behaviour than in the innermost parts . here
@xmath4 is rather insensitive to the actual accreting conditions @xmath38 and @xmath18 ( as long as the outer disc remains photon - pressure dominated ) , but it is a _
stronger _ function of temperature and it decreases with radius .
however , both regions have in common @xmath232 : unlike x - ray binary and agn discs , hyper - accreting discs seem not to have a @xmath270 region . beside a high @xmath4 ,
this outer disc flow has other characteristics in common with the neutrino - cooled region .
the conductivity is high with @xmath271 and the hall effect dominates even for modest magnetic fields ( @xmath272 ) .
these features are caused by the still high temperature that makes coulomb coupling rather inefficient ( though in the outer region it is still more efficient than nuclear coupling ) .
in this section , we attempt to relate the plasma properties that we have computed to the global properties of discs .
there are many steps required to connect the microscopic scales where resistivity and viscosity act to the global evolution of the disc , and simulations and analytic arguments provide only an imperfect guide
. however , plausible guesses are possible .
we focus on the possible consequences of magnetic prandtl number greater than unity for the magnetisation and evolution of hyper - accreting discs .
let us first summarise the argument that suggests that the magnetic prandtl number matters . in turbulent plasmas with @xmath273 ,
the absence of turbulence below the viscous scale @xmath274 means that the resistive scale may not be easily accessible to field lines and reconnection could be slow with respect to magnetic amplification .
we can understand the dependence on @xmath4 with the following simple argument .
turbulence at scale @xmath275 exponentially amplifies the magnetic field on an eddy turnover time - scale @xmath276 , while the magnetic field dissipation on that same scale takes @xmath277 .
the ratio of these timescales is .
so , for @xmath273 , magnetic energy can build up , first on the viscous scale and subsequently in an upward cascade toward larger scales @xcite .
this may have important consequences for accretion discs . within discs ,
the turbulent growth of magnetic fields can be limited in principle by either resistive dissipation ( on small scales ) or buoyant expulsion of the flux from the disc .
numerical simulations suggest that in low @xmath4 discs resistive dissipation ultimately sets the saturation level ( e.g. * ? ? ?
if , however , the small - scale dissipation of magnetic fields is frustrated by a high magnetic prandtl number , the only channel available to limit the flux may be macroscopic buoyancy . since buoyancy is relatively inefficient the timescale is likely to modestly exceed the alfvn crossing time @xmath278 of the disc a high magnetic prandtl number disc could saturate at a level that involves stronger magnetic fields , more vigorous turbulence and angular momentum transport , and ongoing field expulsion .
these expectations are supported by the available numerical evidence . @xcite and @xcite
find that the parameter @xmath18 increases with the magnetic prandtl number in the range @xmath279 and @xmath280 . in the case
where @xmath18 approaches unity , @xcite find a highly time dependent disc , with large fluctuations in @xmath18 and the mean magnetic pressure .
this appears to be a consequence of the well - known dependence of the strength of the mri on the magnetic field : as the ratio @xmath196 of thermal to magnetic pressure decreases @xmath18 at first increases until the mri is quenched around equipartition @xcite .
based on these arguments , two evolutionary consequences can be contemplated for hyper - accreting discs formed after stellar core collapse or compact object mergers . in the initial phase ,
a high magnetic prandtl number in the neutrino cooled inner disc results in stronger magnetic field and a higher efficiency of angular momentum transport .
since , as we have shown , a _ higher _ @xmath18 in such discs results in a yet _ higher _ @xmath4 , a runaway occurs until @xmath20 .
this regime is sustainable for many dynamical timescales , provided that the excess magnetic energy that could quench the mri is removed from the disc , for example , by buoyancy .
this is plausible since close to equipartition the buoyancy timescale @xmath281 is similar to the orbital timescale and can provide cooling by removing magnetic energy in which a fixed fraction of the cooling occurs non - radiatively via loss of magnetic energy .
we found that @xmath4 is only reduced by a factor of @xmath282 with respect to a disc with @xmath283 in which only neutrino cooling and gas pressure are considered .
hence reasonable amounts of non - radiative cooling do not avert a runaway to a high @xmath18 state . ] .
the disc , which would be in a similar regime to that simulated by @xcite , would be highly variable .
this initial stage , in which a highly variable magnetic corona " is built above the disc via flux expulsion , could be associated with the production of a magnetic outflow , dynamically similar to those studied by @xcite and @xcite ( see also * ? ? ?
* ) , polluted with some disk matter , that eventually will be observed as a grb
. the longer term evolution of very high @xmath4 discs is less clear .
one possibility is that eventually a _
net _ vertical field , which can not be removed locally by buoyancy , builds up and suppresses the mri .
this depends upon the ability of field lines to diffuse outwards .
turbulent diffusion can depend in turn on the strength of the magnetic field and on non - linear mhd effects . in our case
the dominance of the hall effect may influence the behaviour of the turbulence , by introducing an asymmetry between magnetic fields that are aligned or anti - aligned with the direction of the angular momentum vector @xcite .
the outcome is thus uncertain and the investigation requires numerical methods . from a purely phenomenological perspective , one might note that there are similarities between the state transitions in x - ray binaries and grb prompt and late - time activity .
our results show the presence of a high @xmath4 number inner region in both discs .
one may then speculate that the grb prompt phase could involve similar physical conditions , insofar the magnetic field behaviour is concerned , to the transition to the `` low / hard state '' in x - ray binaries , which is observationally associated with the emergence of an outflow @xcite . however , the theoretical analogy with x - ray binaries is imperfect , since it is unclear that what causes a change in state may be the same physical mechanism in both sources . for x - ray binary discs the temperature dependence of @xmath4 may allow a local thermal runaway , and
it has been suggested that this ultimately causes global state transitions @xcite .
the conditions for such a mechanism to work in hyper - accreting discs are present only in the outer radiatively inefficient regions , since in the neutrino - cooled regions @xmath4 is primarily a function of density .
however , a classical thermal instability is not possible in radiatively inefficient flows .
in this paper , we calculated the magnetic prandtl number for plasma conditions in hyper - accreting discs .
this dimensionless number @xmath284 expresses the relative size of two critical turbulent flow scales : the scale on which velocity fluctuations are viscously damped @xmath275 , and the scale @xmath285 on which ohmic losses result in magnetic field dissipation .
we find that in the inner neutrino - cooled regime : * electric resistivity involves relativistic , mildly degenerate electrons . as a consequence ,
the resistivity is very low and weakly dependent on density and temperature . *
the main source of viscosity is coulomb collisions between mildly _ degenerate electrons _ and protons .
it decreases as the flow gets denser or cooler , since the electron fraction decreases . *
the magnetic prandtl number is always _
greater _ than unity , unless the angular momentum transport efficiency , parameterized via the shakura - sunyaev @xmath18 , is very small ( @xmath286 ) .
@xmath4 ranges between a few tens to a few @xmath68 , and it typically increases with @xmath18 . in the optically _ thin _ regions for neutrinos , the main dependence of @xmath4 is on density ( inversely , via viscosity ) since the disc temperature varies very slowly as a function of @xmath38 and radius .
this causes @xmath4 to _ decrease _ with increasing accretion rates . in the optically _ thick _ regions , the main dependence is on temperature , that varies more rapidly than density . as a consequence , @xmath4 _ increases _ with @xmath38 in the outer , radiatively inefficient region : * resistivity and viscosity have the classical `` spitzer '' dependencies on temperature and density , since electrons are non - relativistic and non - degenerate and viscosity is mainly given by coulomb interactions between ionised helium particles .
the magnetic prandtl number is thus a much stronger function of temperature @xmath287 .
in contrast , it does not depend on the overall magnitude of accretion ( via @xmath38 and @xmath18 ) .
the prandtl number is always much larger than unity ( @xmath288 ) and it is unlikely to fall below unity for the entire extent of the disc .
we conclude by noting a number of implications for the global structure of hyper - accreting discs : * for all plausible values of the magnetic field strength in the disc , the hall effect provides the largest non - ideal term in the mhd equations .
these discs are perhaps a unique example of a high magnetic reynolds number flow in which the hall effect is the dominant non - ideal term .
since there are no simulations where these conditions are accounted for , their effect on the magnetic field evolution and mri is unclear . *
the large values of the magnetic prandtl number mean that the evolution of magnetic field within hyper - accreting discs is likely to be qualitatively different from discs with lower magnetic prandtl numbers .
the results of forced dynamo and mri simulations suggest that , in the high @xmath4 regime , small scale field dissipation is suppressed and the saturation level of the magnetic field is enhanced , perhaps dramatically @xcite .
numerical simulations of hyper - accreting discs that do not account for microphysical dissipation may therefore have underestimated the magnetic field strength . *
the _ immediate _ consequence of a high value of @xmath4 for models of the central engines of grbs is that disc formation is likely to be accompanied by vigorous expulsion of magnetic flux .
this favours models in which energy is liberated in the form of a magnetic jet or outflow .
the authors thank a. potekhin for providing help with his code and for useful discussions .
they also acknowledge useful discussion with c. thompson , j. mckinney , y. levin and a. beloborodov .
emr acknowledges support from nasa though chandra postdoctoral fellowship grant number pf5 - 60040 awarded by the chandra x - ray center , which is operated by the smithsonian astrophysical observatory for nasa under contract nasa8 - 03060 .
pja acknowledges support from nasa under grants nng04gl01 g and nnx07ah08 g from the astrophysics theory program . | hyper - accreting discs occur in compact - object mergers and in collapsed cores of massive stars .
they power the central engine of @xmath0-ray bursts in most scenarios .
we calculate the microphysical dissipation ( the viscosity and resistivity ) of plasma in these discs , and discuss the implications for their global structure and evolution . at the temperatures ( @xmath1 ) and densities ( @xmath2 ) characteristic of the neutrino - cooled innermost regions ,
the viscosity is provided mainly by mildly degenerate electrons , while the resistivity is modified from the spitzer value due to the effects of both relativity and degeneracy . under these conditions
the magnetic reynolds number is very large ( @xmath3 ) and the plasma behaves as an almost ideal magneto - hydrodynamic ( mhd ) fluid . among the possible non - ideal mhd effects the hall term is relatively the most important , while the magnetic prandtl number , @xmath4 ( the ratio of viscosity to resistivity ) , is typically larger than unity : @xmath5 .
inspection of the outer radiatively inefficient regions indicates similar properties , with magnetic prandtl numbers as high as @xmath6 .
numerical simulations of the magneto - rotational instability ( mri ) indicate that the saturation level and angular momentum transport efficiency may be greatly enhanced at high prandtl numbers .
if this behaviour persists in the presence of a strong hall effect we would expect that hyper - accreting discs should be strongly magnetised and highly variable .
the expulsion of magnetic field that can not be dissipated at small scales may also favour a magnetic outflow .
we note that there are limited similaries between hyper - accreting discs and x - ray binary discs which also have a high magnetic prandtl number close to the black hole which suggests that a comparison between late - time activity in @xmath0-ray bursts and x - ray binary accretion states may be fruitful .
more generally , our results imply that the possibly different character of high prandtl number mhd flows needs to be considered in studies and numerical simulations of hyper - accreting discs .
black hole physics accretion , accretion discs mhd instabilities plasmas |
over the past several years there has been a renewal of interest in the physics of lorentz symmetry violation .
these investigations are largely motivated by the enduring problem of quantum gravity and also cosmological mysteries such as the dark matter and dark energy problems .
lorentz violation ( lv ) effects in standard model extensions and modified particle dispersion relations are currently tightly constrained by high precision tests @xcite , however constraints on gravitational lv effects are not as severe .
various approaches to incorporating lv into gravity have been investigated , see for example refs .
@xcite . in this paper we will consider einstein - aether " theory ( or ae - theory , for short )
@xcite , which consists of a dynamical unit timelike vector field @xmath0 coupled to einstein gravity .
the @xmath0 aether " field can be thought of as the 4-velocity of a preferred frame .
the lv here is a form of spontaneous symmetry breaking , since the unit timelike constraint requires the vector field to have a non - zero value everywhere in any solution , even flat spacetime .
the lagrangian of ae - theory depends on four dimensionless coupling constants @xmath1 . by studying the phenomenology of ae - theory
it can be compared with observations in order to constrain the @xmath2 values and possibly rule the theory out entirely as a description of nature .
observational constraints on ae - theory have been determined from parameterized post - newtonian analysis @xcite , stability and energy positivity @xcite , primordial nucleosynthesis @xcite , and vacuum cerenkov radiation @xcite .
the combined constraints from all of these are reviewed in ref .
they are all consistent with a large ( order unity ) 2d region in the 4d parameter space .
agreement with weak field radiation damping ( such as in binary pulsar systems ) @xcite is expected to restrict the parameters to a narrow band of width @xmath3 in that 2d region , and combined constraints from many such systems may further narrow the allowed region to a small neighborhood of the origin , with size of order @xmath4-@xmath5 .
it is likely that theories with lv can generally be more tightly constrained by examining strong field solutions .
in addition , the properties of such solutions may also be of interest from a purely mathematical physics perspective since the additional fields that couple to gr can produce interesting new sorts of solutions and phenomena .
black hole solutions have been considered in the ghost condensate theory @xcite , and spherically symmetric solutions in the relativistic mond theory were studied in @xcite .
work on this subject in ae - theory was begun long ago in special cases @xcite , and a study of the solutions for general values of the coupling parameters was completed recently in @xcite and @xcite . in these papers
the time independent spherically symmetric solutions in ae - theory were characterized by examining the field equations as a set of coupled ordinary differential equations .
it was demonstrated that there is generically a three parameter family of solutions to these equations .
imposing the boundary condition of asymptotic flatness at spatial infinity reduces the solution space to a two parameter family .
this differs from gr , in which there is a unique spherically symmetric solution , the schwarzschild metric , which is automatically asymptotically flat .
the one parameter subset of solutions where @xmath0 is aligned with the timelike killing vector was found analytically and discussed in detail in @xcite .
it was also shown that this solution describes the exterior of a static fluid star .
regular black holes in ae - theory must be described by a different solution , since the aether is a timelike vector so it can not coincide with the killing vector which is null on the horizon and spacelike inside .
that is , the aether must be flowing into the black hole .
moreover , the theory possesses spin-2 , spin-1 , and spin-0 massless " wave modes that travel at speeds that differ from one another and from the metric speed of light @xcite . to qualify as a black hole ,
a solution must therefore contain a region where all of these wave modes are trapped .
it was found in @xcite that in spherical symmetry , regularity at the metric , spin-2 , and spin-1 horizons is automatic , but regularity at the spin-0 horizon is a supplementary condition .
this condition reduces the two - parameter family of asymptotically flat solutions to a one - parameter family characterized by the total mass or horizon area .
particular black hole solutions with regular spin-0 horizons were found by numerical integration for special cases of the coupling parameters . like the schwarzschild solution
, these solutions have spacelike singularities at the origin , and in the exterior the metric functions are very close to the familiar schwarzschild solution , even for fairly large coupling constants @xmath2 .
( the differences become more pronounced in the black hole interior . ) for certain large values of the coupling constants no black hole solution with a regular spin-0 horizon was found . from an astrophysical point of view ,
an important question is which solutions describe the final state of a gravitational collapse of matter fields . in @xcite
it was conjectured that the solutions with regular spin-0 horizons when they exist are the ones that arise from collapse of initially nonsingular data .
the main objective of this paper is to test that conjecture by numerical evolution of the time dependent spherically symmetric field equations given appropriate collapsing initial data .
a massless scalar field coupled only to the metric will serve as the matter field .
in section [ aetheory ] we review the nature and status of einstein - aether theory , and discuss the nature of black holes in this theory .
section [ ivform ] together with the appendix develops the initial value formulation under the assumption that the aether vector is hypersurface orthogonal , as is always the case in spherical symmetry .
the results of numerical collapse simulations are presented in section [ numresults ] .
we find that the collapsing scalar field does indeed reach a final stationary black hole state with a regular spin-0 horizon , for the range of coupling constants found in @xcite to allow such states . outside this range of @xmath2
we find solutions that develop a finite area singularity at the would - be spin-0 horizon .
the numerical evolution procedure also allows for wider explorations of space of @xmath2 than was done in @xcite .
the results appear to support the conjecture that formation of black holes with regular spin-0 horizons is generic in ae - theory for coupling parameters that are not too large .
the action for einstein - aether theory is the most general generally covariant functional of the spacetime metric @xmath6 and aether field @xmath0 involving no more than two derivatives ( not including total derivatives ) , s = ( l_ae+l_matter ) d^4x [ action ] where l_ae = [ r - k^ab_mn _ a u^m _ b u^n + ( g_abu^a u^b + 1 ) ] and @xmath7 denotes the matter lagrangian . here
@xmath8 is the ricci scalar , @xmath9 is defined as = c_1 g^abg_mn+c_2^a_m ^b_n + c_3^a_n^b_m - c_4u^a u^b g_mn where the @xmath2 are dimensionless coupling constants , and @xmath10 is a lagrange multiplier enforcing the unit timelike constraint on the aether .
the convention used in this paper for metric signature is @xmath11 and the units are chosen so that the speed of light defined by the metric @xmath6 is unity . in the weak - field , slow - motion limit ae - theory reduces to newtonian gravity @xcite , with a value of newton s constant @xmath12 related to the parameter @xmath13 in the action ( [ action ] ) by @xmath14 . in this paper
we take the matter field to be a minimally coupled massless scalar field @xmath15 , with lagrangian @xmath16 .
however , to simplify the form of the equations of motion , we introduce the quantity @xmath17 so that the matter lagrangian takes the form l_matter = _ a ^a .
the field equations from varying ( [ action ] ) with respect to @xmath18 , @xmath0 , @xmath19 , and @xmath10 are given respectively by & = & t_ab [ efe ] + _ a j^a_b + + c_4 a_a _ b u^a & = & 0 [ evolveu ] + ^a_a & = & 0 + u^au_a & = & -1.[unit ] here @xmath20 is the einstein tensor of the metric @xmath6 ; and the quantities @xmath21 and the stress - energy @xmath22 are given by & = & k^ab_mn_bu^n + a_a & = & u^b_bu_a + & = & - g_ab ( j^c_d_cu^d + _ c ) + & + & _ a + c_4 a_aa_b + + & + & c_1 ( _ au_c_bu^c - ^cu_a _
cu_b ) + & + & _ c .
[ aestress ] in this paper we will perform numerical simulations of this system in spherical symmetry .
the condition of spherical symmetry implies that the twist @xmath23 of the vector field @xmath0 vanishes .
this makes the four parameters of the lagrangian redundant : the action depends only on @xmath24 and @xmath25 where we use the abbreviation @xmath26 to stand for @xmath27 ( and correspondingly for @xmath25 ) .
( in ref .
@xcite it is shown that when the twist vanishes the @xmath28 term can absorbed by the shifts @xmath29 and @xmath30 .
thus the action depends only on @xmath31 , @xmath25 and @xmath32 .
but since the latter is equal to @xmath33 , the action is also determined by @xmath31 , @xmath26 , and @xmath25 . ) because the twist vanishes , it follows that @xmath0 is hypersurface orthogonal . for our study of scalar field collapse we considered a case where the parameters @xmath2 satisfy the requirements that ( i ) the einstein - aether theory has precisely the same post - newtonian parameters as general relativity , so it agrees in the post - newtonian approximation with all experimental tests of general relativity , ( ii ) the theory is linearly stable , ( iii ) the wave modes all carry positive energy , and ( iv ) the wave modes travel at greater than or equal to the matter speed of light , so there is no vacuum erenkov radiation . together
these impose the conditions @xcite & = & c_3 ^2 - 2c_1+c_3 3 + c_4 & = & - c_3 ^2 + 0 & < & c_13 < 1 + 0 & < & ( c_1-c_3 ) < c_13 . we chose @xmath34 and @xmath35 ( essentially in the midrange of the inequalities given above ) and then determined @xmath31 and @xmath28 by the above equalities .
this yields the parameter set c_1=,c_2=-,c_3=,c_4=- .
[ ci ] this choice of the constants @xmath2 is different from those used in the studies of static black holes in @xcite .
however , for the einstein - aether theory without matter , the transformation @xmath36 ( with @xmath37 a constant and with @xmath38 remaining unchanged ) takes einstein - aether theory into itself but with changes in the parameters @xmath2 @xcite .
this transformation and the restriction to spherical symmetry allows one to transform the results with our parameter choice ( [ ci ] ) to one with @xmath39 and @xmath28 vanishing and with @xmath40 and @xmath41 .
this new set of parameters falls in the class that lends itself to a numerical integration of the static , spherically symmetric einstein - aether equations using the methods of @xcite , and thus allows a direct comparison of a static solution with the end state of our numerical simulations of gravitational collapse . in this paper
we examine the collapse process to determine the nature of the black hole that is formed .
this issue is somewhat more subtle in einstein - aether theory than in general relativity .
usually in a gravitational collapse simulation one looks for apparent horizons : marginally outer trapped surfaces of the metric @xmath6 .
typically the spacetime outside the apparent horizon settles down to a stationary ( static in the spherically symmetric case ) black hole .
however , einstein - aether theory contains a spin-0 mode that travels at the speed @xmath42 where @xcite = c_123(2-c_14 ) .
[ v0]for the equalities that we have imposed on the @xmath2 this expression reduces to = c_13 .
[ v02 ] thus , the boundary of the black hole should be a horizon that just barely traps the spin-0 mode .
this turns out to be equivalent to a marginally outer trapped surface of the metric @xmath43 .
thus in searching for black holes we look for marginally outer trapped surfaces of the metric @xmath44 , which for brevity we will call spin-0 horizons . for the parameter choice ( [ ci ] ) ,
the spin-0 speed from ( [ v02 ] ) is @xmath45 , so the spin-0 horizon lies inside the metric horizon .
numerical evolution requires a choice of the surfaces of constant time , and we will choose those surfaces orthogonal to @xmath0 . this choice is always possible in spherical symmetry since @xmath0 is necessarily hypersurface orthogonal . the spatial metric @xmath38 and extrinsic curvature @xmath46 of these hypersurfaces
are then given by = g_ab + u_au_b [ hdef ] + k_ab = - [ kdef ] where @xmath47 denotes lie derivative .
notice that the constraint ( [ unit ] ) that @xmath0 be a unit vector is built into our initial value formulation by its identification with the unit normal . the use of @xmath46 as a variable allows us to write einstein s equation in a form that is first order in time .
similarly we introduce the quantity @xmath48 which allows us to do the same for the wave equation .
the time evolution vector field takes the form @xmath49 .
we use as a radial coordinate @xmath50 the length in the radial direction ( rather than an area coordinate ) .
this makes the spatial metric = _ a r _ b r + ^2 _ ab where @xmath51 is the unit two - sphere metric .
the quantities that are evolved are ( @xmath52 ) .
all other quantities are found from these . in the appendix ,
equations of motion for these quantities are derived from the euler - lagrange equations ( [ efe]-[unit ] ) .
these equations are & = & p + ^r_r [ dtpsi2 ] + p & = & ^r_r p + & + & [ dtp2 ] + k & = & ^r _ r k + 3 k^2 + & + & [ dtk2 ] + & = & ^r _
r a_r + [ dta2 ] + _ t & = & ^r _ r + ( q/2-k/3 ) .
[ dtphi2 ] here @xmath53 is the trace - free part of the extrinsic curvature .
as shown in the appendix , the quantities @xmath54 and @xmath55 are determined by q & = & - 3 q _
r + ( 1-c_13)^-1 [ drq2 ] + _ r & = & a_r [ dralpha2 ] + _ r^r & = & ( q + k/3 ) . [ drbeta2 ]
the hamiltonian initial value constraint is & = & _ r_r + ( _ r ) ^2 - 1 + c_14 a_r _
r + & + & 4 + & = & 0 .
[ hamreltext ] for initial data , we choose a moment of time symmetry , so that @xmath56 and @xmath57 are zero . both @xmath58 and @xmath19 can be freely specified .
we choose @xmath58 to vanish and @xmath19 to be = a_0 where @xmath59 and @xmath60 are constants .
thus the scalar field is initially at rest and is essentially a spherical shell of radius @xmath61 , thickness @xmath60 and amplitude @xmath62
. then equation ( [ hamreltext ] ) is integrated outwards to find @xmath63 , subject to the boundary conditions @xmath64 and @xmath65 at @xmath66 .
the other boundary conditions that need to be imposed are @xmath67 and @xmath68 for smoothness at @xmath66 , and @xmath69 for a gauge fixing as @xmath70 .
the evolution proceeds as follows : given ( @xmath52 ) at one time step , equations ( [ dralpha2 ] ) and ( [ drq2 ] ) are integrated to find @xmath71 and @xmath72
. then equation ( [ drbeta2 ] ) is integrated to find @xmath73
. then equations ( [ dtpsi2 ] ) , ( [ dtp2 ] ) , ( [ dtk2 ] ) , ( [ dta2 ] ) and ( [ dtphi2 ] ) are used to find respectively ( @xmath74 ) at the next time step .
since @xmath63 is evolved using equation ( [ dtphi2 ] ) but must also satisfy equation ( [ hamreltext ] ) this means that equation ( [ hamreltext ] ) can be used as a code check in a convergence test .
the specific parameters we used for the initial scalar field are @xmath75 and @xmath76 .
initial data with these parameters leads to gravitational collapse and the formation of a black hole . in the numerical simulations
spatial derivatives were implemented with standard centered finite differences .
the time evolution was done with the iterated crank - nicholson ( icn ) method .
kreiss - oliger type dissipation was used for stability .
the first grid point was at @xmath66 .
of the quantities that are evolved by the code , smoothness requires that @xmath77 and @xmath63 are odd functions of @xmath50 and that the other quantities are even .
the code evolves the quanties on all grid points except the first and last .
the odd quantities are set to zero on the first grid point , while the even quantities are set as follows : = ( 4f_2-f_3)/3 which implements the condition that the derivative of @xmath78 vanish at the origin .
here @xmath78 denotes any even quantity , and the subscript denotes the corresponding gridpoint . at the last gridpoint , @xmath79 , all the evolved quantities are set to zero with the exception of @xmath63 which is set by = 2_n-1 - _ n-2 which is the appropriate boundary condition for a quantity that asymptotically satisfies @xmath80 .
the simulations were run on a sunblade 2000 in double precision .
the number @xmath79 of gridpoints used was 12800 and the initial value of @xmath50 at the outermost gridpoint was 80 .
plot of @xmath81 _ vs _
@xmath50 for a simulation with 12800 points ( solid line ) and @xmath82 for a simulation with 6400 points ( dashed line ) .
the agreement of the two curves demonstrates second order convergence of the code .
the curves are for a time when the scalar field is in the process of collapsing . ]
figure ( [ ctest ] ) shows the result of a convergence test of the code .
it is a plot of the constraint quantity @xmath81 of equation ( [ hamreltext ] ) done at a time when the scalar field is in the process of collapsing .
the solid curve represents @xmath81 for a simulation with 12800 points , while the dashed curve represents @xmath82 for a simulation with 6400 points . the fact that the two curves argee demonstrates that the code is second order convergent .
the simulations show that for the initial data used the collapse does form a spin-0 horizon ( which is contained within an ordinary horizon , that is a marginally outer trapped surface of the metric @xmath6 ) .
furthermore , outside of the spin-0 horizon the metric and aether field do settle down into a time independent state .
in addition , all the scalar field either falls into the black hole or escapes , so that near the horizon the time independent state is that of a pure einstein - aether black hole with no matter other than the aether field @xmath0 .
we are able to follow the time evolution to late times with our horizon crossing time slices , orthogonal to the timelike aether field @xmath0 , because the lapse @xmath71 is driven to zero as the singularity is approached .
the reason for this can be traced to eqn .
( [ dralpha2 ] ) . evidently at the singularity the aether acceleration component @xmath58 goes to positive infinity . in reference @xcite the static black holes of einstein - aether theory with a regular spin-0 horizon were found numerically , and it was conjectured that these would be the endstates of gravitational collapse .
our simulations verify that this is indeed what happens .
plot of @xmath58 _ vs _
@xmath83 for the end state of the collapse simulation ( solid line ) and the static solution found using the method of @xcite ( dashed line ) ] plot of @xmath57 _ vs _
@xmath83 for the end state of the collapse simulation ( solid line ) and the static solution found using the method of @xcite ( dashed line ) ] plot of @xmath72 _ vs _
@xmath83 for the end state of the collapse simulation ( solid line ) and the static solution found using the method of @xcite ( dashed line ) ] in figures ( [ arfig ] ) , ( [ kfig ] ) and ( [ qfig ] ) are plotted respectively the quantities @xmath84 and @xmath72 for this simulation ( solid line ) and for the static solution found using the method of @xcite ( dashed line ) . here
the static solution is found using the transformed choice of the parameters @xmath2 , and its metric and aether variables are then transformed back as explained in section [ aetheory ] to make the comparison with the result of the collapse simulation . in all cases
the quantities are plotted as functions of @xmath85 where @xmath86 is the value of @xmath63 at the spin-0 horizon .
note that the agreement is extremely good , so the end result of collapse is indeed the static black hole solution found using the method of @xcite .
the quantities @xmath87 and @xmath72 completely determine the aether field , and since we are treating spherically symmetric spacetimes where there are no separate gravitational degrees of freedom , these quantities also completely determine the metric .
nonetheless , it is helpful to do a direct comparison of the metric of the endstate of the collapse simulation and the static metric found using the method of @xcite . for a static , spherically symmetric metric with area radius @xmath63 , the metric is determined by the quantities @xmath88 and @xmath89 . in figures ( [ grrfig ] ) and ( [ boxrfig ] ) these quantities
are respectively plotted .
plot of @xmath90 _ vs _
@xmath83 for the end state of the collapse simulation ( solid line ) and the static solution found using the method of @xcite ( dashed line ) ] plot of @xmath91 _ vs _
@xmath83 for the end state of the collapse simulation ( solid line ) and the static solution found using the method of @xcite ( dashed line ) ] as with figures ( [ arfig]-[qfig ] ) the quantities are plotted as functions of @xmath83 with the endstate of the collapse simulation as a solid line and the static metric found using the method of reference @xcite as a dashed line .
the agreement between the curves shows that the metric in the final state of the collapse simulation agrees with the static metric .
we now present the results of a slightly different simulation that uses directly a different parameter choice of @xcite , namely @xmath92 , and @xmath31 chosen so that the spin-0 horizon coincides with the usual metric horizon .
that is , the speed @xmath42 in ( [ v0 ] ) is set equal to one , which requires = -c_1 ^3 . if @xmath93 such theories meet all the linear stability conditions ( stable modes , positive energy , and no vacuum erenkov radiation ) . in this class
the post - newtonian preferred frame parameter @xmath94 vanishes identically , but @xmath95 does not vanish unless @xmath96 ( see @xcite for @xmath97 ) .
nevertheless , since the stability conditions are met , it seems likely that the behavior of collapse in the theory is qualitatively similar to the previous case where also @xmath95 vanishes .
plot of @xmath57 _ vs _
@xmath50 for simulations with @xmath98 ( solid line ) and @xmath99 ( dashed line ) . ] in @xcite , static , asymptotically flat black hole solutions were found in the range @xmath100 , by integrating from a regular horizon to spatial infinity using a shooting method .
the asymptotically flat boundary condition could not be met for @xmath101 .
we performed collapse simulations for the case @xmath98 and the case @xmath99 . in the @xmath98 case a regular horizon forms
however , in the @xmath99 case , no horizon forms and the evolution seems to become singular , thus indicating the formation of a naked singularity . figure ( [ c1fig ] ) shows the value of @xmath57 ( dashed line ) for the @xmath99 simulation at a time shortly before the singularity forms . for comparison ,
the same figure shows ( solid line ) the value of @xmath57 at the same time for the @xmath98 simulation .
since the collapse solutions are by construction asymptotically flat , this is consistent with what was found for static solutions in @xcite .
we have examined the process of dynamical collapse of a spherically symmetric , self - gravitating scalar field pulse in einstein - aether theory . in pure einstein gravity
it is well known that for appropriate initial conditions such a pulse can form a black hole .
ours is the first study of the analogous process in einstein - aether theory . in order to carry out the numerical simulations we developed the initial value formulation for this theory in spherical symmetry .
here we have minimized the initial distortion " of the aether field , by setting its divergence and acceleration to zero .
( at the other extreme , one could study collapse of a pure spherical aether pulse , with no matter field at all . )
these initial conditions are not precisely what would occur in an astrophysical setting , but should be indicative .
we found that for the initial data studied , black holes with regular metric and spin-0 mode horizons form when the parameters in the lagrangian are not too large . for larger parameters ,
a singularity occurs at finite area where the spin-0 horizon would have been .
these results are consistent with findings from previous investigation of static solutions @xcite .
since regular black holes have almost certainly been observed " in nature , the black hole formation found here indicates that a strong - field test has been passed by ae - theory , at least for coupling parameters that are not so large that the spin-0 horizon is singular .
constraints from the absence of vacuum erenkov radiation require that the spin-0 mode travels at a speed greater than or equal to the speed of light " as seen by matter fields @xcite , here the metric " speed of light .
thus a the singular spin-0 horizon would necessarily be on or inside the matter horizon .
whether it could then have any observational consequences remains to be determined .
it would also be interesting and astrophysically important to examine the case of collapse with angular momentum , to a spinning black hole .
the stationary spinning solutions have not yet been studied however , and to follow the time - dependent collapse would require the general initial value formulation , without assuming hypersurface orthogonal aether nor spherical symmetry .
in addition , axisymmetric codes are much more complicated .
this work was supported in part by the national science foundation under grants phy-0601800 through the university of maryland and phy-0456655 through oakland university .
in this section we derive the evolution equations for einstein - aether theory in spherical symmetry . to begin with ,
we use only the hypersurface orthogonality of @xmath0 and the euler - lagrange equations of motion ( [ efe]-[unit ] ) .
equation ( [ evolveu ] ) ( the equation of motion for the aether field ) then becomes ( l_u a_a + 2 k_aba^b - k a_a ) + c_13d^bk_ab + c_2 d_a k = 0 .
[ evolvea ] here @xmath102 denotes the lie derivative and @xmath103 is the spatial covariant derivative .
the usual initial value formulation for general relativity requires that we evolve @xmath38 and @xmath46 as well as the matter fields . from equation ( [ kdef ] ) we have = - 2 k_ab [ evolveh ] which is used to evolve @xmath38 .
the quantity @xmath57 ( the trace of @xmath104 ) sasfies the evolution equation k = - d^aa_a - a^aa_a + k^abk_ab + t_ab(h^ab + u^au^b ) .
[ evolvek ] in addition , the quantity @xmath46 satisfies the constraint equation - d_a k = - h_a^bu^ct_bc .
[ constraink ] the extrinsic curvature @xmath104 can be decomposed into a trace and a trace - free part .
we will use equation ( [ evolvek ] ) to evolve the trace @xmath57 and will use equation ( [ constraink ] ) to find the trace - free part
. however , there is a subtlety associated with the einstein - aether theory that makes the initial value problem somewhat complicated : the stress energy tensor contains terms that have second time derivatives of the aether field and second time derivatives of the metric .
thus after writing out the einstein field equations one must solve them for these second time derivatives .
in particular , some straightforward but tedious algebra applied to equation ( [ aestress ] ) yields ( h^ab + u^au^b ) & = & - ( c_13 + 3c_2)l_u k + c_14(d_aa^a+a_aa^a ) + & + & 2p^2 - 2 c_13k_abk^ab + c_123 k^2 [ splusrho ] + - h_a^bu^ct_bc & = & - p d_a + & - & c_14 ( 2 k_aba^b - k a_a + l_ua_a ) [ tzeroi ] where @xmath105 and @xmath106 . using equation ( [ splusrho ] ) in equation ( [ evolvek ] ) and solving for @xmath107 yields ( 2+c_13 + 3c_2 ) & l_u k & = ( c_14 - 2)(d_aa^a+a_aa^a ) + 2 p^2 + & + & 2(1-c_13)k_abk^ab + c_123k^2 .
[ evolvek2 ] similarly , equation ( [ constraink ] ) , upon substituting the expression in equation ( [ tzeroi ] ) , does not directly yield a constraint since the right hand side of equation ( [ tzeroi ] ) contains a term proportional to @xmath108 .
however , eliminating this term using equation ( [ evolvea ] ) yields the following constraint = ( 1+c_2)d_ak - p d_a[constraink2 ] which can be used to rewrite the equation of motion for @xmath109 as & = & - 2 k_aba^b+ka_a + c_13 p d_a + & - & c_123 d_a k. [ evolvea2 ] the usual hamiltonian constraint of general relativity is r+k^2-k^abk_ab = 2 t_abu^au^b [ hamrel ] where @xmath110 is the scalar curvature of the spatial metric .
however , it follows from equation ( [ aestress ] ) that 2 t_abu^au^b & = & 2 c_14 d_aa^a + p^2 + d_a + & + & c_14 a_aa^a - c_2 k^2 - c_13k_abk^ab .
[ aerho ] on substituting equation ( [ aerho ] ) into equation ( [ hamrel ] ) one finds the constraint r & = & c_14 ( 2 d_aa^a+a_aa^a ) + p^2 + d_a + & + & ( 1-c_13)k_abk^ab - ( 1+c_2)k^2 .
[ hamrel2 ] we now impose spherical symmetry .
we use as a radial coordinate @xmath50 the length in the radial direction ( rather than an area coordinate ) .
this makes the spatial metric = _ a r _ b r + ^2 _ ab where @xmath51 is the unit two - sphere metric .
the time evolution vector field takes the form @xmath111 . from the definition of @xmath112 we find = p + ^r_r [ dtpsi ]
while from the wave equation we obtain p = ^r_r p + .
[ dtp ] it is helpful to define @xmath113 so that @xmath72 is the component of the trace - free part of @xmath104 . from equation ( [ evolveh ] )
, we have the standard result = - 2 + l _ h_ab . [ evolveh2 ] the @xmath114 component of equation ( [ evolveh2 ] ) yields = ( q + k/3 ) [ drbeta ] while the @xmath115 component yields = ^r _ r + ( q/2-k/3 ) .
[ dtphi ] from the definition of @xmath71 we have = a_r .
[ dralpha ] equation ( [ constraink2 ] ) yields q = - 3 q _
r + ( 1-c_13)^-1 [ drq ] while equations ( [ evolvea2 ] ) and ( [ evolvek2 ] ) yield respectively & = & ^r _
r a_r + [ dta ] + k & = & ^r _ r k + 3 k^2 + & + & .
[ dtk ] the hamiltonian constraint ( equation ( [ hamrel2 ] ) ) becomes the vanishing of the quantity @xmath81 where & = & _ r_r + ( _ r ) ^2 - 1 + c_14 a_r _
r + & + & 4 .
[ hamrel3 ] d. mattingly , `` modern tests of lorentz invariance , '' living rev .
* 8 * , 5 ( 2005 ) [ arxiv : gr - qc/0502097 ] .
will and k. nordtvedt , jr . ,
conservation laws and preferred frames in relativistic gravity .
i. preferred - frame theories and an extended ppn formalism , " astrophys .
j. * 177 * , 757 ( 1972 ) ; k. nordtvedt , jr . and c.m .
will , conservation laws and preferred frames in relativistic gravity .
ii . experimental evidence to rule out preferred - frame theories of gravity , " astrophys .
j. * 177 * , 775 ( 1972 ) ; r.w .
hellings and k. nordtvedt , jr .
, vector - metric theory of gravity , " phys .
rev . * d7 * , 3593 ( 1973 ) .
m. gasperini , `` classical repulsive gravity and broken lorentz symmetry , '' phys .
d * 34 * , 2260 ( 1986 ) .
v. a. kostelecky and s. samuel , `` gravitational phenomenology in higher dimensional theories and strings , '' phys .
d * 40 * , 1886 ( 1989 ) .
m. a. clayton and j. w. moffat , `` scalar - tensor gravity theory for dynamical light velocity , '' phys .
b * 477 * , 269 ( 2000 ) [ arxiv : gr - qc/9910112 ] .
t. jacobson and d. mattingly , `` gravity with a dynamical preferred frame , '' phys .
d * 64 * , 024028 ( 2001 ) [ arxiv : gr - qc/0007031 ] .
n. arkani - hamed , h. c. cheng , m. a. luty and s. mukohyama , `` ghost condensation and a consistent infrared modification of gravity , '' jhep * 0405 * , 074 ( 2004 ) [ arxiv : hep - th/0312099 ] .
j. d. bekenstein , `` relativistic gravitation theory for the mond paradigm , '' phys .
d * 70 * , 083509 ( 2004 ) [ erratum - ibid .
d * 71 * , 069901 ( 2005 ) ] [ arxiv : astro - ph/0403694 ] .
b. m. gripaios , `` modified gravity via spontaneous symmetry breaking , '' jhep * 0410 * , 069 ( 2004 ) [ arxiv : hep - th/0408127 ] .
r. bluhm and v. a. kostelecky , `` spontaneous lorentz violation , nambu - goldstone modes , and gravity , '' phys .
d * 71 * , 065008 ( 2005 ) [ arxiv : hep - th/0412320 ] . c. heinicke , p. baekler and f. w. hehl , `` einstein - aether theory , violation of lorentz invariance , and metric - affine gravity , '' phys .
d * 72 * , 025012 ( 2005 ) [ arxiv : gr - qc/0504005 ] .
v. a. rubakov , `` phantom without uv pathology , '' arxiv : hep - th/0604153 .
h. c. cheng , m. a. luty , s. mukohyama and j. thaler , `` spontaneous lorentz breaking at high energies , '' jhep * 0605 * , 076 ( 2006 ) [ arxiv : hep - th/0603010 ] . t. g. zlosnik , p. g. ferreira and g. d. starkman , `` modifying gravity with the aether : an alternative to dark matter , '' arxiv : astro - ph/0607411 . c. eling , t. jacobson and d. mattingly , `` einstein - aether theory , '' in _ deserfest _ , eds .
j. liu , m. j. duff , k. stelle , and r. p. woodard ( world scientific , 2006 ) arxiv : gr - qc/0410001 . c. eling and t. jacobson , `` static post - newtonian equivalence of gr and gravity with a dynamical preferred frame , '' phys .
d * 69 * , 064005 ( 2004 ) [ arxiv : gr - qc/0310044 ] .
m. l. graesser , a. jenkins and m. b. wise , `` spontaneous lorentz violation and the long - range gravitational preferred - frame effect , '' phys .
b * 613 * , 5 ( 2005 ) [ arxiv : hep - th/0501223 ] .
b. z. foster and t. jacobson , `` post - newtonian parameters and constraints on einstein - aether theory , '' phys .
d * 73 * , 064015 ( 2006 ) [ arxiv : gr - qc/0509083 ] .
t. jacobson and d. mattingly , `` einstein aether waves , '' phys.rev .
d * 70 * , 024003 ( 2004 ) [ arxiv : gr - qc/0402005 ] .
e. a. lim , `` can we see lorentz - violating vector fields in the cmb ?
, '' phys .
d * 71 * , 063504 ( 2005 ) [ arxiv : astro - ph/0407437 ] .
j. w. elliott , g. d. moore and h. stoica , `` constraining the new aether : gravitational cherenkov radiation , '' jhep * 0508 * , 066 ( 2005 ) [ arxiv : hep - ph/0505211 ] . c. eling , `` energy in the einstein - aether theory , '' phys .
d * 73 * , 084026 ( 2006 ) [ arxiv : gr - qc/0507059 ] .
s. m. carroll and e. a. lim , `` lorentz - violating vector fields slow the universe down , '' phys .
d * 70 * , 123525 ( 2004 ) [ arxiv : hep - th/0407149 ] .
b. z. foster , `` radiation damping in einstein - aether theory , '' phys .
d * 73 * , 104012 ( 2006 ) [ arxiv : gr - qc/0602004 ] . b. z. foster , in preparation .
s. mukohyama , `` black holes in the ghost condensate , '' phys .
d * 71 * , 104019 ( 2005 ) [ arxiv : hep - th/0502189 ] .
d. giannios , `` spherically symmetric , static spacetimes in teves , '' phys .
d * 71 * , 103511 ( 2005 ) [ arxiv : gr - qc/0502122 ] . c. eling and t. jacobson , `` spherical solutions in einstein - aether theory : static aether and stars , '' class .
* 23 * , 5625 ( 2006 ) [ arxiv : gr - qc/0603058 ] . c. eling and t. jacobson , `` black holes in einstein - aether theory , '' class .
* 23 * , 5643 ( 2006 ) [ arxiv : gr - qc/0604088 ] .
b. z. foster , `` metric redefinitions in einstein - aether theory , '' phys .
d * 72 * , 044017 ( 2005 ) [ arxiv : gr - qc/0502066 ] . | we study gravitational collapse of a spherically symmetric scalar field in einstein - aether theory ( general relativity coupled to a dynamical unit timelike vector field ) .
the initial value formulation is developed , and numerical simulations are performed .
the collapse produces regular , stationary black holes , as long as the aether coupling constants are not too large . for larger couplings a finite area singularity occurs .
these results are shown to be consistent with the stationary solutions found previously . h l |
there is a strong historical basis for the consideration of composite models which is presently mirrored in the proliferation of fundamental particles . in attempts to explain the repetition of generations or the large number of arbitrary parameters within the standard model ( sm ) , several levels of substructure have been considered@xcite , including composite fermions , higgs bosons , and even weak bosons . here
we focus on the possibility that leptons and quarks are bound states of more fundamental constituents , often referred to as preons in the literature .
the preon binding force should be confining at a mass scale @xmath11 which also characterizes the radius of the bound states .
experimentally , @xmath11 is constrained to be at least in the tev range .
theoretically , numerous efforts have been made to construct realistic models for composite fermions , but no consistent or compelling theory which accounts for @xmath12 presently exists . at energies above @xmath11 the composite nature of fermions would be revealed by the break - up of the bound states in hard scattering processes . at lower energies , deviations from the sm may be observed via form factors or residual effective interactions induced by the binding force .
these composite remnants are usually parameterized by the introduction of contact terms in the low - energy lagrangian .
more generally , four fermion contact interactions represent a useful parameterization of many types of new physics originating at a high energy scale , such as the exchange of new gauge bosons , leptoquarks , or excited particles , or the existence of anomalous couplings .
these contact interactions are described by non - renormalizable operators in the effective low - energy lagrangian .
the lowest order four - fermion contact terms are dimension-6 and hence have dimensionful coupling constants proportional to @xmath13 .
the fermion currents are restricted to be helicity conserving , flavor diagonal , and @xmath14 invariant .
these terms can be written most generally as@xcite @xmath15 where the generation and color indices have been suppressed , @xmath16 , and @xmath17 is inserted to allow for different quark and lepton couplings but is anticipated to be @xmath18 . since the binding force is expected to be strong when @xmath19 approaches @xmath20 , it is conventional to define @xmath21 .
the subscript @xmath22 indicates that the currents in each parenthesis can be either left- or right - handed and various possible choices of the chiralities lead to different predictions for the angular distributions of the reactions where the contact terms contribute .
interference between the contact terms and the usual gauge interactions can lead to observable deviations from sm predictions at energies lower than @xmath11 .
they can affect _
e.g. _ , jet production at hadron colliders , the drell - yan process , or lepton scattering .
the size of this interference term relative to the sm amplitude is @xmath23 , where @xmath24 represents the strength of the relevant gauge coupling .
one may hence neglect modifications of the gauge couplings due to form factors .
it is clear that the effects of the contact interactions will be most important in the phase space region with large @xmath19 . at hadron
colliders these terms manifest themselves in the high @xmath25 region in jet and lepton - pair production and deviations from the sm can unfortunately often become entangled in the uncertainties associated with the parton densities@xcite .
cdf has recently constrained@xcite @xmath26 contact interactions from the measurement of the dijet angular distribution ; it is found to be in good agreement with qcd , thereby excluding at 95 % c.l . a contact interaction among the up and down
type quarks with scale @xmath27 tev or @xmath28 tev .
composite scales of 25 tev can be reached in future runs at the tevatron@xcite and a search limit for @xmath11 in the 1520 tev range is expected at the lhc @xcite .
cdf also found the restrictions@xcite @xmath29 tev and @xmath30 tev at 95% c.l . on @xmath31 contact interactions from @xmath32 of data on drell - yan production .
run ii of the tevatron is expected to improve these limits to @xmath33 tev .
hera also constrains @xmath31 contact terms , with the exclusion@xcite from h1 of @xmath34 tev at 95% c.l . , where the range takes into account various helicity combinations . a review of the bounds on four lepton contact interactions from pep
, petra , tristan , and aleph is given in buskulic _
et al._@xcite , but are superseded by recent results from opal at @xmath35 gev which bounds @xmath11 in the range @xmath36 tev ( again , for the various helicity states ) from @xmath37 and combined @xmath38 pair production @xcite .
this search also constrains @xmath39 tev and @xmath40 tev at 95% c.l . from identified b - quark final states .
there is an earlier analysis@xcite of @xmath41 contact terms from the forward - backward asymmetry of @xmath42 and @xmath43 mesons which yields a bound of @xmath44 tev . in this contribution , we study the compositeness search reach on @xmath0 and @xmath1 contact terms using the processes @xmath45 where @xmath46 or @xmath47 at future lepton colliders .
we shall consider @xmath3 colliders with center of mass energy 0.5 , 1 , 1.5 , 5 tev and luminosity 50 , 200 , 200 , 1000 fb@xmath48 , respectively , as well as muon colliders with @xmath49 tev and luminosity 0.7 , 50 , and 1000 fb@xmath48 .
we build on earlier studies@xcite of compositeness searches at lepton colliders .
the reactions @xmath50 where @xmath51 and @xmath52 ( @xmath53 proceed via @xmath54-channel exchanges of @xmath55 and @xmath56 bosons , as well as the @xmath57 contact interaction .
thus , not only the squared term of the contact interaction but also the interference terms between the @xmath55 , @xmath56 exchanges and the contact interaction will contribute to the differential cross section and yield deviations from the sm .
we explicitly rewrite the contact terms for @xmath58 @xmath59 \,.\end{aligned}\ ] ] the polarized differential cross sections for @xmath60 are @xmath61 @xmath62 where @xmath63 is the scattering angle in the center of mass frame and @xmath64 represents the color factor being the usual 3(1 ) for quarks(leptons ) .
the helicity amplitudes are @xmath65 with @xmath66 , @xmath67 , @xmath68 and @xmath69 are the sine and cosine of the weak mixing angle , and @xmath70 and @xmath71 represent the fermion s charge and third component of the weak isospin , respectively .
the expressions for @xmath72 and @xmath73 can be obtained by interchanging @xmath74 .
the use of polarized beams , combined with the angular distributions , can thus clearly determine the helicity of the contact term .
the unpolarized differential cross section is given simply by @xmath75\,.\ ] ] the polarized and unpolarized total cross sections are obtained by integrating over @xmath76 , resulting in the spin - averaged unpolarized cross section : @xmath77\,.\ ] ] the forward - backward and left - right asymmetries are easily obtained and can be written as @xmath78 figure 1 displays the @xmath76 distribution for @xmath79 at @xmath80 tev for the sm and with a contact term present .
the effects of a contact term are qualitatively similar for other final states . in all curves we set @xmath81 . in fig .
1a we take @xmath82 tev which shows that a finite value of @xmath11 alters the angular distribution particularly in the forward direction .
1b displays the angular distributions for right - handed polarized electrons .
although the effects of contact interactions are more dramatic here , because the right - handed cross section is smaller , the relative contribution of right - handed contact terms will be smaller in unpolarized cross sections .
thus , not only will polarization be important for disentangling the helicity structure of a contact interaction should deviations be seen , but polarization will also enhance the sensitivity to contact interactions . in fig .
1c distributions are shown for @xmath83 and @xmath84 demonstrating that opposite signs for the @xmath85 s results in opposite interference . finally , in fig .
1d the angular distribution is shown for @xmath86 but with @xmath87 5 tev , 10 tev , 20 tev , and 30 tev to give a feeling for the sensitivity to the scale of new physics .
we note that the effects of the contact term on @xmath88 are relatively small when all quark flavors are summed , compared to the individual deviations in _
e.g. _ , @xmath89 or @xmath90 , because cancelations occur in the interference term between the up - type and down - type quarks .
we thus concentrate on the heavy quark final states , taking a 60% identification efficiency for detecting b - quarks and 35% identification efficiency for detecting c - quarks at the nlc@xcite .
the detection efficiency of heavy flavor final states at a muon collider has yet to be determined , but is expected to be worse than what can be achieved at the nlc due to the inability to put a vertex detector close to the interaction point and due to the heavier backgrounds .
for now we assume canonical lep values , @xmath91 for the muon collider but warn the reader that these numbers are quite arbitrary and are only used for illustrative purposes .
we assume 100% identification efficiency for leptons .
although we do not take into account the purity of the tagged heavy flavor samples in our results , we note that the purities that can be achieved at a linear collider are higher than can be achieved at lep .
to gauge the sensitivity to the compositeness scale we assume that the sm is correct and perform a @xmath92 analysis of the @xmath76 angular distribution . to perform this we choose the detector acceptance to be @xmath93 ( corresponding to @xmath94 ) for the @xmath3 collider and @xmath95 ( corresponding to @xmath96 ) for the muon collider@xcite .
we note that angular acceptance of a typical muon collider detector is expected to be reduced due to additional shielding required to minimize the radiation backgrounds from the muon beams .
we then divide the angular distribution into 10 equal bins .
the @xmath92 distribution is evaluated by the usual expression : @xmath97 ^ 2\ ] ] where @xmath98 is the luminosity and @xmath99 is the efficiency for detecting the final state under consideration which is discussed above . for polarized beams we assume @xmath100 of the total integrated luminosity listed in the tables for each polarization .
we assume that only one of the @xmath85 s is nonzero at a time .
the 95% c.l .
bounds on @xmath11 are tabulated in tables i , ii , and iii .
generally , high luminosity @xmath101 and @xmath5 colliders are quite sensitive to contact interactions with discovery limits ranging from 5 to 50 times the center of mass energy . for unpolarized beams with leptons in the final state ,
the slightly higher sensitivity to contact interactions at @xmath101 colliders than at @xmath5 colliders with the same @xmath102 can be attributed to the larger expected acceptance for @xmath3 detectors . for @xmath103 final states
the sensitivities for @xmath101 colliders are roughly 20% higher than for @xmath104 colliders while for @xmath105 final states the difference can be up to a factor of two .
these differences are due to the different tagging efficiencies assumed for @xmath3 and @xmath5 colliders .
polarization in the @xmath3 colliders can offer even higher limits depending on the final state being considered .
more importantly , if deviations are observed polarization would be crucial for determining the chirality of the new interaction .
finally , we note that for @xmath103 and @xmath105 final states sometimes very specific , relatively low values of @xmath11 , give rise to angular distributions indistinguisiable from the sm .
however , we expect that these values will be ruled out by other measurements before high energy lepton colliders become operational so we only include the higher values in the tables .
in this report we presented the results of a preliminary study on the sensitivity to contact interactions at future high energy @xmath101 and @xmath5 colliders . depending on the specific collider and final state
, contact interactions can be detected up to 5 - 50 times the center of mass energy of the collider with the lowest number coming from the low luminosity 500 gev @xmath5 collider and the highest numbers from high luminosity @xmath3 colliders with polarization .
these results should be taken as preliminary .
first and foremost the sensitivities were based only on statistical errors and systematic errors were not included .
in addition , a more thorough analysis should include potentially important effects like initial state radiation and should consider heavy quark final state purities .
these considerations are under study and will be presented elsewhere @xcite .
s.g . thanks dean karlen for helpful conversations .
lllll process & @xmath106 & @xmath107 & @xmath108 & @xmath109 + + @xmath110 ( p=1.0 ) & 33 & 30 & & + @xmath110 ( p=0.9 ) & 32 & 29 & 10 & 10 + @xmath111 ( p=1.0 ) & & & 30 & 33 + @xmath111 ( p=0.9 ) & 11 & 10 & 29 & 31 + @xmath112 & 28 & 26 & 26 & 27 + @xmath113 ( p=1.0 ) & 39 & 32 & & + @xmath113 ( p=0.9 ) & 38 & 30 & 5.3 & 9.1 + @xmath114 ( p=1.0 ) & & & 35 & 38 + @xmath114 ( p=0.9 ) & 17 & 12 & 33 & 33 + @xmath115 & 37 & 28 & 28 & 25 + @xmath116 ( p=1.0 ) & 34 & 29 & & + @xmath116 ( p=0.9 ) & 33 & 28 & 5.0 & 8.5 + @xmath117 ( p=1.0 ) & & & 27 & 34 + @xmath117 ( p=0.9 ) & 12 & 11 & 24 & 32 + @xmath118 & 31 & 28 & 18 & 26 + + @xmath110 ( p=1.0 ) & 66 & 60 & & + @xmath110 ( p=0.9 ) & 63 & 57 & 20 & 21 + @xmath111 ( p=1.0 ) & & & 61 & 66 + @xmath111 ( p=0.9 ) & 22 & 20 & 58 & 62 + @xmath112 & 57 & 51 & 51 & 55 + @xmath113 ( p=1.0 ) & 78 & 64 & & + @xmath113 ( p=0.9 ) & 75 & 61 & 11 & 18 + @xmath114 ( p=1.0 ) & & & 70 & 76 + @xmath114 ( p=0.9 ) & 34 & 24 & 65 & 67 + @xmath115 & 74 & 56 & 55 & 50 + @xmath116 ( p=1.0 ) & 68 & 59 & & + @xmath116 ( p=0.9 ) & 65 & 57 & 9.9 & 17 + @xmath117 ( p=1.0 ) & & & 55 & 68 + @xmath117 ( p=0.9 ) & 24 & 22 & 49 & 62 + @xmath118 & 61 & 55 & 37 & 51 + lllll process & @xmath106 & @xmath107 & @xmath108 & @xmath109 + + @xmath110 ( p=1.0 ) & 81 & 74 & & + @xmath110 ( p=0.9 ) & 77 & 70 & 25 & 26 + @xmath111 ( p=1.0 ) & & & 74 & 81 + @xmath111 ( p=0.9 ) & 27 & 25 & 70 & 76 + @xmath112 & 70 & 63 & 63 & 67 + @xmath113 ( p=1.0 ) & 95 & 80 & & + @xmath113 ( p=0.9 ) & 92 & 77 & 15 & 23 + @xmath114 ( p=1.0 ) & & & 84 & 94 + @xmath114 ( p=0.9 ) & 41 & 31 & 78 & 82 + @xmath115 & 90 & 70 & 66 & 61 + @xmath116 ( p=1.0 ) & 83 & 72 & & + @xmath116 ( p=0.9 ) & 80 & 69 & 14 & 20 + @xmath117 ( p=1.0 ) & & & 67 & 83 + @xmath117 ( p=0.9 ) & 29 & 26 & 60 & 76 + @xmath118 & 75 & 67 & 44 & 63 + + @xmath110 ( p=1.0 ) & 220 & 200 & & + @xmath110 ( p=0.9 ) & 210 & 190 & 70 & 71 + @xmath111 ( p=1.0 ) & & & 200 & 220 + @xmath111 ( p=0.9 ) & 75 & 70 & 190 & 210 + @xmath112 & 190 & 170 & 170 & 180 + @xmath113 ( p=1.0 ) & 260 & 220 & & + @xmath113 ( p=0.9 ) & 250 & 210 & 49 & 66 + @xmath114 ( p=1.0 ) & & & 230 & 250 + @xmath114 ( p=0.9 ) & 110 & 89 & 210 & 220 + @xmath115 & 250 & 200 & 180 & 170 + @xmath116 ( p=1.0 ) & 220 & 190 & & + @xmath116 ( p=0.9 ) & 210 & 190 & 43 & 46 + @xmath117 ( p=1.0 ) & & & 180 & 220 + @xmath117 ( p=0.9 ) & 78 & 38 & 160 & 210 + @xmath118 & 200 & 180 & 110 & 170 + lllll process & @xmath106 & @xmath107 & @xmath108 & @xmath109 + + @xmath119 & 9.8 & 9.1 & 9.1 & 9.4 + @xmath120 & 10 & 8.8 & 4.9 & 7.6 + @xmath121 & 5.6 & 3.6 & 4.2 & 2.7 + + @xmath119 & 28 & 25 & 25 & 27 + @xmath120 & 29 & 22 & 21 & 20 + @xmath121 & 19 & 16 & 5.7 & 15 + + @xmath119 & 170 & 150 & 150 & 160 + @xmath120 & 180 & 140 & 120 & 120 + @xmath121 & 110 & 92 & 42 & 90 + b. schrempp _
et al . _ ,
* b296*,1 ( 1988 ) ; e.n .
_ , nucl . phys . *
b354 * , 1 ( 1991 ) ; j. ellis , and f. pauss , _ proceedings of the workshop on physics at future accelerators _ , la thuile , italy , cern 87 - 07 ( 1987 ) .
nlc accelerator design group and physics working group , ( s. kulhman _ et al .
_ ) , _ physics and technology of the next linear collider _ , bnl-52502 ( 1996 ) ; @xmath5 collider collaboration ( r.palmer _ et al .
_ ) , _ @xmath5 collider : a feasibility study _ , bnl-52503 ( 1996 ) . | fermion compositeness and other new physics can be signalled by the presence of a strong four - fermion contact interaction .
here we present a study of @xmath0 and @xmath1 contact interactions using the reactions : @xmath2 at future @xmath3 linear colliders with @xmath4 tev and @xmath5 colliders with @xmath6 tev .
we find that very large compositeness scales can be probed at these machines and that the use of polarized beams can unravel their underlying helicity structure .
-0.5 in 6.5 in 8.50 in -0.5 in 4.0=0.5 in epsfig.sty # 1@xmath7 # 1 # 1 # 1[#1 ] * probing contact interactions at high energy lepton colliders * kingman cheung@xmath8 , stephen godfrey@xmath9 , and joanne hewett@xmath10 + .3 cm @xmath8university of texas , austin tx 78712 + .3 cm @xmath9ottawa carleton institute for physics , carleton university , ottawa , canada + .3 cm @xmath10stanford linear accelerator center , stanford , ca 94309 .3 cm 0.75 in to appear in the _ proceedings of the 1996 dpf / dpb summer study on new directions for high energy physics - snowmass96 _ , snowmass , co , 25 june - 12 july , 1996 . |
it is by now an accepted paradigm that the formation of globular clusters ( gcs ) is closely linked to the formation of their host galaxies ( e.g. searle & zinn 1978 , ashman & zepf 1992 , kissler - patig 1997 , hilker _
et al . _ 1999b , forbes 1997 , ct _ et al
_ 1998 , beasley 2002 ) . in recent years
, observational data for extragalactic gc systems have made a huge leap forward in quality , thanks mainly to high - resolution imaging from the hubble space telescope , and to wide field imaging and multi - object spectroscopy from 8m - class ground based telescopes ( see west 2004 and references therein ) . as a rule ,
studies of extragalactic gcs have tended to focus on correlations between the global photometric properties of the gc systems ( gcss ) and those of their host galaxy .
for instance , peng @xcite , examined the color distributions for gcs belonging to 100 galaxies observed in the acs virgo cluster survey ( acsvcs ; ct 2004 ; see also below ) , finding nearly all of these galaxies ( @xmath7 mag ) to possess bimodal , or at least asymmetric , gc color distributions .
consistent with previous findings ( e.g. , gebhardt & kissler - patig 1999 ; kundu 2001 ; larsen 2001 ) , the mean gc color was found to correlate with galaxy luminosity in the same sense as the well - known color - magnitude relation for early type galaxies ( e.g. bower , lucey & ellis 1992 ; karick 2003 ; hilker 2003 ; ferrarese 2006 ) .
a corresponding gc color - galaxy color relation was also found by peng @xcite , confirming previous findings ( e.g. , burgarella 2001 ; larsen 2001 ; lotz 2004 ) . at the same time
, analyses of the color magnitude diagrams for the gcss of early - type galaxies revealed some tentative evidence for a dependence of _ individual _ gc color on luminosity .
ostrov @xcite found from washington photometry of ngc 1399 that the color peaks of the gc color distribution seem to merge at the highest luminosities .
dirsch et al .
@xcite confirmed this finding for the same galaxy based on an expanded sample of gcs .
there have also been hints of such a behavior in other galaxies such as ngc 5128 ( harris 2004 ) and m87 ( whitmore 1995 ; larsen 2001 ) . with the improved imaging capabilities of hst provided by the advanced camera for surveys ( acs ; ford 1998 ) , it has become possible to investigate this issue in more detail .
harris @xcite used acs imaging to investigate the distribution of gcs in the @xmath8 color magnitude plane for eight brightest cluster galaxies in the range @xmath9 km s@xmath10 . for the four galaxies with the largest number of gcs and also the joint sample of eight gc systems they found a significant trend towards redder colors for increasing luminosity for the subpopulation of blue gcs .
this finding led them to suggest a mass - metallicity relation of the form @xmath11 for this subpopulation .
curiously , no color - magnitude relation was found for the red gcs .
strader @xcite used the publicly available data from the acsvcs to look for color - magnitude trends in the three brightest virgo cluster galaxies m49 , m87 and m60 . for the blue gc populations of m87 and m60
, they found a significant trend of redder color with brighter gc @xmath12-band luminosity similar to that found by harris @xcite .
they too interpreted this trend as evidence for a mass - metallicity relation among the blue ( metal - poor ) gcs and suggested self - enrichment as a possible explanation .
no significant color - magnitude trend was found for m49 , the brightest member of the virgo cluster .
both the harris @xcite and strader @xcite studies used heteroscedastic mixture models to fit the colors of an assumed double - gaussian color distribution to the observed gc color - magnitude diagrams ( cmds ) as a function of magnitude .
et al . _
use ( e.g. ashman , bird & zepf 1994 ) , while strader _
et al . _
use nmix ( richardson & green 1997 ) .
two important issues that were not addressed by these studies are the subject of this paper : first , we will investigate the color - magnitude correlations not only in the red ( i.e. , @xmath12 ) , but also in the blue band ( i.e. , @xmath13 ) ; second , we will investigate how the blue peak slope varies as a function of host galaxy luminosity and galactocentric distance .
note that the harris @xcite study targeted only brightest cluster galaxies , so it spanned a relatively limited range in galaxy luminosity .
we finally point out that there is recent evidence for the existence of a gc color - magnitude relation also in spiral galaxies . based on acs imaging , spitler _
et al . _
@xcite found a significant color - magnitude trend for the sombrero galaxy ( ngc 4594 ) implying @xmath14 , which is a somewhat weaker trend than suggested by harris @xcite and strader @xcite for giant ellipticals .
the aim of this paper is to investigate the color - magnitude trend of gcs over the entire host galaxy luminosity range covered by the acsvcs .
special emphasis will be put on environmental dependencies of this trend and checks of the fitting technique .
the paper is structured as follows : in [ gcsel ] we outline how gc candidates are selected in the acsvcs . in [ realvssim ]
we compare fits to simulated cmds with fits to observed cmds . in
[ environment ] we investigate the environmental dependence of the color - magnitude trend . in [ discussion ]
we examine a variety of physical effects that may create these trends .
we finish this paper with the summary and conclusions in [ conclusions ] .
the acsvcs represents the most complete and homogeneous study of extragalactic gcs performed to date .
the sample consists of 100 early - type ( e , s0 , de , de , n , ds0 ) members of the virgo cluster .
each galaxy was imaged in the f475w and f850lp filters ( @xmath15 and @xmath16 , respectively ) for a total of 750 seconds and 1210 seconds , respectively .
this filter combination gives roughly a factor - of - two improvement in wavelength baseline and metallicity sensitivity compared to the `` canonical '' @xmath17 color index .
the identification of bona - fide gcs from these images is performed in the size - magnitude plane as described in peng @xcite .
this selection procedure which is possible because the half - light radii of gcs are marginally resolved ( jordn @xcite ) at the distance of virgo ( @xmath18 mpc ; tonry @xcite ; mei @xcite ) greatly reduces contamination from foreground stars and background galaxies .
the acs images had been reduced using a dedicated pipeline described in detail by jordn ( 2004ab ; papers ii and iii ) . in brief , the reduction steps are : image combination , galaxy modeling and model subtraction , rejection of obvious background galaxies , and the measurement of magnitudes and sizes for candidate gcs using the program kingphot ( jordn 2005 ; paper x ) . the result is a catalog of integrated @xmath13 and @xmath12 magnitudes , @xmath2 colors and half - light radii , @xmath19 , for each candidate gc .
magnitudes and colors are corrected for foreground extinction using the reddening maps of schlegel @xcite . to estimate the contamination by background galaxies ,
an identical reduction procedure was applied to 17 blank high - latitude fields observed by acs in the @xmath13 and @xmath12 filters . because background galaxies are typically larger at comparable luminosities ( or , conversely , fainter at comparable sizes )
, a statistical decontamination is performed in size - magnitude space ( see peng 2006 and jordn 2006 , in prep . for details ) : to this end , first a non - parametric density model is fitted to the size - magnitude distribution of sources detected in background fields . for the gc candidate magnitude distribution , a gaussian luminosity function is assumed .
the gc candidate size distribution is fitted by a non - parametric kernel estimate plus a power law tail .
the resulting surface densities in the size - magnitude plane for background sources and gcs then make it possible to assign a gc probability , @xmath20 , to every source detected in the galaxy image .
all candidates with probabilities @xmath21 are considered gcs .
the remaining contamination by background galaxies that are assigned @xmath21 is negligible for the sake of this study .
we have tested this by statistically cleaning the gc cmds for each galaxy , using the color - magnitude distribution of sources with @xmath22 in the 17 blank background fields .
figure [ cmd1316 ] shows @xmath6-@xmath2 and @xmath3-@xmath2 cmds for the gcs of m87 (= ngc4486 = vcc1316 ) , the central galaxy of the virgo cluster and the galaxy with the singlest largest population of gcs . for comparison ,
figure [ cmdclasses ] shows the co - added cmds for the 79 acsvcs galaxies that have reliable sbf distances ( mei 2005 , 2006 ) and which have a discernible gc population ( see peng 2006 ) .
we use the sbf distance moduli from mei ( 2005 , 2006 ) , which have an internal precision of @xmath23 mag , to assign absolute magnitudes to the associated gc candidates . to better visualize trends across host galaxy magnitude ,
the 79 galaxies are subdivided into four broad luminosity groups such that the number of gcs per group falls in the range @xmath24 2400 to 3300 .
luminosity group 1 consists of the three brightest avcvcs galaxies m49 (= ngc4472 = vcc1226 ; @xmath25 mag ) , m87 (= ngc4486 = vcc1316 ; @xmath26 mag ) , and m60 (= ngc4649 = vcc1978 ; @xmath27 mag ) .
group 2 consists of six acsvcs galaxies with @xmath28 mag .
group 3 consists of 17 acsvcs galaxies with @xmath29 mag .
the remaining 53 acsvcs galaxies with @xmath30 mag make up group 4 .
the three galaxies in group 1 are all so - called `` core - srsic galaxies '' ( ferrarese 2006 ) , while those in groups 2 and 3 are a mixture of core - srsic and srsic elliptical and lenticular galaxies .
the 53 galaxies in group 4 consist mainly , but not exclusively , of early - type dwarfs ( e.g. , de , de , n , ds0 and ds0,n ) .
we have applied to the combined cmds for these four groups , as well as to the individual cmds for the three galaxies in group 1 . to this end
, we subdivided the cmds into luminosity bins , each bin containing the same number of data points .
for the cmds of m49 and m60 , the size of each luminosity bin is 100 , for m87 the size is 150 , and for the combined cmds the bin size is 200 gcs .
was then run on the gc color distribution in each luminosity bin . as in the paper by harris _
et al . _
@xcite , we use the heteroscedastic option of kmm : we allow for gaussians of different widths to be fitted to the blue and red gc populations .
the fitted mean positions are plotted over the respective cmds in figures [ cmd1316 ] and [ cmdclasses ] .
we show the results for two different initial guesses for the mean colors of the blue and red gc populations : one for the pair @xmath31 and 1.2 mag , another one for @xmath32 and 1.4 mag .
these different guesses were chosen to cover : ( 1 ) the range of peak positions between cmds of the various luminosity groups ; and ( 2 ) the range caused by the color - magnitude trend itself ( see figure [ cmdclasses ] ) .
linear slopes between color and absolute magnitude are obtained from ordinary least square fits to the peak positions for both pairs of initial guesses , fitting color as a function of magnitude .
gcs fainter than @xmath33 mag and @xmath34 mag were excluded from the fits , to avoid effects of photometric incompleteness on the measured slopes .
the mean of both fits is adopted as the final best - fit slope .
the fit errors of the slope are derived from re - sampling the points using as dispersion the observed scatter around the fitted relation .
we note that this scatter is for all luminosity groups larger ( by 30% for luminosity group 1 and by 100% for group 4 ) than the average statistical error estimate for the peak position , given by @xmath35 .
here , @xmath36 is the width of the gaussian peak and @xmath37 the average number of gcs per magnitude bin associated to the peak .
we are therefore confident that our error estimates are not too optimistic .
for the few cases where the difference between the results obtained using the two initial guesses was larger than the formal fit errors , we adopted the difference as the error .
the resulting slopes of those fits are shown in table [ fitresults ] .
we first consider @xmath3 vs. @xmath2 for group 1 . in this case
, we detect a very significant slope @xmath38 for the blue peak position . when restricting the fit to @xmath39 mag ( i.e. , rejecting the two brightest magnitude bins in figure [ cmdclasses ] ) , the result is @xmath40 .
this is significant at the 4.5@xmath36 level , indicating that the color - magnitude trend is _ not _ restricted to just the brightest gcs . for m87 and m60 alone ,
a comparable slope is derived , albeit at somewhat lower ( @xmath41 ) significance .
the slope for m49 is consistent with zero , in agreement with the findings of strader @xcite . when excluding m49 from group 1 , the slope for the blue population rises slightly to @xmath42 .
the non - detection of a color - magnitude trend for m49 is intriguing , since this galaxy has a similar number of gcs as m60 and is similar to m60 and m87 in terms of luminosity , color , surface brightness and gc dynamics ( peng 2006 ; ferrarese 2006 ; mei 2005 ; ct 2001 ; 2003 ) .
the fact that m49 is located at the center of its own sub - cluster being offset by @xmath43 from the virgo cluster center may hint at some environmental effect on the strength of the observed color - magnitude trend .
in contrast to the strong correlation @xmath5 exhibited by the blue gcs in group 1 , @xmath5 is insignificant for the red gcs in this group .
however , the situation changes when using @xmath6 instead of @xmath3 as magnitude in the color - magnitude diagram ( see table [ fitresults ] ) .
the slope @xmath44 for the blue population is more than 50% smaller than @xmath5 and , as a result , has a lower level of significance ( 1.7@xmath36 . )
considered individually , @xmath45 of the blue population becomes insignificant for all three galaxies in this group . in turn , the slope @xmath45 of the red population becomes marginally positive i.e. , opposite in sense to that defined by @xmath5 for the blue population . [ cols="<,>,>,>,>",options="header " , ] notes : analogous to table [ fitresults ] .
difference is that peak positions are now defined by the biweight of the color distribution blue- and redwards of a luminosity independent limiting color , see fig .
[ normal ] .
[ slopes ] graphically compares the fits to and biweight peak positions for all four luminosity groups .
the difference in slopes between @xmath12 and @xmath13 raises the question as to what the `` true '' slope actually is . to investigate this issue in more detail ,
we compare in figure [ cmd3brightsim ] the actual measurements for group 1 with two sets of 20 monte - carlo simulations for the m87 cmd .
the first set of cmds has an implemented slope @xmath46 for the blue gcs corresponding to @xmath47 is of course not identical to @xmath5 , but this difference is negligible for the sake of this study : @xmath48 and @xmath49 for the red ones ; the other set has @xmath49 implemented for either gc population .
these artificial cmds were created by assuming the average peak widths derived by for the blue and red gcs , taking into account also the photometric errors on the colors .
the luminosity function is sampled from a kernel estimator of the original cmd of m87 , separately for each color peak .
figure [ compsim ] compares the real and simulated m87 cmd with an implemented slope .
note an important feature of those cmds : the red peak is populated to brighter absolute luminosities in @xmath3 than in @xmath6 , while the blue peak does not show such a difference .
this is because , relative to gcs in the blue population , gcs in the red population are brighter in @xmath3 than in @xmath6 .
table [ fitresults ] gives the slopes fitted to the mean colors from for the two simulated sets of m87 cmds , together with the slopes for the real data .
this comparison shows that both the cmd of m87 alone , and that for the co - added cmd of group 1 , are inconsistent with a magnitude independent color for the blue gcs . the data are consistent with a simulated slope of 0.040 for the blue gcs and no slope for the red gcs . note the positive slope found for the red gcs in both the real and simulated data in @xmath6 .
averaging over the 20 simulated cmds , the mean red slope of @xmath500.03 is different from zero at the 5.7@xmath36 level . that finding is noteworthy since the simulations do _
not _ implement a slope for the red peak .
this exercise therefore demonstrates an important caveat of the fitting : at a given magnitude , as one of the two peaks becomes de - populated ( in this case the red peak in @xmath6 ) , its position will artificially be dragged towards the position of the more prominent peak . as a consequence
, the mean color of the blue population will be artificially shifted towards bluer colors .
this results in a weaker slope for the blue peak in @xmath6 than in @xmath3 . in @xmath3 ,
both peaks are populated up to comparable absolute luminosities , and therefore the fitted slopes for both red and blue peaks correspond to the simulated ones .
that is to say , only the _ weaker _ slope in @xmath6 is caused by a bias .
does not artificially introduce the very significant negative slope in the blue @xmath3 peak : this is clearly a real trend that is intrinsic to the data .
when interpreting the slope as a mass - metallicity relation for gcs in the blue peak , we derive a scaling @xmath51 by combining the slope @xmath52 - 0.037 found in the previous section with @xmath53}}{d(g - z ) } = 5.14~{\rm dex~mag^{-1}}\ ] ] from peng @xcite and assuming a constant mass - to - light ratio over the luminosity range investigated .
this is in good agreement with the scaling relation @xmath54 found by harris @xcite . as an alternative to kmm
, we have calculated the biweight location ( beers 1990 ) for the gc subpopulations blue - wards and red - wards of a magnitude independent limiting color .
this limiting color is the mean of the position of the blue and red populations , weighted by the average width of either peak .
the results for this alternative method are shown in table [ fitresultsnaive ] .
as an example , we show in fig .
[ normal ] the peak positions for groups 1 and 4 in @xmath3 .
generally speaking , the errors for this biweight method decrease significantly compared to the data points , and the results confirm the significant slope found for the blue gc population .
they also confirm that no significant correlation is found for m49 .
table [ fitresultsnaive ] and figure [ slopes ] show that the difference between fits in @xmath6 and @xmath3 is considerably smaller than in the case of kmm .
nevertheless , this alternative method is itself not immune to biases , as is clear when it is applied it to the simulated cmds ( see table [ fitresultsnaive ] and figure [ slopes ] ) . in the case of the blue population exhibiting a negative slope , a fixed limiting magnitude between both populations will systematically decrease the absolute slope of the blue population , and introduce an artificial positive slope for the red .
having examined how specific features of the cmd may influence the fitted color - magnitude relations for the blue and red gc populations , we now investigate the extent to which possible environmental effects may influence the observed trends .
figure [ cmdclasses ] shows cmds for groups 14 with the peaks over - plotted .
derived slopes for the color - magnitude relations in each group are presented in tables [ fitresults ] and [ fitresultsnaive ] .
the dependence of slope on host galaxy magnitude is illustrated in figure [ slopes ] .
the fits in @xmath3 show that the slope of the blue gcs in groups 2 and 3 are indistinguishable from that for group 1 ; in all three groups , the slope is inconsistent with zero at about the 3@xmath36 level .
this indicates that the color - magnitude trend among metal - poor gcs is _ not _ restricted to just the brightest galaxies . only for group 4
does the measured slope of the blue population become consistent with zero .
the biweight fits in the lower panels of figure [ slopes ] paint a similar picture : the slope decreases only gradually from group 1 to 4 , and apparently does not drop to zero .
given that the biweight fits systematically underestimate the blue slope , this is an indication that the color magnitude trend persists over the entire magnitude range of the acsvcs galaxies .
the trend for the red gcs determined using becomes progressively more uncertain for fainter galaxy magnitudes .
this is because of the decreasing fraction of red ( metal - rich ) gcs with decreasing host galaxy luminosity ( peng 2006 ) .
the firmest result from figure [ slopes ] regarding the color - magnitude relation for red gcs is that there is no strong @xmath3 slope for the two brightest groups ; the slopes for the two faintest galaxy groups have large uncertainties .
the large positive slopes found in @xmath6 are particularly prone to biases , a conclusion that is confirmed by looking at the more robust biweight fit results .
the comparison between the biweight fits to the real data and the simulations shows that the slope of the red population is consistent with zero over the full luminosity range .
an interesting feature of the observed color - magnitude relations is illustrated in figure [ radial ] .
here we subdivide the gc samples of groups 1 and 2 by galactocentric radius into equally - sized , inner " and outer " subsamples .
as the dividing radius , we choose @xmath55 , which translates to a physical distance of @xmath56 kpc at the distance of the virgo cluster . in terms of the effective radii , @xmath57 , of the host galaxies in groups 1 and 2 ,
this fixed dividing radius corresponds to @xmath58 0.30.6 @xmath57 ( ferrarese 2006 ) .
tables [ fitresults ] and [ fitresultsnaive ] report the slopes for the different subsamples . for group 1 ,
the slope found for the inner gcs is a factor of @xmath58 23 higher than that for the outer gcs . both the and biweight methods support this conclusion , at significance levels of 2.2@xmath36 and 2.5@xmath36 , respectively . for the combined group 1 + 2 samples ,
the fits still obtain a difference of about a factor of three ( at 2.9@xmath36 significance ) , while the differences are almost insignificant for the biweight fits . in figure [ sizecolor ] , we plot color vs. half - light radius for gcs with @xmath59 mag , separated into inner and outer samples .
there is a small size offset between the inner and outer sample , in the sense that outer gcs are larger by about 10% .
this dependence of @xmath19 on galactocentric radius has already been investigated by jordn @xcite . in figure
[ colorhist ] , we plot the color histograms for two luminosity ranges : in the left panel , we show gcs in the same luminosity range as in figure [ sizecolor ] ; in the right panel , we show gcs in the range @xmath60 mag .
clearly , the gap between the two peaks is larger among the fainter sample of gcs , a consequence of the color - magnitude trend .
furthermore , for the brighter gcs , the inner and outer color histograms appear to differ more than for the fainter gcs .
the ks probability with which the inner and outer sample of the bright gcs have the same underlying distribution is 1% , while it is 14% for the faint sample . indeed , while the bimodality feature is still notable for the outer bright gcs , it is slightly smeared out for the bright , inner gcs
this suggests that the dependence of the color - magnitude relation on galactocentric distance is driven mainly by the bright gcs ( @xmath61 mag ) .
figure [ colorhist ] also shows that the fraction of blue - to - red gcs decreases slightly for smaller galactocentric distance ( see also , e.g. , dirsch 2003 ) .
the corresponding fractions of blue - to - red gcs are 0.45:0.55 and 0.50:0.50 for the inner and outer gcs , respectively .
it is apparent from the color histograms in figure [ colorhist ] that this small effect can not be responsible for the steeper color - magnitude relation found for the inner sample of gcs . before proceeding ,
we pause to compare the distribution of gcs in the cmd with those of other stellar systems : i.e. , the nuclei of acsvcs galaxies , the acsvcs galaxies themselves , and some candidate ultra - compact dwarf galaxies ( ucds ) identified in the course of the acsvcs . figure [ cmdallobjects ] compares the combined cmd of gcs ( red circles ) , compact stellar nuclei ( asterisks ; ct 2006 ) , acsvcs galaxies ( green circles ; ferrarese 2006 ) and ucds ( open squares ; haegan 2005 ) . in this figure , we have converted from @xmath3 to stellar masses using the population synthesis models of bruzual & charlot @xcite under the assumption of a 10-gyr old population and a chabrier imf ( chabrier 2003 ) . for comparison , we also show the color - magnitude relation for early - type galaxies in the fornax cluster from hilker @xcite , after converting from @xmath17 to @xmath2 using the transformation of fukugita @xcite . the majority of the blue gcs are not consistent with an extrapolation of the galaxy color - magnitude relation , nor with that of the nuclei .
moreover , the color - magnitude relations of the nuclei and the galaxies are steeper than that of the gcs .
only the faintest nuclei appear to be roughly consistent with the color - magnitude relation observed for the blue gcs . in any case , we point out that the existence of a color - magnitude trend for the blue gcs results in a paucity of gcs at high luminosities and blue colors : i.e. , @xmath62 mag and @xmath63 mag .
it is difficult to understand how contamination of the observed cmds by some unrecognized stellar system stripped nuclei , ultra - compact dwarfs ( ucds ) , or otherwise can lead to an underpopulation of this region of the cmd .
as a starting point for this discussion , we summarize the two main findings thus far , both of which concern dependencies of the measured color - magnitude slope on external factors : ( 1 ) the slope for the blue gcs decreases gradually towards fainter host galaxy luminosities , whereas there is no significant color - magnitude relation for the red gcs ; and ( 2 ) the slope is steeper for gcs which lie closer to the centers of their host galaxies .
we now discuss a number of scenarios that that might help explain the observed trend : ( 1 ) contamination by super star clusters , stripped nuclei , or ucds ; ( 2 ) accretion of gcs from low - mass galaxies ; ( 3 ) stochastic effects ; ( 4 ) capture of field stars by individual gcs ; and ( 5 ) gc self enrichment . in [ metallicity ] , we briefly discuss how the observed color - magnitude trends appear after transforming into the metallicity - mass plane . [ discussion ] one idea that may explain the slope difference between inner and outer gcs is the presence of overluminous star clusters of intermediate color that may lie close to the galaxy centers . such a putative population would populate the region between the two color sequences at high luminosities . merged super star clusters ( e.g. , fellhauer & kroupa 2002 , 2005 ; kissler - patig 2006 ) could , in principle , be those objects .
they are expected to be more extended than their monolithic counterparts of equal mass ( e.g. , bekki 2004 ; fellhauer & kroupa 2002 ; bastian 2005 ) . however , in the size vs. color plot shown in figure [ sizecolor ] it is the _ outer _
gcs that are slightly larger .
more importantly , there is no offset in the size distribution for the inner gcs with intermediate colors ( i.e. , @xmath64 mag ) .
this makes the presence of a large number of merged star clusters unlikely .
another possibility is the presence of naked " or tidally stripped de , n nuclei , as suggested by harris @xcite ( see also bassino _ et al .
_ 1994 ) . the de , n nuclei of virgo cluster galaxies do exhibit a color - magnitude trend and have colors comparable to those of the blue gcs ( see , e.g. , lotz 2004 ; ct 2006 ; figure [ cmdallobjects ] ) .
ct @xcite find a median half - light radius of @xmath65 pc for nuclei with @xmath59 mag .
this is comparable to the sizes of gcs in this same luminosity range ( figure [ sizecolor ] ) .
stripped nuclei might therefore be confused with
normal " blue gcs at the highest luminosities .
however , as pointed out in [ comparison ] , the overall distribution of gcs with the cmd shows a _ lack _ of bright , blue objects .
very generally , this dearth is not explained by the addition of an extra " population of contaminators , but rather requires an underlying trend for most of the objects .
stripped nuclei would have to dominate the population of blue gcs .
this seems unlikely to us for several reasons .
first , the nuclei luminosity function peaks @xmath58 24 mag brighter than that of gcs ( see lotz 2004 ; , ct 2006 , see also figure [ cmdallobjects ] ) . more seriously , though , is the shear number of nuclei that would be required in this scenario .
observationally , there is an upper limit on the number of naked nuclei brighter than @xmath66 mag from jones _
et al . _
these authors found nine compact objects ucds in their notation with @xmath67 mag in a survey that is almost complete within the central degree of the virgo cluster ( excluding the 1 - 2 surrounding m87 ) .
their nine compact objects are distributed over an area that is @xmath68 100@xmath69 the acs field of view , in qualitative agreement with the predictions by bekki _
et al . _
@xcite for the spatial distribution of tidally stripped de , n nuclei .
therefore , a presence of dozens , or possibly hundreds , of naked nuclei in our acs fields for the most luminous galaxies seems unlikely . a final note regarding the possibility of ucds as contaminators : the two most discussed formation channels for these compact stellar systems are stripped de , n nuclei and merged stellar super - clusters ( e.g. minniti _ et al .
1998 , hilker 1999a , drinkwater _ et al . _
2003 , haegan 2005 , kissler - patig 2006 , mieske 2006 ) . since these two formation channels correspond to the two kinds of contaminators discussed in this section , we conclude that also ucds are unlikely to account for the observed trends .
a population of gcs accreted from galaxies with a range of masses ( e.g. ct 1998 or hilker 1999b ) will exhibit a color - magnitude relation if : ( 1 ) the mean gc color of a single galaxy scales with galaxy mass ; and ( 2 ) more massive gcs preferentially form in more massive galaxies .
indeed , both conditions are satisfied in actual early - type galaxies ( see , e.g. , peng 2005 ; kundu & whitmore 2001 ; jordn _ et al .
_ 2006 in prep . and
references therein ) .
moreover , this simple scenario may offer , at least in principal , a way to qualitatively understand the slope differences for individual galaxies ( e.g. , the non - existence of a correlation for m49 ) as a consequence of different accretion / merger histories . in order to quantify the tilt in the cmd created by the accretion of gcs from low - mass galaxies that do not themselves have color - magnitude relations , we consider the combined gc sample of groups 24 ( see figure [ cmdall ] ) .
we first calculate the mean host galaxy magnitude for each bin , and then calculate the color in each gc bin according to the relation between host galaxy magnitude and mean color of the blue and red gc subpopulations ( table 3 of peng 2006 ) .
the outcome of this exercise is shown in figure [ cmdall ] . as expected
, there is a ( weak ) trend between mean color with luminosity for the gc subpopulations .
this trend is compared to the fits for group 1 .
there is a substantial offset in the sense that the fainter luminosity groups have bluer gc colors than group 1 . in addition , the simulated slope is substantially smaller than measured for the ensemble of galaxies in group 1 .
all in all , we conclude that the observed color - magnitude relations are unlikely to have been created by accretion of gcs from low - mass galaxies , although we can not rule out the possibility that accretion has contributed to the observed trend .
the light of old gcs is in the red optical to near - infrared pass - bands dominated by red giant stars .
generally , the number of red giants is only a small fraction of the total number of stars in a given stellar population .
it is therefore worth discussing to which extent stochastic effects due to small numbers of red giants can skew the probability distribution of the integrated gc luminosity .
this may cause a trend of gc color with integrated luminosity , because any stochastic effect is weaker for higher gc luminosity ( larger total number of stars ) and for bluer bands ( lower average stellar luminosity ) . in the works by cervio ( 2002 ) , and cervio & luridiana ( 2004 ; 2006 ) ,
the effect of discrete isochrone sampling on integrated photometric properties is discussed in detail . in cervio & luridiana ( 2006 ) ,
the width , skewness and kurtosis for the probability distribution of integrated luminosity of a stellar population are calculated as a function of star number and age .
the faint magnitude limit of our gcs ( @xmath70 mag ) corresponds to about 10@xmath71 , and hence about @xmath72 stars . in the @xmath73-band taken to represent the @xmath12 band due to lack of estimates for the latter band
, the photometric uncertainty arising from sampling effects amounts to about 0.06 mag for such a population ( cervio & luridiana ( 2006 ) ; and cervio , private communications ) .
the skewness of the distribution in integrated luminosity is about 0.25 .
the product of these two numbers about 0.015 mag is a measure of the difference between mode and mean of the probability distribution in integrated luminosity . given that the colour - magnitude trend in our data requires an almost ten times larger color shift , stochastic effects do probably not have a significant contribution to the observed trend .
there is also a phenomenological counterargument against the importance of stochastic effects : the color - magnitude trend is shaped more by brighter gcs ( see figure [ colorhist ] ) , while stochastic effects are stronger for fainter gcs .
furthermore , the environmental dependencies of the trend can not be explained by purely intrinsic stochastic effects .
given that the strength of the color - magnitude trend depends on local environment , it is reasonable to examine the extent to which capture of field stars may help to explain the observed slopes .
the capture of field stars has previously been considered by bica @xcite for the case of a 10@xmath74m@xmath75 star cluster orbiting within the galactic bulge at a distance of 1 kpc .
they found that such a cluster could absorb a substantial fraction of its total mass during its lifetime .
similarly , kroupa @xcite has shown that stellar super - clusters created in disks like that of the milky way can capture several tens of percent of their total mass in form of disk stars .
field stars of the brightest early - type galaxies are significantly redder than their blue gc population , typically by about 0.5 to 0.6 mag ( peng 2006 ) . therefore
, there will be a color shift towards the red by about 0.10 - 0.15 for a blue gc by the time that @xmath58 20 - 25% of the gc light originates from field stars .
this effect will be notable mainly for the blue gc population , since the color of red gcs is much closer to that of field stars . as will become clear from the subsequent discussion , the capture rate per unit gc mass is higher for more massive gcs , which will lead to a slope in color - magnitude space .
this qualitatively explains finding ( 1 ) from above .
in addition , the effect will be stronger in the inner part of the host galaxy because of the higher field star density in such regions , consistent with finding ( 2 ) .
thus , field star capture seems qualitatively consistent with the observed trend .
we now quantify this scenario by considering a gc with a mass of 10@xmath76 , a mass that roughly corresponds to the brightest clusters in groups 1 and 2 in figure [ cmdclasses ] .
for an assumed m / l=2.5 , this corresponds to a total of 4@xmath77 solar luminosities .
we assume the gc to be on a circular orbit around m87 .
this is certainly an extreme assumption , given that ct @xcite rule out such tangentially biased orbits for the gc system _ as a whole_. we nevertheless use it for the sake of this simple estimate .
we adopt a radius of 3 kpc , representative for the sample within 5 kpc projected galactocentric distance .
we may estimate the number of captured stars by multiplying the total number of stars within the volume of influence of the gc with the fraction of stars having relative velocities below the gc escape velocity .
as the volume of influence , we define a torus around the orbit of the gc with an inner radius of the gc s tidal radius with respect to its host galaxy m87 . for calculating the tidal radius , we use the mass profile of m87 derived by romanowsky & kochanek @xcite .
they quote the integrated mass at a distance of 3 kpc to be 1.1@xmath6910@xmath78 @xmath79 , which is comparable to the value found by ct @xcite .
together with the 10@xmath76 mass of the gc , this gives a tidal radius of 9 pc .
the total volume of the torus then is 4.8@xmath80 pc@xmath81 .
we use the same paper by romanowsky & kochanek to get the 3-dimensional stellar density of m87 at 3 kpc , which is about 0.06 l@xmath82 pc@xmath83 .
that is , along one orbit there will be a total of 2.9 @xmath77 solar luminosities within the volume of influence of our test gc .
however , only a small fraction of those will be captured : the average escape velocity @xmath84 of the gc within its tidal radius of 9pc is about 37 km s@xmath10 , while the stellar velocity dispersion in m87 is about 275 km s@xmath10 ( romanowsky & kochanek 2001 ) . the gc population in m87 has a velocity dispersion of @xmath68 400 km s@xmath10 .
the probability distribution for the relative velocity between the field stars and a gc is given by the convolution of both distributions , which is a gaussian of width @xmath36 = 485 km s@xmath10 .
the fraction of stars ( in solar luminosities ) that can be captured in the 1d case is given by the area below the gaussian between + /-
@xmath85 , which is 0.035 . for the full 3d case , this is @xmath86 , or one out of 2.3 @xmath87 stars
. given that in total 2.9 @xmath77 stars are in the volume of influence in each orbit , this implies a capture rate of one dozen stars per orbit - body simulations , which is beyond the scope of this paper . ] .
for a keplerian orbit at 3 kpc , one revolution lasts about 2.5@xmath88 years . in a hubble time
, one can therefore capture about 6000 stars .
this is 1.5% of the total assumed gc luminosity , not sufficient to fully explain the observed effects .
only when considering a more extreme orbital distance of 1 kpc does the total number of captured stars increase to about 4@xmath87 , i.e. @xmath8910% of all stars .
we now briefly examine the opposite effect of field star capture : namely cluster star evaporation .
as is well known , mass segregation in gcs ( gunn & griffin 1979 ; jordn 2004 ) will result in evaporation affecting mainly the lowest mass stars .
lamers et al .
@xcite find for clusters of solar metallicity that evaporation will cause the overall gc color to become bluer with time ( over most of their lifetime ) due to the preferential loss of red , low - mass stars .
could such a color change help to explain our observations ?
two effects are important : ( 1 ) the _ maximum _ blueward shift due to evaporation is larger for more massive clusters ( lamers et al .
@xcite ) , reaching up to @xmath90 0.10 mag ; and ( 2 ) the dissolution time , @xmath91 , is longer for more massive clusters . to qualitatively explain our observations one would require the massive gcs to be in an early stage of evolution , so that evaporation has not yet had a significant effect
the fainter gcs would mainly define the color magnitude trend .
however , figure [ colorhist ] shows that the slope changes between inner and outer gcs are mainly driven by the _
brightest _ gcs .
that is , evaporation is unlikely to account fully for the observed color - magnitude trend .
even more serious , evaporation can not naturally explain the simultaneous presence of a trend for the blue gcs and its absence for the red
. generally speaking , self enrichment will occur in any stellar system , gcs included , if the system is able to retain a fraction of the gas expelled by the first generation of supernovae and subsequently form a new generation of stars . to create a metallicity - mass ( or , equivalently , a color - magnitude ) trend via self - enrichment , one would therefore require that more massive gcs are more efficiently self - enriched . the extent to which gcs in the milky way may have self - enriched
, however , is a controversial subject : e.g. , frank & gisler @xcite , smith @xcite , gnedin @xcite , parmentier & gilmore @xcite , dopita & smith @xcite , morgan & lake @xcite , thoul @xcite . in this context , it is interesting to note that the most massive gc in the galaxy , @xmath92 cen , is known to have a metallicity distribution function that shows a long tail towards high metallicities ( e.g. , norris , freeman & mighell 1996 ; stanford _
various studies have found this high - metallicity population to be either coeval , or perhaps 24 gyr younger , than the cluster s primary ( metal - poor ) population ( e.g. , ferraro 2004 ; sollima et al .
2005 ; hughes & wallenstein 2000 ; hilker & richtler 2000 ; rey 2004 , stanford 2006 ) .
this is consistent with @xmath92 cen having been at least partially self - enriched .
note , however , that because of its peculiar star formation history , numerous investigators have suggested that the cluster may be the stripped nucleus of a now - dissolved dwarf galaxy ( e.g. , freeman 1993 ; hilker & richtler 2000 ; bekki & freeman 2003 ) .
if true , then it may be incorrect to consider @xmath92 cen to be a typical example of a self enriched gc .
this caveat aside , if self enrichment is common to the formation of the most massive gcs , then it may indeed help to explain our observations .
most significantly , the lack of a correlation among the red gc subpopulation is naturally explained in this scenario : i.e. , the redder gcs are already metal - enriched , and a fixed amount _ d_z of supernovae ejecta will cause a relatively weaker metallicity increase , _
d_[fe / h ] , for metal - rich gcs .
for instance , for a system with [ fe / h ] = @xmath932 ( z = 0.0002 ) , adding a fixed amount of metals _
d_z = 0.0002 will result in _ d_[fe / h][email protected] dex . according to peng @xcite this corresponds to a shift in @xmath2 of about 0.06 mag and is about the observed amount of the color magnitude trend .
however , for a system with [ fe / h ] = @xmath930.5 and the same _ d_z , we only have _ d_[fe / h][email protected] dex , with a corresponding color change in @xmath2 well below 0.01 mag .
a possible shortcoming of this self - enrichment scenario is that is does not explicitly account for environmental effects such as a dependence on galactocentric radius . on the other hand , feedback induced by pressure - confinement from the surrounding medium may be able to help reconcile this scenario with observations . to conclude this discussion ,
we show in figure [ met ] a metallicity - luminosity plot for the co - added sample of gcs in group 1 . to estimate gc metallicities , we have applied a quartic color - metallicity transformation based on the empirical calibration data from peng @xcite .
note that the resulting metallicity distribution is _ not _ bimodal , but rather shows a single pronounced peak at [ fe / h]@xmath89 @xmath930.4 dex and a long , extended tail towards lower metallicities .
this confirms recent warnings ( e.g. , richtler 2005 ; yoon , yi & lee 2006 ) that bimodality in color does not necessarily imply bimodality in metallicity .
using the data in figure [ met ] , we can compare the metallicity distributions of gcs brighter and fainter than @xmath95 mag .
for the metal - poor population , ( [ fe / h]@xmath96 - 0.6 ) , the fainter gcs have a significantly broader tail towards lower metallicities than do the bright ones .
a ks - test gives a 0.1% probability for a common underlying distribution between both samples . in the metal - rich part of the diagram ,
the agreement is very good , with a ks probability of 91% . in other words ,
the empirical finding of a color - magnitude trend for the blue gcs can hence be re - phrased in terms of metallicity : the gc systems appear to have a skewed , but unimodal , metallicity distribution function whose metal - poor tail becomes increasingly extended at lower luminosities / masses .
in this paper we have analysed the color - magnitude relation for the gc subpopulations in early - type galaxies belonging to the virgo cluster .
our sample consists of 79 galaxies , spanning the range @xmath0 mag , drawn from the acs virgo cluster survey ( ct 2004 ) .
our principal findings can be summarized as follows : 1 .
based on fits , we find a highly significant correlation @xmath1 for the blue gcs in the co - added cmd of the three brightest virgo cluster galaxies ( m49 , m87 , m60 ) .
the sense of the correlation is such that brightest gcs in this subpopulation are redder than their faint counterparts . for m87 and m60 alone
, we find comparable correlations ; by contrast , the blue gcs in m49 do not appear to follow such a trend . in no galaxy
do we find a significant correlation between @xmath2 and @xmath3 for the gcs associated with the red subpopulation .
2 . based on fits ,
we find the slope @xmath97 within the blue subpopulation to be much weaker than @xmath5 . in turn
, we find a mild _ positive _ correlation @xmath45 for the red subpopulation , in the opposite sense to what is found for the blue subpopulation .
3 . we test on artificial cmds with , and without , implemented slopes for the blue subpopulation , and no slope for the red population .
correctly reproduces the implemented slopes when using the cmd @xmath2 vs. @xmath3 . for @xmath2 vs. @xmath6 , gives a biased estimate for the color of the blue population and yields a positive slope for the red subpopulation , although no slope had been implemented .
we trace this bias to the fact that in @xmath6 unlike in @xmath3 the luminosity function of blue gcs is populated to significantly brighter luminosities than for the red gcs .
we conclude that for the real data , the strong slope @xmath5 is _ not _ an artifact of the fitting method , while the weaker slope @xmath45 is an artificial feature of the fit .
we derive correlations independent of by using a biweight estimator for gcs divided on the basis of a luminosity independent color .
the results confirm the significant slope found by in @xmath3 and reveal no significant discrepancies between the slope measured using @xmath3 and @xmath6 .
5 . the slope measured from the blue subpopulation decreases with host galaxy luminosity . however , even among the faintest galaxies ( @xmath30 mag ) the slope is nonzero .
the color - magnitude relation exhibited by the blue gcs is stronger for clusters with smaller projected galactocentric distances .
this difference is driven mainly by gcs with @xmath61 mag .
we examine a number of physical mechanisms that might give rise to a color - magnitude relation with these characteristics : ( 1 ) contamination by super - star - clusters , stripped galactic nuclei , or ultra - compact dwarf galaxies ; ( 2 ) accretion of gcs from low - mass galaxies ; ( 3 ) stochastic effects ; ( 4 ) capture of galactic field stars by gcs ; and ( 5 ) self - enrichment of individual gcs .
neither contaminants nor accretion of low - mass galaxies appear likely to fully explain our observations .
self - enrichment in more massive gcs may be able to account for the observed color - magnitude relation , although more work is needed to determine if the dependence on galactocentric radius can be explained within the context of the self - enrichment scenario .
although the capture of field stars _ does _ lead to a dependence on galactocentric radius that is at least qualitatively in agreement with our observations , this mechanism does not appear efficient enough to explain the overall size of the observed trend . while none of these scenarios can fully account for the full set of observations in a straightforward manner , self - enrichment and capture of field stars seem the most promising approaches .
there is , of course , no _ a priori _ reason that a single mechanism is responsible for the observed color - magnitude relation , and conceptually , the combined effects of self - enrichment and field star capture may be able to explain the observations : i.e. , field star capture could act as a second order effect , producing environmental variations of a global mass - metallicity relation induced by self - enrichment .
this would alleviate the discrepancy between real and required field star capture rate estimated in this work . dedicated n - body simulations going beyond these simple estimates
will be required to accurately address the efficiency of field star capture in different environments .
[ conclusions ] we would like to thank pavel kroupa , mike fellhauer , holger baumgardt , h.j.g.l.m .
lamers and miguel cervio for fruitful discussions .
we thank the anonymous referee for her / his comments .
support for program go-9401 was provided through a grant from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas5 - 26555 .
p.c . acknowledges support provided by nasa ltsa grant nag5 - 11714 .
m.j.w . acknowledges support through nsf grant ast-0205960 .
s.m . acknowledges additional support from nasa grant nag5 - 7697 to the acs team . d.m . acknowledges support provided by nsf grants ast-0071099 , ast-0206031 , ast-0420920 and ast-0437519 , by nasa grant nng04gj48 g , and by grant hst - ar-09519.01-a from stsci .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
ashman , k. m. , zepf , s. e. 1992 , apj , 384 , 50 ashman , k. m. , bird , c. m. , zepf , s. e. 1994 , aj , 108 , 2348 bastian , n. , emsellem , e. , kissler - patig , m. , maraston , c. 2006 , a&a , 445 , 471 bassino , l. p. , muzzio , j. c. , & rabolli , m. 1994 , , 431 , 634 beasley , m. a. _ et al .
_ 2002 , mnras , 333 , 383 beers , t. c. , flynn , k. , gebhardt , k. 1990 , aj , 100 , 32 bekki , k. , couch , w.j .
, drinkwater , m.j . ,
shioya , y. , 2003a , mnras , 344 , 399 bekki , k. , & freeman , k. c. 2003 , mnras , 346 , l11 bekki , k. , couch , w.j . ,
drinkwater , m.j . ,
shioya , y. , 2004 , apjl , 610 , 13 bica , e. _ et al .
_ 1997 , apj , 482 , 49 bower , r.g . ,
lucey , j.r . , ellis , r.s .
1992 , mnras , 254 , 589 bruzual , g. , charlot , s. 2003 , mnras , 344 , 1000 burgarella , d. , kissler - patig , m. , buat , v. 2001 , aj , 121 , 2647 cervio , m. , valls - gabaud , d. , luridiana , v. , & mas - hesse , j. m. 2002 , a&a , 381 , 51 cervio , m. , luridiana , v. 2004 , a&a , 413 , 145 cervio , m. , luridiana , v. 2006 , a&a , 451 , 475 chabrier , g. , 2003 , pasp , 115 , 763 ct , p. , marzke , r. o. , west , m. j. 1998 , apj , 501 , 554 ct , p. _ et al .
_ 2001 , apj , 559 , 828 ct , p. , mclaughlin , d. e. , cohen , j. g. , & blakeslee , j. p. 2003
, apj , 591 , 850 ct , p. _ et al .
_ 2004 , apjs , 153 , 223 ( paper i ) ct , p. _ et al .
_ 2006 , apjs , accepted , astro - ph/0603252 ( paper viii ) dirsch , b. _ et al .
_ 2003 , aj , 125 , 1908 dopita , m. a. , smith , g. h. 1986 , apj , 304 , 283 drinkwater , m.j . ,
gregg , m.d . ,
hilker , m. et al . , 2003 , nature , 423 , 519 fellhauer , m. , kroupa , p. , 2002 ,
mnras , 330 , 642 fellhauer , m. ; kroupa , p. , 2005 ,
mnras , 359 , 223 ferraro , f.r .
_ et al .
_ 2004 , apj , 603 , l81 ferrarese , l. _ et al .
_ 2006 , apjs , 164 , 334 ( paper vi ) freeman , k. c. 1993 , in asp conf .
48 , the globular cluster - galaxy connection , ed .
g. h. smith & j. p. brodie ( san francisco : asp ) , 27 forbes , d. a. , brodie , j. p. , grillmair , c. j. 1997 , aj , 113 , 1652 ford , h. c. , bartko , f. , bely , p. y. _ et al .
_ 1998 , proc .
3356 , p. 234 - 248 ,
space telescopes and instruments v , pierre y. bely ; james b. breckinridge ; eds .
frank , j. , gisler , g. 1976 , mnras , 176 , 533 fukugita , m. , ichikawa , t. , gunn , j.e .
1996 , aj , 111 , 1748 gebhardt , k. , kissler - patig , m. 1999 , aj , 118 , 1526 gnedin , o. y. _ et al .
_ 2002 , apj , 568 , 23 gunn , j. e. , griffin , r. f. 1979 , aj , 84 , 752 harris , g. l. h. , harris , w. e. , geisler , d. 2004 , aj , 128 , 723 harris , w. e. _ et al .
_ 2006 , apj , 636 , 90 hasegan , m. , jordn , a. , ct , p. et al .
2005 , apj , 627 , 203 hilker , m. , infante , l. , vieira , g. , kissler - patig , m. , richtler , t. 1999 , a&as , 134 , 75 hilker , m. , infante , l. , richtler , t. 1999 , a&as , 138 , 55 hilker , m. , & richtler , t. 2000 , a&a , 362 , 895 hilker , m. , mieske , s. , & infante , l. 2003 , a&al , 397 , l9 hughes , j. , & wallerstein , g. 2000 , aj , 119 , 1225 jones , j. b. , _
_ 2006 , aj , 131 , 312 jordn , a. , blakeslee , j. p. , peng , e. w. _ et al .
_ 2004a , apjs , 154 , 509 ( paper ii ) jordn , a. , ct , p. , ferrarese , l. _ et al .
_ 2004b , apj , 613 , 279 ( paper iii ) jordn , a. 2004 , apj , 613 , l117 jordn , a. , ct , p. , blakeslee , j. _ et al .
_ 2005 , apj , 634 , 1002 ( paper x ) karick , a. , drinkwater , m.j . , gregg , m.d .
2003 , mnras , 344 , 188 kissler - patig , m. 1997 , a&a , 319 , 83 kissler - patig , m. , jordn , a. , bastian , n. 2006 , a&a , 448 , 1031 kravtsov , a. v. , gnedin , o. y. 2005 , apj , 623 , 650 kroupa , p. 1998 , mnras , 300 , 200 kundu , a. , whitmore , b. 2001 , aj , 121 , 2950 lamers , h.j.g.l.m , anders , p. , de grijs , r. 2006 , a&a , 452 , 131 larsen , s. s. , brodie , j. p. , huchra , j. p. , forbes , d. a. , grillmair , c. j 2001 , aj , 121 , 2974 lotz , j. m. , miller , b. w. , ferguson , h. c. 2004 , apj , 613 , 262 mei , s. _ et al .
_ 2005 , apj , 625 , 121 ( paper v ) mei , s. _ et al .
_ 2006 , submitted to apj ( paper xiii ) mieske , s. , hilker , m. , infante , l. & jordn , a. 2006 , aj 131 , 2442 minniti , d. , kissler - patig , m. , goudfrooij , p. , meylan , g. 1998 , aj , 115 , 121 morgan , s. , lake , g. 1989 , apj , 339 , 171 norris , j. e. , freeman , k.c .
, mighell , k.j .
1996 , apj , 462 , 241 odenkirchen , m. , grebel , e. k. , dehnen , w. 2003 , aj , 126 , 2385 ostrov , p. g. , forte , j. c. , geisler , d. 1998 , aj , 116 , 2854 parmentier , g. , gilmore , g. 2001 , a&a , 378 , 97 peng , e. , _ et al .
_ 2006 , apj , 639 , 95 ( paper ix ) pryor , c. , mcclure , r. d. , fletcher , j. m. , hesser , j. e. 1989 , aj , 98 , 596 rey , s. c. _ et al .
_ 2004 , aj , 127 , 958 richardson , s. , & green , p. g. 1997 , jr .
b , 1997 , 59 , 731 richtler , t. 2005 , in press at bull .
india , astro - ph/0512545 romanowsky , a. , & kochanek , c.s .
2001 , apj 553 , 722 schlegel , d.j . ,
finkbeiner , d.p . , &
davis , m. 1998 , apj , 500 , 525 searle , l , zinn , r. 1978 , apj , 225 , 357 smith , g.h .
1996 , pasp , 108 , 176 sollima , a. _ et al .
_ 2005 , apj , 634 , 332 spitler , l. r. , _ et al .
_ 2006 , aj accepted , astro - ph/0606337 stanford , l. m. , da costa , g. s. , norris , j. e. & cannon , r. d. 2006 , apj , in press , astro - ph/0605612 strader , j. , brodie , j.p . , spitler , l. , beasley , m.a .
2006 , aj submitted , astro - ph/0508001 thoul , a. _ et al .
_ 2002 , a&a , 383 , 491 tonry , j.l . , dressler , a. , blakeslee , j.p .
2001 , apj , 546 , 681 west , m. j. , ct , p. , marzke , r. o. , jordn , a. 2004 , nature , 427 , 31 whitmore , b. c. _ et al .
_ 1995 , apj , 454l , 73 yoon , s. , yi , s. k. , lee , y. 2006 , science , 311 , 1129 | we examine the correlation between globular cluster ( gc ) color and magnitude using hst / acs imaging for a sample of 79 early - type galaxies ( @xmath0 mag ) with accurate surface - brightness fluctuation distances from the acs virgo cluster survey . using the mixture modeling algorithm , we find a highly significant correlation , @xmath1 , between color and magnitude for the subpopulation of blue gcs in the co - added gc color - magnitude diagram of the three brightest virgo cluster galaxies ( m49 , m87 and m60 ) .
the sense of the correlation is such that brighter gcs are redder than their fainter counterparts . for the single gc systems of m87 and m60 , we find similar correlations ; m49 does not appear to show a significant trend .
there is no correlation between @xmath2 and @xmath3 for gcs belonging to the red subpopulation .
the correlation @xmath4 for the blue subpopulation is much weaker than @xmath5 . using monte carlo simulations
, we attribute this finding to the fact that the blue subpopulation in @xmath6 extends to higher luminosities than does the red subpopulation , which biases the fit results .
the highly significant correlation between color and @xmath3 , however , is a real effect : this conclusion is supported by biweight fits to the same color distributions .
we identify two environmental dependencies which influence the derived color - magnitude relation : ( 1 ) the slope of the color - magnitude relation decreases in significance with decreasing galaxy luminosity , although it remains detectable over the full luminosity range of our sample ; and ( 2 ) the slope is stronger for gc populations located at smaller galactocentric distances .
these characteristics suggest that the observed trend is , at least partially , shaped by external agents .
we examine several physical mechanisms that might give rise to the observed color - magnitude relation including : ( 1 ) presence of contaminants like super - clusters , stripped galactic nuclei , or ultra - compact dwarfs ; ( 2 ) accretion of gcs from low - mass galaxies ; ( 3 ) stochastic effects ; ( 4 ) the capture of field stars by individual gcs ; and ( 5 ) gc self - enrichment .
although none of these scenarios offers a fully satisfactory explanation of the observations , we conclude that self - enrichment and field - star capture , or a combination of these processes , offer the most promising means of explaining our observations . |
the siegel - weil formula identifies the integral of a certain theta function as an eisenstein series on a symplectic group . in modern language , this is an example of a global theta correspondence for the dual pair @xmath15 in which the trivial representation of @xmath16 corresponds to an automorphic representation of @xmath17 generated by the values of eisenstein series .
the group @xmath16 has many other one dimensional automorphic representations ( automorphic characters ) and it is an open problem to describe where they all go in the correspondence
. this paper deals with the dual pair , @xmath18 .
the local and global versions of the theta correspondence , in this case , were studied by waldspurger @xcite @xcite and he obtained very explicit information about corresponding representations .
however he did not consider global one dimensional representations of @xmath0 .
our main results describes explicitly the lifts of automorphic characters @xmath19 of @xmath20 in terms of theta series coming from various @xmath21 s , these results provide a complement to the work of waldspurger and are analagues to the seigel - weil formula .
let @xmath22 be a number field and let @xmath23 be a quaternion algebra over @xmath22 . for any place @xmath24 of @xmath22 ,
let @xmath25 be the completion of @xmath22 , and let @xmath26 .
the local algebra @xmath27 can be either split , @xmath28 or non - split , so that @xmath27 is a division algebra .
the invariant @xmath29 is defined as @xmath30 when @xmath27 is split and @xmath31 when @xmath27 is non - split . @xmath32
let @xmath33 by the product formula @xmath32 the cardinality of @xmath34 , is even .
the isomorphism class of @xmath23 is determined uniquely by @xmath34 .
let @xmath3 be the trace zero elements of @xmath23 , @xmath35 and equip @xmath3 with a quadratic form @xmath36 , where @xmath37 is the reduced norm .
the space @xmath3 is thus a three dimensional quadratic space over @xmath22 and , up to scaling @xmath38 , all three dimensional spaces over @xmath22 are constructed this way . since the dimension of @xmath3 is odd , the orthogonal group factors as @xmath39 .
moreover , the group @xmath40 acts on @xmath3 by conjugation and the action gives rise to the exact sequence @xmath41 where @xmath42 is the center of @xmath43
. we can thus identify @xmath43 with @xmath44 let @xmath45 be the adeles of @xmath22 and @xmath46 be the ideles of @xmath22 .
let @xmath9 be a character of @xmath47 such that @xmath48 .
such a character is given by @xmath49 for some @xmath50 where the coset @xmath51 in @xmath52 is uniquely determined and where the global hilbert symbol @xmath53 is defined by the local hilbert symbols @xmath54 . since @xmath55 is the map @xmath56 the composition @xmath57 is a quadratic character on @xmath58 which is trivial on @xmath59 .
hence it is an automorphic character of @xmath58 which descends to @xmath60 .
all automorphic characters of @xmath60 are obtained in this way .
let @xmath1 , be the @xmath61-fold metaplectic cover of @xmath62 fix a nontrivial additive character @xmath63 . then @xmath1 acts on the schwartz space @xmath64 by the weil representation @xmath65 determined by @xmath66 .
the action of @xmath1 commutes with the linear action of @xmath58 on @xmath64 given by @xmath67 . using the commuting actions , we define for @xmath68 and @xmath69 the theta function @xmath70 this function
is left invariant by @xmath71 .
we now assume that @xmath72 is a division algebra , i.e. @xmath73 . in this case
, the orthogonal group @xmath74 is anisotropic and the quotient @xmath75 is compact .
thus , for a quadratic character @xmath9 as above , the integral @xmath76 is absolutely convergent and defines an automorphic form on @xmath1 .
the space @xmath77 is a subspace of @xmath78 where @xmath78 is the space of genuine automorphic forms on @xmath1 .
furthermore , the space @xmath79 is right invariant under the action of @xmath1 by right translation of the functions in @xmath78 .
our main result is a description of the space @xmath79 for all nontrivial automorphic characters @xmath9 .
this description will be given in terms of another theta correspondence which we outline next .
let @xmath5 be the one dimensional space with a quadratic form @xmath80 .
the group @xmath81 is just @xmath82 and the adelic group @xmath83 is the unrestricted direct product .
the group @xmath84 is compact in the usual product topology and @xmath85 is embedded diagonally in @xmath86 a continuous character @xmath12 of @xmath84 determines a set of local characters @xmath87 of @xmath88 by restriction to the subgroup @xmath89 and continuity of @xmath12 implies that at almost all places @xmath90 , the trivial representation of @xmath88 . at the places where @xmath91 it is the sign character @xmath92 . let @xmath93 the automorphy condition on @xmath12 i.e. @xmath94 implies that @xmath95 is even .
conversely , for any finite set of places @xmath34 where @xmath95 is even there is a unique automorphic character @xmath96 the group @xmath1 acts on the schwartz space @xmath97 by the weil representation @xmath98 as in the case above , the linear action of @xmath84 given by , @xmath99 commutes with the action of @xmath1 on @xmath97 . using the commuting action ,
we define for @xmath100 the theta function @xmath101 as above , this function is left invariant by @xmath102 and for an automorphic character @xmath12 of @xmath84 , we define the automorphic form on @xmath1 @xmath103 note that in this case , @xmath84 is compact so the integral is absolutely convergent .
the space @xmath104 is a right @xmath1 invariant subspace of @xmath105 .
such spaces for varying @xmath5 and @xmath12 , are colloquially referred to as coming from @xmath21 .
we can now state our main result .
main theorem[main ]
let @xmath23 be a quaternion algebra over a totally real field @xmath22 let @xmath9 be a quadratic character of @xmath47 given by @xmath106 , with @xmath50 .
assume that @xmath19 and that @xmath73 so that @xmath23 is a division algebra .
let @xmath79 be the corresponding automorphic representation of @xmath1 constructed above .
then @xmath79 is cuspidal irreducible and @xmath107 where @xmath5 is the one dimensional space with quadratic form @xmath108 and @xmath109 is the automorphic representation of @xmath84 determined by @xmath34 .
the case of @xmath110 is covered by the classical siegel - weil formula . in that case
@xmath111 is given by eisenstein series whereas in our case @xmath79 is a space of cusp forms .
one similarity between our result and the siegel weil formula is that in the classical case , the theta integral attached to @xmath14 is identified with an explicit eisenstein series constructed from @xmath14 .
similarly , we are able to construct a function @xmath11 from @xmath14 and identify the theta lift attached to @xmath14 with a theta lift attached to @xmath13 .
this is made precise in the following [ matching ] suppose that @xmath112 and corresponding @xmath113 are as in the main theorem .
then there is a unique map @xmath114 such that @xmath115 for all @xmath116 .
more explicitly , for factorizable functions @xmath117 the map @xmath118 is given by @xmath119 where @xmath120 is defined on @xmath121 by @xmath122 we will say that the pair of functions @xmath123 in theorem [ matching ] are matching . in fact theorem [ main ] follows from theorem [ matching ] and the later is proved via a matching of whittaker functions .
it turns out that in our case , the automorphic form @xmath7 is distinguished @xcite , i.e. , involves only one square class of fourier coefficients @xmath124 .
that square class is the one where @xmath125 , and @xmath126 note that although many authors ( e.g. gelbart and piatetski shapiro @xcite for @xmath127 and vignras , for classical automorphic forms , ) proved that `` distinguished automorphic forms come from @xmath21 '' , their results do not apply in our case . the automorphic forms in @xmath128 are also distinguished .
for any @xmath11 the fourier expansion of @xmath10 is @xmath129 where , in fact , @xmath130 . to establish theorem [ matching ] for a given @xmath14 it suffices to construct @xmath13 such that @xmath131 for all @xmath116 .
thus the problem of matching reduces to a problem of matching the whittaker functions @xmath132 then , we are reduced to the problem of local matching : given @xmath133 , find @xmath134 such that @xmath135 for all @xmath136 .
the analysis of the local matching depends on the bruhat decomposition @xmath137 in the following way .
the equality in equation for @xmath138 determines @xmath120 uniquely for @xmath139 by the formula @xmath140 there are then two issues .
first we must show that @xmath13 defined on @xmath121 as in equation can be extended to a schwartz function on @xmath25 .
then , with this schwartz function @xmath141 , we must show that the equality in holds on the big cell @xmath142 as well and hence on all of @xmath143 .
the calculation that establishes the local matching on the big cell amounts to establishing an identity involving orbital integrals , gauss sums and weil indexes . for the global identity
we must also check the fundamental lemma which says that , in the unramified situation , the standard unramified vector @xmath144 is matched by the standard unramified vector @xmath145 .
one technical point which should be noted is that the usual theta correspondence is between @xmath16 and @xmath1 whereas we have only discussed automorphic characters of @xmath146 .
however , in our situation , @xmath147 where @xmath148 and this product allows us to extend our results to @xmath16 .
for an automorphic character @xmath149 of @xmath150 , let @xmath151 where @xmath152 the space @xmath153 is the usual theta lift of the automorphic character @xmath154 of @xmath16 and as a consequence of our calculation we can determine it .
[ ue ] there is a unique character @xmath149 of @xmath150 such that @xmath155 and @xmath156 for @xmath157 .
more precisely , @xmath158 with @xmath159 a final note about the type of representation we get is that the local representations coming from @xmath21 which we match are the irreducible even or odd components of the weil representation for a one dimensional quadratic space .
in fact , what we show is that , when @xmath160 , then in the local matching , @xmath161 is matched by an even @xmath162 and , when @xmath163 , then @xmath161 is matched by an odd @xmath162 .
this means that the local component @xmath164 of @xmath128 is the image of the irreducible representation of @xmath1 on either the space of even schwartz functions @xmath165 or the space of odd schwartz functions @xmath166 according to the invariant of @xmath23 . by theorem
[ matching ] the local component @xmath167 of @xmath79 coincides with @xmath164 .
therefore the local components @xmath167 and the global representation @xmath79 are exactly the ones coming from @xmath21
. in @xcite waldspurger considers the space @xmath168 which is the orthogonal complement in @xmath169 to the representation coming from @xmath21 which he calls the elementary weil representations .
these elementary representations are exactly the ones that correspond to @xmath79 in our case and so our result complements the classification carried out by waldspurger .
our result can be applied to the question of representations by spinor generas to obtain results similar to those of r. schulze - pillot in @xcite .
details will be included in a future article .
this paper is based on a thesis directed by stephen kudla who conceived of the idea and directed me in carrying it out .
i would like to take this opportunity to thanks him for all his help .
thanks are also due to jeff adams for advice on the archemedian places and to larry washington for advice on gauss and kloosterman sums .
all errors or misstatements that remain are the responsibility of me alone .
when @xmath25 is non archimedean we denote the ring of integers by @xmath170 and a generator of the maximal subring @xmath171 of @xmath170 by @xmath172 .
let the modular degree of @xmath25 over @xmath173 be @xmath174 so that @xmath175 fix @xmath66-a non trivial additive character of @xmath45 which is trivial on @xmath22 .
the local components of @xmath66 , will be denoted by @xmath176 except when it is clear from context that @xmath66 means its local component .
the character @xmath66 is defined in terms of the following parameters .
we can define a map @xmath177 as the composition of the maps @xmath178 . where the map from @xmath25 to @xmath173 is the trace map .
we then define characters of @xmath25 @xmath179 the product of the local characters @xmath180 is a character of @xmath181 and our fixed character @xmath66 and its local factors are defined as @xmath182 for some @xmath183 . for @xmath184 , let @xmath185 . then @xmath186 is the space of all non trivial characters on @xmath187 .
the conductor of @xmath188 is the smallest power of @xmath171 on which @xmath188 is trivial .
it is equal to the inverse different @xmath189 .
the differental exponent is @xmath190 , it is defined such that @xmath191 .
note that the conductor of @xmath66 is @xmath192 .
@xmath23 is a quaternion algebra over @xmath22 , and @xmath193 .
the invariant @xmath194 if @xmath195 and @xmath163 if @xmath27 is the ( unique up to isomorphism ) division quaternion algebra over @xmath25 . the isomorphism class of @xmath23 is uniquely determined by the set of places @xmath196 .
@xmath3 is the 3-dimensional @xmath22-vector space @xmath197 we also define @xmath198 and @xmath199 .
we will often abuse notation and write @xmath3 instead of @xmath200 when the meaning of @xmath3 is clear from context .
the group @xmath43 is the group of invertible quaternions @xmath201 the group @xmath202 is the centralizer of @xmath203 @xmath204 the standard definitions of schwartz spaces are [ scdef ] when @xmath3 is defined over a non archimedean field , the schwartz space @xmath205 is the set of locally constant compactly supported functions on @xmath3 .
when @xmath3 is defined over an archimedean field , the schwartz space @xmath205 is the set of infinitely differentiable rapidly decreasing functions .
when @xmath3 is split and archimedean , the `` restricted schwartz space '' is the subspace of functions of the form @xmath206 where @xmath207 is a polynomial .
when @xmath3 is non split and archimedean , the `` restricted schwartz space '' is the space of functions @xmath208 where @xmath207 is a polynomial and @xmath209 .
this section concludes with some basic geometric properties of @xmath23 .
[ comp ] if @xmath210 then @xmath211 is not compact and if @xmath212 then @xmath211 is compact .
the quotient @xmath213 is compact unless @xmath214 .
this follows from corollary 3.2 in @xcite .
the product of the reduced norms is isotropic if and only if the reduced norm is isotropic at every @xmath24 .
[ ext ] for @xmath215 we have @xmath216 lemma [ ext ] also holds over all local fields @xmath25 and over the ring of adeles , @xmath217 .
[ compactness ] if @xmath218 then @xmath219 if and only if @xmath220 is compact .
the group @xmath1 is the unique non trivial 2-fold topological central extension of @xmath221 @xmath222 group elements are realized as pairs @xmath223 .
multiplication is defined by : @xmath224 where @xmath225 is a 2-cocycle on @xmath221 .
the generators of @xmath1 are : @xmath226 when calculating the action of group elements by the weil representation ( see below ) we will drop the @xmath227 from the notation .
recall that we have fixed a division quaternion algebra @xmath23 with associated quadratic space @xmath3 and hence indentified @xmath229 .
the actions of @xmath1 and @xmath58 on @xmath230 are the following : @xmath58 acts on elements of @xmath231 by conjugation and on elements of @xmath230 by acting on their arguments i.e. for @xmath232 , and @xmath233 , the action is @xmath234 the weil representation of @xmath1 as described in kudla s notes ( @xcite p. 32
- 34 ) is determined by an additive character @xmath66 of @xmath217 which we fixed in chapter [ dtl ] and by the quadratic space @xmath235 , determined by the algebra @xmath72 .
the formulas for the actions of the generators of @xmath1 on schwartz functions are : @xmath236 here @xmath237 is the quotient of the weil indices @xmath238 . see the appendix in @xcite for the definitions and properties of the weil index . also , @xmath239 is the determinant of the matrix of the quadratic form on @xmath3 and @xmath240 where @xmath241 is the hasse invariant of @xmath3 , see @xcite .
the measure @xmath242 in equation is the unique haar measure which makes the transform @xmath243 self dual . for more details
see @xcite . for the 1-dimensional quadratic space @xmath244 with @xmath245 the weil representation @xmath246 of @xmath1 , commutes with the action of @xmath84.this representation is given by the formulas @xmath247 as above @xmath237 is the quotient of the weil indices @xmath238 and @xmath248
our matching of the spaces @xmath79 and @xmath128 will be reduced to matching their whittaker models so we first realize the two spaces as such .
we will now calculate the fourier expansion of functions @xmath249 .
this calculation will show that , like @xmath128 , the space @xmath79 is distinguished .
we start by writing the automorphic forms as sums . @xmath250
we make the convention that @xmath251 should be read as `` a square '' and @xmath252 be read as a non zero square .
as mentioned in sections [ quat ] , we exclude the case of the unramified global algebra @xmath253 and therefore by lemmas [ comp ] and [ compactness ] the quotient @xmath75 is compact and the integral in equation is absolutely convergent . also , @xmath3 is anisotropic so that the norms of @xmath215 are not equal to a negative of a non zero square .
therefore @xmath254 is not equal to a non zero square .
the term @xmath255 in the sum in equation is zero .
the space @xmath3 is anisotropic so the only point of norm 0 is 0 and the the @xmath255-term in the sum is the integral of the character @xmath57 on the compact space @xmath75 so the term is zero .
the integral in equation can be switched with the summation since the domain of integration is compact and the function has compact support .
thus we have @xmath256 where @xmath257 is some point such that @xmath258 notice that the summation over the set @xmath259 is replaced by the sum over @xmath260 . since @xmath23 is a division algebra and @xmath3 is anisotropic @xmath261 acts transitively on @xmath262 and
so the choice of @xmath263 does not matter .
also , since @xmath9 is trivial on the rational points @xmath264 for @xmath265 so we can include @xmath266 in the argument of @xmath267 .
@xmath268 we now combine the inner sum and integral by a change of variables @xmath269 .
@xmath270 the next step is to factor @xmath271 with @xmath272 and , @xmath273 .
this gives , @xmath274 to proceed , we need the following description of @xmath275 .
[ ker ] let @xmath9 is a nontrivial quadratic character of @xmath276 given by the hilbert symbol @xmath277 and @xmath278 .
let @xmath279 with @xmath257
@xmath280 if and only if @xmath281 2 .
@xmath282 if and only if @xmath283 the first result is contained in proposition 4.8.2 in @xcite when we notice that since @xmath284 the kernel of @xmath285 is the set @xmath286 and since @xmath287 , @xmath288 where @xmath289 .
the second statement follows from the fact that if @xmath290 and @xmath291 are quadratic extensions of @xmath22 then they determine characters @xmath292 and @xmath293 , where @xmath294 , and @xmath295 if and only if @xmath296 . [ ramtriv ] if @xmath297 for some place @xmath298 , the set of ramified places for @xmath23 then @xmath299 .
if @xmath285 is trivial at a ramified place @xmath24 , the square class of @xmath300 is @xmath301 .
but since @xmath302 and @xmath303 is not a norm of @xmath304 , @xmath303 is not in the square class of @xmath300 .
therefore , by lemma [ ker ] @xmath305 and the integral over @xmath306 in equation will be zero for any @xmath307 and @xmath14 .
using lemma [ ker ] we can see that the inner integral in is nonzero exactly when @xmath283 .
if we pick @xmath308 such that @xmath309 then elements with norm @xmath310 are represented by @xmath311 , therefore @xmath312 where we normalized the volume such that @xmath313 when we replace @xmath307 in equation by @xmath314 then the @xmath124 term in the sum is multiplied by @xmath315 .
therefore the sum in is actually the fourier expansion of @xmath7 .
we make this explicit in the following definitions : [ wvarphi ] the whittaker function corresponding to the form @xmath7 is the function @xmath316 where @xmath317 such that @xmath309 [ ed ] the whittaker model of the automorphic representation @xmath79 is the space @xmath318 we collect the result of the above discussion in the proposition [ dist ] let @xmath319 then 1 . the function @xmath7 is a cuspidal automorphic form and therefore @xmath320 2 . the function @xmath7 is distinguished in the sense of gelbart and piatetski - shapiro @xcite with fourier coefficients in the square class of @xmath321 .
the information we have so far , allows us to determine some information about the space @xmath5 .
[ setm ] in order that @xmath322 it is necessary that the quadratic form on @xmath5 is defined by @xmath108 . in this section
we compute the fourier expansion of the function @xmath323 .
first , we write @xmath324 the group @xmath84 is compact , so the representation on @xmath64 is a direct sum of isotypic components and it is not hard to see that the inner integral is the volume of @xmath84 times the projection from @xmath97 onto the @xmath109 isotypic subspace . recall that at places @xmath298 the factor @xmath87 is the sign representation and at places @xmath325 the factor @xmath87 is the trivial character .
therefore the @xmath109 isotypic subspace is the subspace @xmath326 where @xmath327 are the subspaces of odd and even functions respectively .
this means that we can replace @xmath13 by its projection to @xmath328 which we also denote by @xmath13 and then @xmath329 the space @xmath128 is isomorphic to @xmath328 . to construct the whittaker model for @xmath128
we consider the fourier expansion of the forms @xmath10 .
let @xmath330 be any element then for any @xmath331 @xmath332 the term @xmath333 can be removed from the sum since @xmath334 and @xmath335 , so @xmath336 . by equation @xmath337
so we can rewrite the fourier expansion as @xmath338 thus @xmath10 is cuspidal and the fourier expansion is completely determined by the function @xmath339 .
notice that the character @xmath340 appears twice in the sum , at @xmath341 and that @xmath342 is even so the coefficient of the character is actually @xmath343 .
the above calculation justifies the definitions : [ wvarphi0 ] the whittaker function corresponding to the automorphic form @xmath10 is @xmath344 [ whitt ] the global whittaker model of @xmath128 is the space @xmath345 recall that @xmath346 defined the quadratic form on @xmath347 .
we note a result which was evident in the fourier expansion of @xmath10 .
[ dist1 ] the automorphic forms @xmath10 are cuspidal and distinguished in the sense of gelbart and piatetski - shapiro @xcite .
the characters in the fourier expansion of @xmath10 are all in the square class of @xmath346 .
if @xmath8 is factorizable , we write @xmath348 and then @xmath349 where @xmath350 and the @xmath351-whittaker space of all such functions is @xmath352 returning to the case of one dimensional schwartz functions @xmath13 we have a similar factorization where we write an arbitarary function as a sum of factorizable functions .
if @xmath11 is factorizable we write @xmath353 and then @xmath354 where @xmath355 and the @xmath351-whittaker space ( by lemma [ setm ] ) of all such functions is @xmath356 from here on , our calculation will consist of local matching . therefore we will drop the @xmath24 subscript from most notation .
unless stated otherwise , all formulas should be understood to be over a completion @xmath25 .
we will also write @xmath357 instead of @xmath358 .
the various lemmas in this section will be used in the later sections . by lemma [ ramtriv ]
we can assume that , whenever @xmath110 , then @xmath359 .
[ conji ] with notation as above , for every place @xmath24 there is an element @xmath360 such that @xmath361 and @xmath362 . by equation ( 1 ) on page 1 and corollary 2.2 on page 6 .
in @xcite we can realize the quaternion algebra as @xmath363 with @xmath364 such that if @xmath365 then @xmath366 . where @xmath367 is the conjugate of @xmath346 in @xmath368 .
it follows that @xmath369 .
also , by corollary 2.4 on page 6 in @xcite we have @xmath370 if and only if @xmath371 .
this means that @xmath372 .
[ halpha ] for @xmath373 , let @xmath374 be any element such that @xmath375 with @xmath376 .
[ cosets ] let @xmath319 , then the coset space @xmath377 is equal to the disjoint union of right cosets @xmath378 over the set of @xmath379 such that @xmath380 if @xmath381 and @xmath382 are such that for some @xmath124 , @xmath383 and @xmath384 with @xmath203 and @xmath385 perpendicular to @xmath308 then @xmath386 and there exists an element of the orthogonal group of the orthogonal complement to the span of @xmath308 taking @xmath203 to @xmath385 .
this means that @xmath387 so the decomposition into right cosets indexed by @xmath124 is proved . to verify
the second part of the claim realize @xmath27 as @xmath363 as in the proof of [ conji ] to carry out the calculation we use the orthogonal basis @xmath388 for @xmath3 and write an arbitrary element @xmath203 in terms of this basis : @xmath389 . in terms of those coordinates ,
if @xmath203 is in the orbit of @xmath308 then @xmath391 this means that @xmath392 and then @xmath393 . now , since @xmath394 we are done .
the second lemma describes the values of @xmath395 : [ chihalpha ] if @xmath396 is defined as in [ halpha ] then @xmath397 we extend the basis defined in the proof of lemma [ cosets ] for @xmath3 to all of @xmath27 and write @xmath398 then by multiplying out the conjugation action of @xmath396 : and collecting the coefficients of @xmath308 we get @xmath399 from that we get that @xmath400 and since @xmath401 we get @xmath402 we can now derive an explicit formula for the local component @xmath120 of @xmath13 and obtain some parity results from it .
recall that @xmath24 has been fixed and we usually do nt include it in the notation .
also , recall the bruhat decomposition @xmath403 [ bmatch ] in order for the functions @xmath404 and @xmath405 to agree on @xmath406 it is necessary and sufficient that @xmath407 since the quadratic form on @xmath408 is @xmath108 , ( see lemma [ setm ] ) @xmath404 and @xmath405 are in the same whittaker space .
therefore , in order for them to agree on @xmath406 , it is necessary and sufficient that they agree on right @xmath409 cosets , i.e. on the set @xmath410 .
so we compare @xmath411 with @xmath412 solving for @xmath413 gives a necessary and sufficient condition for matching .
@xmath414 the quaternion algebra with structure constants @xmath415 has a reduced norm which can be diagonalized as @xmath416 .
this means that the determinant of the bilinear form associated to the reduced norm is @xmath417 so @xmath418 in order to to define @xmath13 according to the formula in lemma [ bmatch ] we need to show that the formula can be extended to a schwartz function of @xmath25 .
we start with the definition let @xmath14 be a schwartz function in @xmath205 and @xmath9 a quadratic character of @xmath121 . define a function on @xmath121 .
@xmath419 [ ari ] if @xmath25 is non archimedean , and @xmath420 then @xmath421 defined by equation can be extended to a function @xmath162
. the next few lemmas will establish proposition [ ari ] .
the fact that [ varphi0 ] defines a schwartz function at the archimedean place will follow from an indirect argument in lemmas [ archimatch ] and [ archimatch2 ] .
[ lc ] when @xmath25 is non archimedean then @xmath422 is locally constant on @xmath121 and supported inside a compact set of @xmath22 .
since @xmath14 has compact support there will be some @xmath423 such that @xmath424 .
therefore for @xmath425 small enough , @xmath426 for any @xmath427 and so for @xmath425 small enough @xmath428 .
since @xmath14 is locally constant and has compact support there is a neighborhood of @xmath429 such that for all @xmath430 , for all @xmath203 in the support of @xmath14 we have @xmath431 .
thus , for @xmath432 and any @xmath430 , @xmath433 [ lcdiv0 ] when @xmath27 is the division algebra @xmath434 , @xmath25 is non archimedean , and @xmath9 is nontrivial , then @xmath13 vanishes on a neighborhood of @xmath435 . since in the division algebra case @xmath436 is compact , for @xmath437 small enough @xmath438 for all @xmath381 . for such @xmath437 @xmath439 since
@xmath440 is compact . and the integral over @xmath436 is equivalent to an integral over the compact group @xmath441 . [ lcm20 ]
when @xmath28 , @xmath25 is non archimedean , and @xmath9 is non trivial , then @xmath13 is constant on a neighborhood of @xmath435 .
since @xmath9 is nontrivial , by lemma [ compactness ] the set @xmath440 is compact and we can define the function @xmath442 it is not hard to see that @xmath443 depends only on the norm of @xmath203 and on @xmath124 as defined in [ halpha ] .
next , by lemma [ cosets ] we can see that the measure @xmath444 on @xmath445 is a positive scalar multiple of the product measure @xmath446 with @xmath447 and @xmath124 ranging over the set of @xmath124 such that @xmath448 . also , by lemma [ chihalpha ] @xmath449 .
this means that for the purpose of showing that @xmath13 is constant near @xmath435 we can replace the integral over @xmath445 by the integral @xmath450 for some constant @xmath451 . with an abuse of notation we can write @xmath443 as a function of two variables
@xmath452 where @xmath453 and @xmath454 . therefore we can write @xmath455 .
furthermore , the characteristic function of the set @xmath456 is the function @xmath457 $ ] .
when we multiply by the characteristic function we can extend the integral to an integral over all of @xmath25 .
@xmath458 \chi(2(\alpha + 1))d\alpha \\ & = \frac{\chi(2)c}{2}\chi(r)|r| \int_{{\ensuremath{\textup{f}}}_{v}}\bar{\varphi}(\alpha r , r^{2}\kappa ) [ \chi(\alpha+1)+\chi(\alpha-1)]d\alpha\end{aligned}\ ] ] we change variables to @xmath459 and we get : @xmath460d\alpha'\ ] ] to continue , we will need to extend @xmath308 to a basis for @xmath3 with @xmath461 such that @xmath462 and @xmath463 and we also need the following lemma . [ eps ] for some @xmath464 , if @xmath465 and @xmath466 then @xmath467 the condition that @xmath466 means that @xmath468 so the square class of @xmath469 can be either @xmath300 or @xmath30 .
if @xmath470 i.e. the square class of @xmath469 is @xmath30 , let @xmath471 , it is not hard to see that @xmath472 regardless of @xmath473 .
let @xmath474 , then @xmath475 and the @xmath308 coordinate of @xmath203 is @xmath473 so @xmath476 since @xmath14 is locally constant and @xmath477 is compact , for @xmath437 small enough @xmath478 similarly when @xmath479 since @xmath9 is not the trivial character , @xmath480 so it is not possible that @xmath481 where @xmath307 is the index of the coset of squares in @xmath482 , also since in this case @xmath483 , it is not possible that @xmath484 .
let @xmath485 and @xmath486 .
as above , @xmath487 and if we let @xmath488 we get that for small enough @xmath437 @xmath489 to continue with the proof of lemma [ lcm20 ] notice that for @xmath473 in some neighborhood of @xmath435 we have @xmath490 since we normalized the volume of @xmath477 to be @xmath30 . to show that for @xmath437 small enough @xmath491 is independent of @xmath437 let @xmath437 be small enough that @xmath492 for any @xmath473 such that @xmath466 .
this means that @xmath493d\alpha'.\ ] ] if @xmath494 then @xmath495 so we can split the integral into two parts @xmath496d\alpha'\end{aligned}\ ] ] for @xmath437 small enough the integral over @xmath497 becomes @xmath498d\alpha'\ ] ] which splits into the sum of the integrals @xmath499 and when we change variables in the two integrals we get @xmath500 so the sum of the two integrals is @xmath501 we have now established proposition [ ari ] .
the following proposition is needed in order to define the global representation .
[ fundi ] if @xmath25 is nonarchimedean with @xmath502 odd , @xmath9 is unramified , @xmath160 , @xmath503 and @xmath504 then @xmath505 . from lemma [ bmatch ] we have
that when @xmath506 must be @xmath507 if we normalize the measure on @xmath508 such that @xmath509 then the volume of the set @xmath510 is one .
notice that for @xmath511 the set @xmath512 is the same as the set @xmath513 and so by lemma [ volomegal ] its volume is @xmath514 .
this means that for @xmath515 we have @xmath516 . also for @xmath517
so @xmath518 and @xmath519 . when @xmath9 is unramified non trivial we normalize the measure such that @xmath520 then by lemma [ unramfundi ] @xmath505 .
[ volomegal ] when @xmath297 ( and hence @xmath521 ) , @xmath25 is non archimedean and @xmath522 is any lattice then if @xmath523 then @xmath524 .
a few things are clear : it is enough to show the result in the case of @xmath525 , and that the volume is independent of the choice of @xmath308 of norm @xmath300 .
it is also not hard to see that we can assume that @xmath526 . to carry out the calculation , pick an isomorphism of @xmath23 with @xmath527 such that @xmath528 is the trace zero matrices with integer coefficients and @xmath529 let @xmath530 .
it turns out that the double coset @xmath531 is represented by the set @xmath532 with @xmath533 this means that the coset space @xmath534 is the disjoint union of the cosets @xmath535 .
it is clear that @xmath536 and @xmath537 so @xmath538 if and only if @xmath539 and also @xmath540 if and only if @xmath541 .
the only @xmath542 such that @xmath543 is @xmath544 , so that @xmath545 is equal to the volume of the set @xmath546 . similarly @xmath547 and @xmath544 are the only @xmath542 such that @xmath548 and so @xmath545 is the sum of the volumes of @xmath549 and @xmath550 .
comparison of the volumes of the two lattices now follows from a simple group calculation .
[ unramfundi ] if @xmath502 is odd , @xmath9 is unramified non trivial , @xmath551 and @xmath552 then @xmath13 as defined by equation [ varphi0 ] is equal to @xmath553 .
pick @xmath554 with @xmath555 . by lemma [ ext ]
we can see that @xmath556 notice that the function @xmath14 is invariant by the action of @xmath557 so to calculate @xmath558 the integral can be replaced by a sum over coset representative of @xmath559 to actually calculate the integral we need to calculate the volume of the cosets @xmath560 .
since we set @xmath561 we obtain the volume of the other cosets by comparing them to the volume of @xmath549 . and
in fact , for @xmath562 , @xmath563 .
we can now calculate @xmath13 .
first notice that if @xmath564 then @xmath565 and @xmath519 .
when @xmath566 then @xmath567 now , @xmath568 so for @xmath569 is not in the support of @xmath14 so @xmath570 we can now obtain some parity properties for @xmath14 and @xmath13 . [ par ]
the parity of @xmath13 is even when @xmath160 and odd if @xmath163 .
be lemma [ conji ] there is an element @xmath571 such that @xmath361 and @xmath362 . with
that we can calculate @xmath572 with @xmath573 @xmath574 [ par2 ] let @xmath575 with @xmath576 an even function and @xmath577 an odd function . if @xmath578 then @xmath579 and if @xmath580 then @xmath581 .
consider @xmath576 . by corollary [ par ] @xmath582 so @xmath583 and we see that if @xmath584 then @xmath585 . the case of @xmath586 is similar
. recall that the choice of parity at places of @xmath22 determines the character @xmath12
. we can restate corollary [ par ] in terms of @xmath12 and obtain the remaining part of the local data defining @xmath587 . in order that @xmath588 it is necessary that the character @xmath589 is defined by @xmath590
for the following , we will need to recall the bruhat decomposition . from lemma [ bmatch ] we know that if @xmath13 is defined according to equation then @xmath404 and @xmath405 agree on @xmath591 . it remains to show that with the same @xmath13 , the whittaker functions @xmath404 and @xmath405 will also agree on @xmath592 .
the verification of this result is simplified by the following lemma .
[ coin ] let the map @xmath593 be defined by @xmath594 where @xmath13 is given by equation , then for any @xmath595 @xmath596 since @xmath597 , it will suffice to show that for any @xmath598 , @xmath599 and for any @xmath600 @xmath601 to establish equation @xmath602 we compute @xmath603 @xmath604 on the other hand @xmath605 @xmath606 comparing the two sides of equation , we see that they are equal if @xmath607 recall from proof of lemma [ bmatch ] that @xmath608 is in the square class of @xmath61 for both division and matrix algebra so it is true that @xmath609 to establish equation we calculate @xmath610 @xmath611 [ coinvariant ] if for all @xmath612 for some constant @xmath451 , then for all @xmath420 and for all @xmath613 , @xmath614 . the subgroup @xmath615 can be factored as @xmath616 . since @xmath617 the cell @xmath592 can be written as @xmath618 .
consider any @xmath619 , by assumption , for any @xmath620 we have @xmath621 so , using lemma [ coin ] @xmath622 so @xmath404 and @xmath623 coincide on all of @xmath624 . finally ,
since both @xmath625 and @xmath626 are in the same whittaker space by construction , they coincide on all of @xmath627 we will use two different arguments to show that @xmath404 matches @xmath405 . in some cases
we will match a single @xmath14 and use general information about the weil representation to conclude that in fact , all other functions @xmath14 are also matched by @xmath13 .
in other cases we will show directly that for any @xmath628 for some constant @xmath451 and then use lemma [ coinvariant ] to conclude that the @xmath629 on all of @xmath142 for some constant @xmath451 .
finally , we will show that @xmath630 and deduce the actual equality on all of @xmath631 from that relation . from equation and the definition of @xmath405 , ( definition [ wvarphi0 ] .
) we have that @xmath632 notice that we set @xmath346 in equal to @xmath300 according to lemma [ setm ] .
the measure @xmath242 is self dual with respect to the fourier transform @xmath633 . since the conductor of @xmath66 is @xmath634 it follows by a standard argument that @xmath635 where @xmath636 gives @xmath170 volume @xmath30 .
since @xmath637 and @xmath638 we can simplify the expression for @xmath639 to @xmath640 . and
substituting the explicit expression for @xmath13 in terms of @xmath14 , we have @xmath641 on the other hand , by the definition of the action of @xmath642 ( equation ) and the definition of @xmath404 we have @xmath643 as mentioned above , we need to verify the equality of and . notice that both equations involve integrals over @xmath304 and @xmath436 so verification of the equality is independent of the realization of the local algebra @xmath27 as long as the measures @xmath242 and @xmath636 are self dual with respect to the action of @xmath644 and the measure on @xmath436 is the same in both equations .
we continue with the calculation of the right hand side of .
since @xmath645 is nontrivial , and @xmath646 we know from lemma [ compactness ] that @xmath647 is compact . it is also clear that the integrand in is constant on left @xmath648-cosets so we will replace the integral over @xmath377 with an integral over @xmath649 . to do that ,
let @xmath650 and we have : @xmath651 we now change variables in the inner integral from @xmath652 to @xmath653 @xmath654 next we let @xmath655 be the characteristic function of a compact set @xmath656 and we note that for any increasing sequence of compact sets @xmath657 where in the last step we replaced the integral over @xmath304 by the integral over @xmath658 .
this is justified by the fact that the set of elements with @xmath435 norm have measure @xmath435 in @xmath3 .
now , from lemma [ ker ] we know that if @xmath659 is not in the square class of @xmath300 then there is an element @xmath660 in its stabilizer @xmath661 such that @xmath662 . in that case the inner integral is zero .
so we can assume that the support of @xmath14 is contained in the set @xmath663 we state this as a lemma for reference .
[ sup ] let @xmath664 be a nontrivial character and let @xmath665 be the restriction of @xmath14 to @xmath666 then @xmath667 .
to continue , we will parameterize @xmath666 by using the two fold covering @xmath668 the measure on @xmath669 is the measure @xmath242 which makes the fourier transform @xmath670 self dual .
this measure is invariant under the action of @xmath43 and transforms by @xmath671 under multiplication by a scalar @xmath672 . the measure @xmath673 on the cover @xmath674 is also invariant under the action of @xmath43 and also transforms by @xmath671 under multiplication by @xmath672 .
it follows that the two measures are related by a positive constant @xmath451 .
we defer the calculation of @xmath451 since we will soon change variables again and will normalize the measure that results from the two changes of variables .
the relation between the two measures allows us to write @xmath625 in terms of integrals over @xmath675 @xmath676 notice that the integrand in the integral over @xmath441 is left invariant under @xmath647 and recall that the volume of @xmath647 is @xmath677 and so the inner integral can be changed to an integral over @xmath377 .
@xmath678 comparing with equation we see that the third integral is @xmath679 so we get @xmath680 the integrand in @xmath625 depends only on cosets of the form @xmath681 so using lemmas [ cosets ] [ halpha ] we can change the integral in @xmath625 over @xmath377 to an integral over the product of the set of @xmath124 such that @xmath682 and @xmath683 .
we have now changed measures from @xmath242 to the product of the measures @xmath684 , with @xmath124 as above , @xmath685 and @xmath686 .
at this point in the calculation we note that the constant relating the two measures will depend on the type of algebra @xmath27 and on the norm of @xmath308 defining the torus @xmath648 .
we denote the constant relating the measures @xmath242 and @xmath687 by @xmath688 .
so far the formula for @xmath625 looks like : @xmath689 where @xmath124 ranges over elements such that @xmath690 . also recall that @xmath691 with @xmath692 so @xmath693 with the above observation and with lemmas [ cosets ] and [ chihalpha ] the two inner integrals become @xmath694 since we are taking the limit over all increasing sequences of lattices @xmath528 the limit would not change if we considered only lattices which are invariant under the action of the compact group @xmath647 . in that case , the integrand does not depend on @xmath695 .
let @xmath696 and the inner integrals become @xmath697 the characteristic function of the set of @xmath698 such that @xmath690 is the function @xmath699 so we can extend the integral over @xmath124 to an integral over @xmath700 by multiplying the integrand by the characteristic function .
the integral over @xmath25 is only defined as a limit and should be understood that when we write @xmath701 we will mean @xmath702 .
@xmath703 we split the integral into two integrals and change variables to @xmath704 and to @xmath705 respectively .
we also define @xmath706 and extend the integral to all of @xmath25 .
@xmath707 we use another change of variable @xmath708 and we get @xmath709 we give the above integral a name [ gauss ] the gauss integral is the quantity @xmath710 so far the inner integrals in equation [ lookslike ] are @xmath711 putting equation [ sofar1 ] together with equation we get @xmath712 @xmath713 now by lemma [ par ] @xmath714 and so @xmath715 what we showed so far is that on the big cell , the two whittaker functions @xmath404 and @xmath405 are related by a constant which depends on @xmath716 but does not depend on the argument of the function .
i.e. @xmath717 with @xmath718 if we showed that @xmath719 we would be done .
[ nantm ] when @xmath25 is non archimedean and @xmath319 then for any @xmath720 @xmath721 furthermore , @xmath722 and for any @xmath420 we have @xmath723 .
the discussion above shows that for any @xmath14 , and for @xmath13 defined by equation @xmath724 . by lemma [ coinvariant ]
we have that @xmath725 for all @xmath726 .
since the weil representation of @xmath631 is smooth , the stabilizers of @xmath14 and @xmath13 are both open subgroups and so the intersection of the stabilizers @xmath727 is an open neighborhood of @xmath728 .
it is not hard to see that @xmath729 and @xmath730 we can find @xmath731 such that @xmath732 . if we pick @xmath14 such that @xmath733 then @xmath734 and @xmath735 since @xmath736 .
therefore @xmath719 and we are done . the argument for matching in this case is different from the argument in the case of a nontrivial character . in this case , the equality of the spaces @xmath737 and @xmath738 follows from results of waldspurger and rallis .
the explicit matching of the functions @xmath404 and @xmath405 will then follow from lemma [ bmatch ] .
[ natm ] for @xmath110 and @xmath25 non archimedean @xmath739 furthermore , for every @xmath740 and @xmath13 defined by equation we have @xmath741 we define the map @xmath742 by @xmath743 . the map @xmath744 is @xmath143 equivariant and it is easy to see that it factors through the space @xmath745 of @xmath43 coinvariant . since we know from lemma [ par2 ] that @xmath746 , the map @xmath744 actually factors through the @xmath747 coinvariants @xmath748
. when @xmath110 the formula for @xmath404 is @xmath749 if we let @xmath14 be the characteristic function of some neighborhood containing elements of norm @xmath300 then it is easy to see that @xmath750
. therefore @xmath751 .
on the other hand , let @xmath752 where @xmath149 is the character @xmath753 there is an @xmath143 equivariant map @xmath754 given by @xmath755 which also factors through @xmath748 .
let @xmath756 be the image of @xmath757 .
a result of rallis ( @xcite thm ii.1.1 ) says that the map @xmath757 induces an isomorphism @xmath758 .
it is also known ( e.g. gelbart @xcite ) that since @xmath3 is a split space , then in fact @xmath759 .
furthermore , the composition series of @xmath760 has length two , it has a unique irreducible submodule and a unique irreducible quotient . in @xcite , waldspurger analyzed the two factors in the composition series .
his result is that the unique quotient is @xmath761 : the local theta correspondence of the trivial character from @xmath5 ( which waldspurger denotes by @xmath762 . )
this unique quotient has a @xmath351 whittaker model whereas the unique irreducible submodule of @xmath760 ( which waldspurger denotes by @xmath763 has whittaker models for all square classes of characters @xmath764 except for @xmath351 ( lemma 3 on p 227 in @xcite ) . since @xmath744 factors through @xmath765
there is a map @xmath766 and by considering the whittaker models we see that @xmath767 and so @xmath768 .
i.e. @xmath769 is the unique irreducible quotient of @xmath760 which is isomorphic to @xmath761 , that is @xmath770 .
the argument splits into three cases handled separately in the next subsections .
[ [ chicircnu - is - trivial - on - ensuremathrm - oensuremathtextup - vensuremathmathbbr . ] ] @xmath57 is trivial on @xmath771 . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ throughout this subsection , @xmath25 will be isomorphic to @xmath772 and we will pick a particular realizations of the matrix algebra and division algebra such that @xmath773 when @xmath9 is nontrivial and @xmath774 when @xmath9 is trivial .
we will not mention these conditions in the statements of the lemmas , but they should be understood to be stated with respect to these choices of @xmath300 . since the equality we will establish is defined independently of a particular realization
there is no loss of generality .
the cases included in this subsection are the case of @xmath775 ( which by lemma [ ramtriv ] , can only exist when @xmath23 is the matrix algebra ) and the case @xmath19 and @xmath23 is the division algebra ( in which case @xmath776 so @xmath777 ) . note that in both cases , once we know that we can match all the forms coming from the representation of @xmath778 , by lemma [ par2 ] , the parity of all @xmath779 can be assumed to be even .
hence only the trivial extension of @xmath267 to @xmath771 will have a nonzero lift .
this is the reason for describing both cases in this subsection as the case of @xmath57 trivial on @xmath771 .
the matching in this case will be established by the matching of the whittaker functions corresponding to the gaussians and using the fact that the the space @xmath780 is generated by the image of the gaussians .
the matching of the gaussians will depend on calculating their @xmath549-types and using the iwasawa decomposition to conclude that the two whittaker functions agree on all of @xmath631 . the main argument is contained in the following theorem .
[ archimatch ] let @xmath9 be a the trivial character when @xmath23 is a matrix algebra or the nontrivial character when @xmath23 is the division algebra .
then @xmath781 and for any @xmath782 we have and @xmath783 .
as explained above , in this case the characters extend to the trivial character on @xmath747 .
therefore , as mentioned in @xcite , it follows from a result of howe @xcite that @xmath737 is generated by the image of the gaussian : @xmath404 .
the calculation that we carry out in this proof will show that if @xmath14 is the gaussian then @xmath783 and therefore @xmath784 . therefore , for any @xmath14 , there is some @xmath785 such that @xmath786 and by lemma [ bmatch ] we know that @xmath787 must be @xmath13 .
this would also show that the formula for @xmath13 defines a schwartz function in the restricted schwartz space .
the second result is a consequence of lemma [ par2 ] , which implies that the space @xmath738 is the image of the irreducible representation on even functions @xmath788 so it is also irreducible and @xmath781 . by the above argument
, the calculation that remains is to show that when @xmath14 is the gaussian we have @xmath783 . by lemma [ weights ] we know that for all elements of @xmath789 we have @xmath790 and @xmath791 for some @xmath346 .
as we mentioned above we picked a realization of the algebras with @xmath796 when @xmath9 is nontrivial or trivial , respectively . by lemma [ setm ] , this means that we must choose the quadratic form on @xmath5 to be @xmath797 and @xmath798 respectively .
the lie group @xmath549 is connected so its representation is determined by the action of the lie algebra .
this lie algebra is generated by the element @xmath801 so it will be enough to calculate the actions of @xmath802 to show that the actions of the whole algebra are the same .
we denote the lie algebra action of @xmath803 and @xmath246 by @xmath804 and @xmath805 , respectively . instead of calculating the action of @xmath802 directly , we write @xmath806 with @xmath807 and calculate the following actions @xmath808 and @xmath809 on @xmath13 and @xmath14 respectively .
it is not hard to see that @xmath810 and @xmath811 the rest of the proof is a lengthy but standard calculus computation .
just as an illustration of the type of calculation involved we calculate the action on @xmath14 when @xmath9 is trivial . in this case
the gaussian is the function @xmath812 and we calculate the first term in the action of @xmath802 . @xmath813 we now calculate the second term in the action of @xmath802 .
@xmath814 the last step uses a lemma ( which is also omitted ) which says that the gaussian is an eigenfunction of @xmath642 with eigenvalue @xmath815 .
@xmath816 to continue the calculation we let @xmath817 and then @xmath818 using standard formulas from calculus and integrating with respect to the self dual measure @xmath819 so that @xmath820 adding up equation and we get the action of @xmath802 on the gaussian @xmath821 [ [ chicirc - nu - is - nontrivial - on - ensuremathtextup - hensuremathmathbbr . ] ] @xmath267 is nontrivial on @xmath822 . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
the remaining archimedean case is the case of a nontrivial @xmath9 and @xmath23 is a matrix algebra . the argument that will show matching is similar to the one for the other real cases .
we will use the fact that the representation is known to be irreducible and the matching of one function to conclude that the whittaker spaces of @xmath737 and @xmath738 coincide . since in this case the quadratic space @xmath3 is split with signature ( 2,1 ) we can use a theorem of zhu , ( 1.2 in @xcite ) which says that in this case , when @xmath826 , @xmath827 . by lemma [ archimatch ]
we know that @xmath828 is irreducible and therefore so is @xmath737 .
we will show that when @xmath14 is the gaussian then @xmath723 , this would mean that the whittaker spaces are equal and by lemma [ bmatch ] it follows that for any @xmath420 we have , @xmath723 .
it will also follow that @xmath13 is actually a function in the restricted schwartz space of @xmath829 .
as in the proof of theorem [ archimatch ] it will be enough to show that when @xmath14 is the gaussian then @xmath14 and @xmath13 are of the same k - type .
this is done in lemma [ weights2 ] and completes the argument .
the local data in the complex case is as simple as possible .
since every complex number is a square , the quadratic character @xmath831 is trivial .
for the same reason , the quaternion algebra is ramified i.e. @xmath160 .
the correspondence in this case can be deduced from known facts without any additional calculation . in @xcite ( proposition 2.1 ) adams and barbasch show that the theta lift of the trivial character from @xmath830 to @xmath832 is isomorphic to the irreducible representation @xmath833 which is the weil representation of @xmath832 on @xmath834- the space of even schwartz functions . if @xmath835 is the automorphic form in the theta lift of the character @xmath57 extended trivially from @xmath836 to @xmath830 defined similarly to @xmath7 then @xmath837 .
therefore , @xmath838 and to conclude the matching of functions we need to show that the spaces of functions @xmath79 and @xmath128 actually coincide .
this follows from the multiplicity one of spherical representations of @xmath832 .
the fact that the representations are spherical is verified by the existence of the gaussian functions in both representations .
therefore , the matching for the actual function is given by the formula for @xmath13 . | this paper describes the lifting of automorphic characters of @xmath0 to @xmath1 .
it does so by matching the image of this lift with the lift of automorphic characters from @xmath2 to @xmath1 .
our matching actually gives a matching of individual automorphic forms , and not just of representation spaces .
let @xmath3 be a @xmath4 dimensional quadratic vector space and @xmath5 a certain @xmath6 dimensional quadratic space . to an automorphic form @xmath7 determined by the schwartz function @xmath8 in the lift of the character @xmath9
we match an automorphic form @xmath10 determined by the schwartz function @xmath11 in the lift of the character @xmath12 .
our work shows that , the space @xmath5 is explicitly determined by the character @xmath9 . the character @xmath12 is explicitly determined by the space @xmath3 and the function @xmath13 is given by an orbital integral involving @xmath14 .
theta lift , automorohic characters |
electrons traveling in vacuum to a measurement apparatus will repel each other because of their mutual coulomb interaction .
if a cloud of electrons is sufficiently dense , this vacuum space charge effect can alter the time , energy , and spatial spread of the cloud , limiting experimental resolution and resulting in systematic measurement errors.@xcite the space charge effect has been addressed mostly for the design of electron guns for time - resolved electron diffraction experiments @xcite and has been discussed in terms of @xmath0-body numerical simulations , mean - field models , and fluid dynamics .
recently the space charge effect has also been discussed in angle - resolved photoemission spectroscopy ( arpes ) experiments @xcite , where it has been shown to depend primarily on four parameters : photon energy , pulse duration , incident beam spot size , and the number of electrons photo - emitted per pulse .
these previous studies explored two different regimes : i ) the relatively high photon energy ( @xmath1 ev ) and long pulse duration ( @xmath2 ps)@xcite regime , and ii ) the very low photon energy ( two - photon photoemission with 3 ev photons ) and short pulse duration ( @xmath3 fs ) regime.@xcite the shape of the electron cloud is very different in these two regimes . in the first case ,
the wide distribution of electron kinetic energy and long pulse duration produce a cloud extended mainly along the direction normal to the sample surface , while in the second case the narrow electron kinetic energy distribution and short pulse duration produce a disk - shaped electron cloud parallel to the sample surface .
the former case of a more three - dimensional cloud requires numerical simulations while a disk - shaped cloud can be characterized analytically using a simple power law.@xcite with the recent advent of laser - based photoemission experiments , the need to extend these studies to an intermediate regime becomes imperative .
a recent study of the space charge effect using laser based arpes with 7 ev photon energy and 10 ps long pulse suggests that the space charge effect plays a minor role in this regime thanks to the mirror charge effect that nearly cancels the space charge effect in the 1 to 10 ps pulse duration.@xcite here we expand these results by reporting a systematic study of the space charge effect on polycrystalline gold in laser - based photoemission spectroscopy in the intermediate regime of moderate pulse lengths ( @xmath4 6 ps ) and low photon energy ( 6 ev ) .
we find in the limit of small spot size ( @xmath4 20 @xmath5 m diameter ) a shift of the fermi edge on the order of 2 mev and a broadening larger than 10 mev starting already at 100 electrons per pulse .
more generally , we find that a single power law can not describe the broadening or the fermi edge shift as a function of electrons per pulse and spot size in the various arpes regimes . to explore the non - trivial relation between the various experimental parameters , we modeled our data with a molecular - dynamics model which employs full @xmath0-body numerical simulations of the electron propagation from the surface to the detector.@xcite this model is able to reproduce quantitatively the space charge effect seen in the long and short pulse regimes as well as in the intermediate regime explored here .
the large range of applicability of this model allows one to identify the functional dependence of the spectral shift and broadening on the number of electrons per pulse and spot size .
we found that one of the key parameters affecting the different power laws observed in different arpes regimes is the maximum kinetic energy of the initial photo - electron distribution .
we believe that this study will be crucial for guiding next generation high resolution setups .
experiments were performed on a polycrystalline gold sample scratched in air and inserted through a load lock into a uhv analysis chamber at low temperature ( 15 - 20 k ) .
pressure remained below @xmath6 torr throughout the experiment . in fig . 1 we show a diagram of the experimental setup .
the sample geometry is detailed in the inset .
the spot size was characterized by using a wire mesh mounted on the same sample holder as the gold sample .
( color online ) laser - arpes experimental setup .
the focal lengths are : l1=10 cm , l2=20 cm and l3=30cm.,width=302 ] the laser system is a cavity - dumped , mode - locked ti : sapphire oscillator ( coherent mira ) , pumped with a 6 w frequency - doubled nd : yvo@xmath7 laser .
the oscillator generates @xmath4 150 fs pulses at 840 nm at a repetition rate of 54.3/@xmath8 mhz ( @xmath9 ) with pulse energies as high as 100 nj .
the photon energy of the laser beam was frequency - quadrupled via cascaded , type - i phase - matched second harmonic generation in two beta barium borate ( bbo ) crystals of 2 mm and 5 mm thickness .
the resulting 210 nm ( 5.9 ev ) pulse was estimated to be @xmath4 6 ps long due mainly to group velocity mismatch .
the uhv chamber window is made of uv- grade fused silica .
we adjusted the pulse energy by tuning the power applied to the cavity dumper bragg cell .
the focusing lens ( l3 in fig
. 1 ) has a 300 mm focal length .
we varied the spot size by defocussing l3 . assuming a work function of about 4.4 ev , electrons at the fermi energy exit the sample with 1.5 ev kinetic energy , corresponding to a velocity of 0.7 @xmath5m / ps . with a pulse lasting 6 ps ,
the initial electron spread normal to the surface is thus about 4 @xmath5 m . with a spot size diameter of about 20 - 100 @xmath5 m
, the initial electron cloud is a flat cylinder .
this is an intermediate case between the fs laser - based data and the synchrotron - based data.@xcite the integrated photoemission signal was measured using a phoibos 150 mm hemispherical analyzer from specs with a 0.2 mm entrance slit and 4 ev pass energy , yielding an energy resolution of 4 mev .
the sample was electrically grounded through a picoammeter , which allowed the measurement of the total photo - electron yield .
all the spectra were averaged over @xmath10@xmath11 along the analyzer slit except where explicitly stated otherwise . the angle between the analyzer and the laser beam
is fixed at 55@xmath11 and the light incident on the sample is @xmath12-polarized .
( color online ) spot size measurements .
( a ) photo - electron yield as a function of laser power .
( b ) typical * x * and * y * scan of the integrated photoemission intensity from the w mesh .
the * y * scan is vertically shifted , width=302 ] fig .
2 shows the characterization of the pulse energy and spot size .
the total photo - electron yield as a function of the laser power is shown in fig .
the repetition rate was set to 5.4 mhz .
the solid line is a linear fit giving a yield of @xmath13 e@xmath14/photon . for comparison ,
the typical yield for a photon energy of 35 ev measured at bl 10.0.1 at the advanced light source ( als ) is about 0.2 e@xmath14/photon ( @xmath15 photons / s with @xmath16 na yield).@xcite the difference can be mainly attributed to the range of electronic states probed : 6 ev photons only excite states in the au 5d-6s valence band with binding energy less than about 1.5 ev , whereas 35 ev photons can excite the entire valence band.@xcite additionally , in the latter case secondary electrons amplify the photo - electron yield noticeably.@xcite the spot size was measured by scanning a 100@xmath17100 lines / inch tungsten mesh at the sample position while measuring the integrated photoemission intensity .
the mesh wire diameter was 25 @xmath5 m .
the @xmath18 angle ( see fig .
1 ) was set to 55@xmath11 during the mesh scan so that the * x * and * y * scanning directions were perpendicular to the laser beam . fig .
2(b ) shows a typical scan along the * x * and * y * direction obtained from this procedure .
we then estimated the spot size by deconvolving the peak s full width at half maximum ( fwhm ) using the known mesh wire diameter .
the reported diameter is the square root of the * x * and * y * fwhm product , corresponding to the diameter of a disc of equal area .
the ratio between the fwhm along * x * and * y * is not constant but was smaller than 3 .
we note that with an incident gaussian beam of 2 mm diameter , the diffraction limited spot size with a 300 mm focal length lens is about 15 @xmath5 m . for the fwhm along the * x * direction , we used a corrected wire diameter to account for the fact that only the part of the wire facing the analyzer could contribute to the measured signal .
it can be shown that the effective wire diameter in this case is @xmath19 @xmath5 m . the small difference in peak separation between the * x * and * y * scan is due to mesh spacing inhomogeneity .
( color online ) fermi edge of the au sample as a function of the number of photo - electrons per pulse ( @xmath0 ) .
the spectra are normalized and shifted vertically .
the solid lines are fits.,width=302 ] fig .
3 shows typical integrated photoemission spectra from the gold sample at various laser powers showing the fermi level broadening and shift .
the spot size diameter was 18 @xmath5 m .
the edges of these spectra have been fit to a function of the form : @xmath20 the parameters @xmath21 and @xmath22 are fitting parameters and @xmath23 is the electron kinetic energy .
the first two terms account for the background , which includes an offset and a linear contribution for @xmath24 .
the third term is the complementary error function @xmath25 , @xmath26 which is commonly used to fit fermi - dirac distributions convolved with experimental resolution .
the characteristic fwhm is defined to be the full - width - half - maximum value of the associated normalized gaussian .
the top of the panel shows a sharp fermi edge at low power with a fwhm of 7.3 mev , consistent with the experimental energy resolution ( 4 mev from the analyzer and 4 - 5 mev from the laser bandwidth ) and the sample temperature ( about 15 k ) .
the total photo - electron yield measurement indicated an average of 1 electrons per pulse .
as the laser power is increased , a greater number of photo - electrons ( @xmath0 ) are emitted per pulse .
this results in a stronger space charge effect , and the edge becomes broader and shifts to higher kinetic energy .
the shift to higher kinetic energy is due to the fact that the electrons at the fermi energy have the highest initial kinetic energy , and therefore the coulomb repulsion of slower photo - electrons can only push them towards higher kinetic energy ( neglecting any mirror charge effects ) .
we note that within the laser power range used here , the @xmath25 fit function captures the experimental line shape very well .
this was not the case at higher laser power ( not shown ) .
( color online ) space charge effect as a function of the number of electrons per pulse ( @xmath0 ) for various spot diameters ( @xmath27 ) .
( a ) broadening ( @xmath28 ) of the fermi edge as a function of @xmath27 and @xmath0 .
the dashed line shows a linear fit and the dotted line a @xmath29 fit to the triangular markers .
( b ) shift of the fermi edge ( @xmath30 ) as a function of @xmath0 for various values of @xmath27 .
the dashed lines show a linear fit of the triangular markers .
the solid lines are fits to a power law .
the data set with diamond shaped markers was obtained at the als with 35ev photon energy and was adapted from ref . .
the y - axis error bars include the errors on the fit and the instrumental resolutions .
the x - axis error bar is estimated from the sample current measurement resolution .
the spectra are averaged over @xmath418@xmath11 along the analyzer slit with @xmath31@xmath11.,width=283 ] in fig .
4 we report the shift of the fermi edge @xmath30 , and its broadening @xmath28 , as a function of number of electrons per pulse ( @xmath0 ) for several values of the spot size diameter ( @xmath27 ) . here
@xmath30 is defined by the difference between the measured edge position ( @xmath32 ) and the fermi energy ( @xmath33 ) while @xmath28 is defined as the edge fwhm deconvolved from the sample temperature and experimental resolution .
the @xmath28 error bars are mainly estimated from fit standard deviation and temperature measurement uncertainty ( estimated to be about 3 k ) .
the @xmath30 error bars are the standard deviations plus 0.5 mev .
the error on @xmath0 was estimated to be @xmath34 .
we observe that @xmath28 and @xmath30 increase with increasing @xmath0 and that , as expected , this effect becomes more severe as the spot size decreases .
the data from ref . taken in a different regime ( @xmath35 ev and pulse length @xmath36 ps ) , corresponding to the largest @xmath27 value shown in the figure
( see diamond markers ) , fall in line with this trend . to determine the relation between @xmath28 , @xmath30 , and @xmath0 we fit the data with the smallest @xmath27 value using the same power law as the one reported in previous studies.@xcite it is clear that neither the linear ( see dashed line ) nor the @xmath29 dependence ( dotted line ) can describe effectively the data here reported in the ps regime . to model the transition from the linear to the @xmath29 regime
we fit the data of fig . 4 with generalized exponents : @xmath37 @xmath38 ( color online ) plot of the the fitting coefficients @xmath39 and @xmath40 as a function of @xmath27 .
the solid lines in panel ( a ) show the average value of @xmath41 and @xmath42 .
the solid and dashed lines in panel ( b ) are fits to a power law.,width=283 ] in figs .
5(a ) and ( b ) , we report the dependence of the fit coefficients @xmath39 and @xmath40 on @xmath27 , respectively .
remarkably , @xmath40 is essentially independent of @xmath27 for both @xmath28 and @xmath30 .
more specifically , for the broadening we found @xmath43 , which is an intermediate value between the value ( 1 ) reported for synchrotron pulses @xcite and the value ( 0.5 ) reported for fs pulses.@xcite for the fermi edge shift we found @xmath44 . in both cases the prefactor @xmath39 increases with decreasing @xmath27 . to characterize the space charge effect dependence on @xmath27 more quantitatively , we fit @xmath39 with a function of the form @xmath45 where @xmath46 and @xmath47 are fitting parameters . in a log - log plot
, @xmath48 is a straight line with slope @xmath47 . by substituting @xmath48 into eq .
( [ eq : g1 ] ) , we find that the space charge effect dependence on the number of photo - electrons @xmath0 and spot diameter @xmath27 is of the from : @xmath49 more specifically we found for the fermi edge shift : @xmath50 and for the fermi edge broadening : @xmath51 where as in all following formulas @xmath52 and @xmath53 are in mev and @xmath27 is in @xmath5 m .
( a ) @xmath28 and @xmath30 as a function of the emission angle @xmath54 . @xmath55 mev and @xmath56 mev are estimated from the fit of the angle - integrated spectrum .
( b ) photoemission intensity as a function of @xmath54 .
the solid lines are fits with the function @xmath57.,width=283 ] to quantify the angular dependence , the angle - resolved results are shown in fig .
the emission angle ( @xmath54 ) is along the same axis as @xmath18 with @xmath58@xmath11 corresponding to normal emission .
the angular acceptance of the analyzer was @xmath4 25@xmath11 along the entrance slit .
panel ( a ) shows the strong dependence of @xmath30 and @xmath28 from @xmath54 .
we show in panel ( b ) the photoemission intensity measured at @xmath59 mev below the fermi energy . as a guide for the eye , we fit @xmath60 , @xmath61 and the photoemission intensity with empirical functions of the form @xmath62 .
an offset is added in the case of the photoemission intensity .
the strong correlation between the fermi edge broadening and the photoemission intensity suggests that the angular dependence of the space charge effect is mainly due to the photo - electron emission angular distribution .
the effect of the angular integration over the analyzer acceptance angle is different for @xmath30 and @xmath28 .
while the shift of the angle - integrated spectrum ( @xmath55 mev ) is close to the average value of @xmath60 over @xmath54 , the broadening of the angle - integrated spectrum ( @xmath56 mev ) is significantly larger than the average value of @xmath61 .
this shows that the broadening due to the angular dependence of @xmath60 can be larger than just the kinetic broadening .
this angular distribution of the space charge effect has a direct effect on the data reported here since the spectra were integrated over @xmath10@xmath11 with @xmath18=55@xmath11 .
since we reported the space charge effects as a function of @xmath0 , measurements at different emission angles will yield different values for the coefficient @xmath39 .
this should result in a different value of @xmath46 .
the exact contribution of the angle integration window and emission angle on the exponents @xmath40 and @xmath47 is more difficult to estimate .
the numerical simulations are based on a representative electron energy distribution with a sharp fermi edge at 1.56 ev .
the pulse duration was assumed to be rectangular with a duration of 6 ps .
the simulations were repeated at least 10 times for each parameters setting .
@xmath30 and @xmath28 are extracted with the same edge fit function with the exception of the parameters @xmath63 and @xmath64 being set to 0 .
the electron emission was assumed to follow a @xmath65 distribution following the experimental results of fig . 6 and only electrons with an emission angle between 46@xmath11 and 64@xmath11are integrated in order to mimic the experimental condition . the number of electrons per pulse @xmath0 varied between 200 and 4000 .
spot diameters between 18 @xmath5 m and 359 @xmath5 m were used .
( color online ) simulation of the space charge effect .
( a ) @xmath28 and ( b ) @xmath30 as a function of @xmath0 for various @xmath27 value .
the error bars result from multiple simulation runs and represent the standard deviations .
the solid lines are fits to a power law.,width=283 ] ( color online ) simulation of the space charge effect .
( a ) @xmath28 and ( b ) @xmath30 as a function of @xmath27 for various @xmath0 values .
the error bars result from multiple simulation runs and represent the standard deviations .
the solid lines are fits to a power law.,width=283 ] fig .
7 shows the simulated broadening @xmath28 and shift @xmath30 of the fermi edge as a function of the number of electrons per pulse @xmath0 for different spot diameters @xmath27 .
8 shows @xmath28 and @xmath30 as a function of the spot diameter @xmath27 for different numbers of electrons per pulse @xmath0 .
we fit the results for @xmath28 and @xmath30 with the same power law as in the experimental case and show the comparative results in table 1 .
the exponent values are the average value of all the fitting results ( not shown ) .
.summary of the measured and simulated exponent values .
[ cols=">,^,^",options="header " , ] the broadening and the shift of the fermi edge show a good overall agreement between simulation and experiment .
small systematic deviations are observed in the small effect regime . in particular for @xmath66 and/or @xmath67 @xmath5 m .
in the case of the disk - shaped photo - electron cloud produced by a fs pulse , the broadening due to the space charge effect was proposed to be of the form@xcite @xmath68 on the other hand , for the 60 ps synchrotron pulse , the broadening follows a linear relation with @xmath0 .
the @xmath69 ( @xmath70 for the simulation ) dependence we report for this intermediate regime of laser - based arpes with 6 ps pulses is between these two values .
however , these three regimes can not be solely compared on a pulse duration basis .
the fs pulse analysis is based on a surface state peak ( 86 mev fwhm ) measured at the band minimum in the brillouin zone center . in this case , the majority of the photo - electrons have a higher kinetic energy .
this will affect the space charge effect in two ways : i ) the peak position will be shifted to lower kinetic energy ; ii ) the momentum resolution will decrease resulting in a broader spectrum shifted towards higher kinetic energy . while the total shift in energy ends up being negligible , the contribution of the momentum broadening on @xmath28 is not.@xcite this makes a direct comparison difficult but further comparative analysis can be done with our model . according to the simulation results , the nonlinearity @xmath71 with @xmath72 seems to depend largely on the initial photo - electron distribution .
more specifically , the simulations show that the exponent @xmath73 decreases as the maximum electron kinetic energy decreases .
this exponent is around 1 for high kinetic energies , decreases to @xmath74 in the intermediate regime presented here with a maximum electron kinetic energy of about 1.5 ev , and decreases down to @xmath75 for the fs pulse experiment where the maximum electron kinetic energy was around 1.1 ev.@xcite we found also that the simulation shows no indications that the pulse durations affect this behavior of the exponent @xmath73 .
similarly , we find the broadening dependence on spot size to be proportional to @xmath76 ( @xmath77 for the simulation ) , compared to @xmath78 the fs pulse data , though more experimental data are required here to validate the trend .
it is interesting to note also that the ratio of the two coefficients @xmath73 and @xmath79 is not 1 as for the fs pulse case , but closer to 0.6 - 0.8 .
since the error bars on @xmath47 are relatively large , fits with @xmath80 are shown with a dashed line in fig .
5 ( b ) . from this fit
it is clear that additional data points are needed to conclude on the correct value of the ratio between @xmath47 and @xmath40 .
nevertheless , these results set already a limit in the range in which the analytical model proposed in ref .
in conclusion , we report the measurement of the space charge effect in the few - ps laser pulse regime .
we observe a broadening of the spectral feature exceeding 10 mev and an energy shift exceeding 2 mev with pulses of as few as a 100 photo - electrons for small spot sizes ( @xmath4 20 @xmath5 m diameter ) . for larger spot sizes ( @xmath4 165 @xmath5 m diameter ) similar effects
are observed only as the number of photo - electrons per pulse reaches 600 to 800 .
we found that neither the linear model that fits synchrotron data nor an @xmath29 model that fits fs pulses describe our data effectively , and we introduced a simple phenomenological model with generalized exponents that characterizes the broadening and energy shift of the spectra as a function of spot size and number of photo - electrons over the entire range probed .
the results were used as a benchmark for an @xmath0-body numerical simulation with a good overall agreement . in turn the simulation allowed us to explain the difference observed in different arpes regimes in terms of initial electron kinetic energy .
these results represent a reference for the design and interpretation of photoemission experiments with next - generation light sources , such as free - electron lasers and table - top laser sources .
the laser - arpes measurements , data analysis and the laser - arpes chamber setup were supported by the director , office of science , office of basic energy sciences , materials sciences and engineering division , of the u.s .
department of energy under contract no .
de - ac02 - 05ch11231 .
j.g . was supported in part by a muri program of the air force office of scientific research , grant no .
fa9550 - 04 - 1 - 0242 , which partially contributed to the realization of the laser - arpes setup . | we report a systematic measurement of the space charge effect observed in the few - ps laser pulse regime in laser - based solid - state photoemission spectroscopy experiments .
the broadening and the shift of a gold fermi edge as a function of spot size , laser power , and emission angle are characterized for pulse lengths of 6 ps and 6 ev photon energy .
the results are used as a benchmark for an @xmath0-body numerical simulation and are compared to different regimes used in photoemission spectroscopy .
these results provide an important reference for the design of time- and angle - resolved photoemission spectroscopy setups and next - generation light sources . |
much experimental effort has been spent to synthesize super - heavy elements by fusion mechanism of heavy systems @xcite .
it is crucial to choose the right projectile - target combinations in order to have the largest probability for forming a compound nucleus that leads to production of the desired super - heavy element . among different possibilities ,
formation of compound nucleus in collisions with deformed actinide targets with neutron rich projectile , so called hot fusion , provides a suitable choice for synthesizing of these elements @xcite .
however , in heavy systems compound nucleus formation is severely inhibited by the quasi - fission mechanism @xcite . in quasi - fission , the colliding ions attach together for a long time , but separate without going through compound nucleus formation . during the long contact times
many nucleon exchanges take place between projectile and target nuclei .
a number of models are developed for a description of the reaction mechanism in the quasi - fission process @xcite . in a recent work , the quasi - fission mechanism in @xmath0ca
+ @xmath1u and @xmath2ca + @xmath1u collisions was investigated in the mean - field approach of the time - dependent hartree - fock theory by oberacker et al . @xcite .
the calculations carried out at bombarding energies around the fusion barrier exhibit an important difference between the collisions @xmath0ca
+ @xmath1u and @xmath2ca + @xmath1u . in the collisions with neutron rich isotope of calcium , for increasing bombarding energy quasi - fission mechanism diminishes .
hence the cross - section of the composite system formation goes up with bombarding energy as compared to the collisions with the stable calcium projectile . calculations also show in the quasi - fission regime , in both collisions , large number of nucleon transfer takes place from heavy target to light projectile with increasing numbers as the bombarding energy goes up towards the fusion barrier . slightly below the fusion barrier
the mean values of the proton and neutron drift toward symmetry , and can reach large values of @xmath3 , @xmath4 in @xmath0ca induced collisions , and @xmath5 , @xmath6 in @xmath2ca induced collisions , respectively .
the mean - field description of the tdhf can determine only the mean values of the proton and neutron drifts .
on the other hand , it is very interesting and important to provide a description of the fragment mass and charge distributions in the quasi - fission reactions . as seen from fig .
1 , sufficiently below the fusion barrier , di - nuclear structure in the collisions is maintained to a large extend .
this figure shows the density profiles in the reaction plane near maximum overlap obtained in tdhf calculations in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively .
red lines indicate the position of window planes .
the windows are perpendicular to the symmetry lines and pass through the minimum density planes at each instant of the collision process .
consequently , it is possible to describe nucleon exchange as a diffusion process familiar from deep - inelastic collisions @xcite . in this work ,
we investigate nucleon exchange mechanism in the quasi - fission reactions in head - on collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems at a bombarding energy slightly below the fusion barrier by employing the stochastic mean - field ( smf ) approach @xcite .
the smf approach gives rise to a langevin description for neutron and proton exchanges characterized by diffusion and drift coefficients .
we calculate these transport coefficients in the semi - classical framework in terms of the mean - field description of the tdhf solutions without any adjustable parameters . as a result of large contact times in the quasi - fission reaction ,
on the top of large drift toward symmetry , fragment mass and charge distributions have a very broad dispersions in both @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems .
these results indicate that in the collisions of heavy systems , in addition to fusion , nucleon diffusion mechanism may help to populate heavy trans - uranium elements in the quasi - fission regime . in section 2 , we present a brief description of the nucleon diffusion mechanism based on the smf approach . in section 3 ,
we discuss transport coefficients for proton and neutron exchanges . in section 4 ,
the result of calculations presented for central collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems .
conclusions are given in section 5 .
in this work , we consider proton and neutron transfer mechanism in head - on collisions in the @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems at bombarding energies below the fusion barrier near the quasi - fission regime . specifically we carry out calculations for @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems at @xmath7 mev and @xmath8 mev , respectively .
as seen from the tdhf calculations in @xcite , near the quasi - fission regime , colliding ions stick together for a long time . at energies below the fusion barrier , as seen in fig .
1 , system maintains a binary structure to a large extent , and a visible window appears between the projectile - like and target - like partners .
therefore , we can analyze nucleon exchange mechanism , by employing nucleon diffusion concept based on the smf approach . the phenomenological nucleon exchange model and diffusion description has been applied extensively for analyzing deep - inelastic heavy - ion collisions @xcite .
the smf approach provides a more accurate microscopic framework for diffusion mechanism and extracting transport coefficients of relevant macroscopic variables without any adjustable parameters and taking the full collision geometry into account . in the smf approach
, the standard description is extended beyond the mean - field by incorporating the mean - field fluctuations in terms of generating an ensemble of events according to quantal and thermal fluctuations in the initial state ( for details please refer to @xcite ) . in extracting transport coefficients for nucleon exchange , we take the proton and neutron numbers of projectile - like fragments as independent variables
. we can define the proton and neutron numbers @xmath9 , @xmath10 of the projectile - like fragments in each event by integrating over the nucleon density on the projectile side of the window as @xcite , @xmath11 \left(\begin{array}{c } { \rho _ { p}^{\lambda } ( \vec{r},t ) } \\ { \rho _ { n}^{\lambda } ( \vec{r},t ) } \end{array}\right).\ ] ] here , @xmath12 denotes the event label , @xmath13 indicates the location of the window , and @xmath14 , @xmath15 are the local densities of protons and neutrons .
we take @xmath16-axis as the collision direction and the position @xmath13 of the window plane is determined from the tdhf calculations .
as described in @xcite , the local density is projected on the reaction plane and the window is located at the lowest density plane on the neck at each time step .
ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in tdhf calculations .
red lines indicate the position of window planes.,title="fig:",width=302 ] ( -170,45)(a ) ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in tdhf calculations .
red lines indicate the position of window planes.,title="fig:",width=302 ] ( -170,45)(b ) according to the smf approach , the proton and neutron numbers of the projectile - like fragment follows a stochastic evolution according to the following langevin equations , @xmath17 in this expression , the right hand side denotes the proton @xmath18 and neutron @xmath19 drift coefficients in the event @xmath20 , which are determined by the proton current @xmath21 and neutron current @xmath22 through the window . in the smf approach ,
the fluctuating proton and neutron currents are defined as @xcite , @xmath23\rho _ { ji}^{\lambda},\end{aligned}\ ] ] where the summations @xmath24 and @xmath25 run over a complete set of single - particle states for protons and neutrons @xmath26 .
drift coefficients fluctuate from event to event due to stochastic elements of the initial density matrix @xmath27 and also due to the different sets of the wave functions in different events . as a result , there are two sources for fluctuations of the nucleon current : ( i ) fluctuations that arise from the state dependence of the drift coefficients , which may be approximately represented in terms of fluctuations of proton and neutron partition of the di - nuclear system , and ( ii ) the explicit fluctuations @xmath28 and @xmath29 which arise from the stochastic part of proton and neutron currents . for small amplitude fluctuations , we can linearize the drift coefficients around their mean values @xmath30 and @xmath31 to obtain , @xmath32 stochastic part of the fluctuations @xmath28 and @xmath29 are specified by uncorrelated gaussian distributions .
these distribution have zero mean values @xmath33 , @xmath34 and their variances in markovian approximation are determined by @xcite @xmath35 and @xmath36 here , @xmath37 and @xmath38 denote diffusion coefficients for proton and neutron exchange , and the mixed diffusion coefficient is zero @xmath39 . by taking the average over the generated ensemble of the langevin eq .
( [ eq2 ] ) , the mean values evolve according to , @xmath40 the mean values of drift coefficients are determined by the proton and neutron fluxes .
these fluxes are calculated in the mean - field description of the tdhf equations with the mean values of proton and neutron currents , @xmath41n_{j},\end{aligned}\ ] ] where @xmath42 for occupied and @xmath43 unoccupied states . in order to calculate the fluctuations of proton and neutron numbers , we use the stochastic part of eq .
( [ eq2 ] ) around the mean evolution , @xmath44 where the derivatives of drift coefficients are evaluated at the mean trajectory .
the variances and the co - variance of neutron and proton distribution of projectile fragments are defined as @xmath45 , @xmath46 , and @xmath47 .
multiplying both side of langevin equations eq .
( [ eq9 ] ) by @xmath48 and @xmath49 , and taking the ensemble average , we find evolution of the co - variances are specified by the following set of coupled differential equations , @xmath50 these co - variances describe a correlated gaussian function for the proton and neutron distribution @xmath51 of the project - like or the target - like fragments , @xmath52 where @xmath53,\end{aligned}\ ] ] with @xmath54 .
the mean values @xmath55 , @xmath56 denote the mean neutron and proton numbers of the target - like or projectile - like fragments .
the set of coupled equations for co - variances are familiar from the phenomenological nucleon exchange model , and they were derived from the fokker - planck equation for the fragment neutron and proton distributions in the deep - inelastic heavy - ion collisions @xcite .
the proton and neutron diffusion coefficients @xmath57 and @xmath58 , act as sources for determining co - variances in the coupled eqs .
( [ eq10]-[eq12 ] ) . in earlier investigations , expressions these diffusion coefficients in the markovian limit
have been deduced from the smf approach in the semi - classical framework .
we present the results here , and for details we refer @xcite . in the particular case of the head - on collisions ,
the expressions of proton and neutron diffusion coefficients are @xmath59 } \\
{ \bar{f}_{t}^{n } \left(x_{0 } , p_{x } , t\right)\left[1-\frac{1}{\omega } \bar{f}_{p}^{n } \left(x_{0 } , p_{x } , t\right)\right ] } \end{array}\right\ } .\end{aligned}\ ] ] here , @xmath60and @xmath61 are the reduced wigner functions in the collision direction for protons / neutrons , which are obtained by integrating coordinates and momenta over the window plane as discussed in appendix of ref . @xcite .
the wigner functions are calculated with the single - particle wave functions for protons and neutrons , which are originating from target @xmath62 and projectile @xmath63 nuclei , respectively .
the quantity @xmath64 denotes the volume of the phase space on the window plane . in order to solve co - variances from eqs .
( [ eq10]-[eq12 ] ) , in addition to the diffusion coefficients @xmath65 and @xmath66 , we need to know the rate of change of drift coefficients in the vicinity of their mean values . according to the smf approach , in order to calculate rates of the drift coefficients ,
we should calculate neighboring events in the vicinity of the mean - field event . here , instead of such a detailed description
, we employ the fluctuation - dissipation theorem , which provides a general relation between the diffusion and drift coefficients in the transport mechanism of the relevant collective variables as described in the phenomenological approaches @xcite .
proton and neutron diffusions in the n - z plane are driven in a correlated manner by the potential energy surface of the di - nuclear system . as a consequence of the symmetry energy ,
the diffusion in direction perpendicular to the beta stability valley takes place rather rapidly leading to a fast equilibration of the charge asymmetry , and diffusion continues rather slowly along the beta - stability valley .
2 illustrates very nicely the expected the mean - drift paths in the central collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems .
the drift paths is obtained from the solution of the mean - field description of the tdhf equations .
the di - nuclear system drifts towards symmetry during long contact time , but separates before reaching to the symmetry . following this observation and borrowing an idea from references @xcite , we parameterize the @xmath67 and @xmath68 dependence of the potential energy surface of the di - nuclear system in terms of two parabolic forms
, @xmath69 here , @xmath70 , @xmath71 and @xmath72 denotes the angle between beta stability valley and the @xmath73 - axis in the @xmath74 plane .
we can determine these angles from the mean - drift paths in fig .
2 . the quantities @xmath75 and @xmath76 denotes the equilibrium values of the neutron and proton numbers , which are approximately determined by the average values of the neutron and proton numbers of the projectile and target ions , @xmath77 and @xmath78 .
the first term in this expression describes a strong driving force perpendicular to the beta stability valley , while the second term describes a relative weak driving force toward symmetry along the valley .
plane in the central collision of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in tdhf calculations.,title="fig:",width=302 ] ( -40,50)(a ) plane in the central collision of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in tdhf calculations.,title="fig:",width=302 ] ( -40,50)(b ) following from the fluctuation - dissipation theorem , it is possible to relate the proton and neutron drift coefficients to the diffusion coefficients and the associated driving forces , in terms of the einstein relations as follows @xcite , @xmath79\end{aligned}\ ] ] and @xmath80.\end{aligned}\ ] ] here , the temperature @xmath81 is absorbed into coefficients @xmath82 and @xmath83 , consequently temperature does not appear as a parameter in the description .
we can determine @xmath82 and @xmath83 by matching the mean values of neutron and proton drift coefficients obtained from the tdhf solutions .
in this manner , microscopic description of the collision geometry and details of the dynamical effects are incorporated into the drift coefficients . in the liquid drop picture , the potential energy surfaces in perpendicular to the stability valley and along the stability valley have parabolic behaviors .
therefore we expect both coefficients @xmath84 and @xmath85 to be positive . however , as a result of the quantal effects arising mainly from the shell structure , we observe that these coefficients exhibit fluctuations as a function of time , which can also be viewed as a function of the relative distance between ions . in eqs .
( [ eq8]-[eq10 ] ) for co - variances , we also need derivatives of drift coefficients with respect to proton and neutron numbers of projectile - like fragments . a great advantage of this approach , we can easily calculate these derivatives from drift coefficients to yield , @xmath86
employing the diffusion mechanism described in the previous section , we investigate nucleon exchange mechanism in central collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems near the quasi - fission region at bombarding energies @xmath7 mev and @xmath8 mev , respectively . in the collision geometry ,
the elongation of deformed @xmath87u target nucleus is taken as perpendicular direction to the beam ( side collision ) .
these energies are slightly below the fusion barriers . as a result , colliding ions stick together with a visible neck for a long time , and separate without forming a compound nucleus .
time dependent single - particle wave functions are determined from solutions of the tdhf equations by employing the code developed by p. bonche et al . with the sly4d skyrme effective interactions @xcite .
2 shows the mean - drift paths of the projectile - like fragments in @xmath88 plane obtained in the tdhf calculations in the collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems . after a rapid charge equilibration , the system drift toward the symmetric fragmentation , which is specified by proton and neutron numbers @xmath89 and @xmath90 for the @xmath0ca + @xmath1u system and @xmath89 and @xmath91 for the @xmath0ca + @xmath1u system .
the tangent of angle made by the mean - drift path with the @xmath73- axis is about @xmath92 for both systems .
( dashed line ) and neutron drift @xmath93 ( solid line ) coefficients in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7mev and @xmath8 mev , respectively , obtained in tdhf calculations.,title="fig:",width=302 ] ( -35,135)(a ) ( dashed line ) and neutron drift @xmath93 ( solid line ) coefficients in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7mev and @xmath8 mev , respectively , obtained in tdhf calculations.,title="fig:",width=302 ] ( -35,135)(b ) we use these values in parameterization of the driving potential energy in eq .
( [ eq14 ] ) for both systems . in both collisions ,
the system stick together approximately from an initial touching time @xmath94 seconds until separation time at @xmath95 seconds .
we observe that during the contact time , the mean number of proton and neutron drifts are about @xmath96 , @xmath97 in the @xmath0ca + @xmath1u system , and @xmath98 , @xmath99 in the @xmath2ca + @xmath1u system , respectively .
even though the sticking time is about the same in both collisions , the smaller drift in @xmath2ca induced collision is due to the smaller bombarding energy by about @xmath100 mev . dashed and solid lines in fig .
3 show the proton and neutron drift coefficients in the @xmath0ca + @xmath1u system ( a ) and in the @xmath2ca + @xmath1u system ( b ) as a function of collision time , which are obtained by the mean - field description of the tdhf . because of small amplitude vibrations of the window positions , drift coefficients exhibit small fluctuations in time .
this figure illustrates smoothed drift coefficients obtained by averaging over short time intervals . probably due to shell effects , in @xmath0ca induced collision protons
exhibit a rapid drift toward asymmetry during the initial phase of the collision , followed by persistent drift toward symmetry in both proton and neutron numbers in both systems .
we determine the dimensionless parameters @xmath101 and @xmath102 in eqs .
( [ eq15],[eq16 ] ) by matching the proton and neutron drift coefficients to the results of obtained in tdhf calculations . in @xmath0ca induced collision
, the parameter @xmath101 during the early times takes relatively large positive values around @xmath103 until about @xmath104 seconds while at later times it takes small fluctuating values around @xmath105 . on the other hand parameter @xmath102
take much smaller positive values @xmath106 as expected , also fluctuating in time .
these coefficients exhibit similar behavior in the @xmath2ca induced collision .
as noted above , we believe that fluctuations in these parameters are due to quantal effects arising mainly from the underlying shell structure .
( dashed line ) and neutron diffusion @xmath107 ( solid line ) coefficients in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in semi - classical framework of the smf approach.,title="fig:",width=302 ] ( -35,135)(a ) ( dashed line ) and neutron diffusion @xmath107 ( solid line ) coefficients in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively , obtained in semi - classical framework of the smf approach.,title="fig:",width=302 ] ( -35,135)(b ) we calculate diffusions coefficients in eq .
( [ eq13 ] ) for proton and neutron exchange in the semi - classical framework by employing the wigner functions , which are determined in terms of the time - dependent single - particle wave functions of the tdhf solutions .
the reduced wigner functions are obtained by integrating over the phase - space volume on the window plane .
the reduced wigner functions exhibit fluctuations as a function of single - particle momentum and can take small negative values in classically forbidden regions .
we eliminate these fluctuations and negative values of wigner functions by performing a smoothing procedure as outline in @xcite . dashed and solid lines in fig .
4 show the proton and neutron diffusion coefficients in @xmath0ca ( a ) and @xmath2ca ( b ) induced collisions as a function of time .
diffusion coefficients also exhibit small fluctuations in time due to small amplitude vibrations of the window positions .
this figure illustrates smoothed diffusion coefficients obtained by averaging over short time intervals .
mainly as a result of the coulomb barrier , the neutron diffusion coefficients are nearly twice as large as compared to the proton diffusion coefficients , in both systems .
( dashed line ) , neutron dispersion @xmath108 ( solid line ) and mixed dispersion @xmath109 ( dotted line ) of the fragment distributions ( projectile - like or target - like ) in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively.,title="fig:",width=302 ] ( -35,135)(a ) ( dashed line ) , neutron dispersion @xmath108 ( solid line ) and mixed dispersion @xmath109 ( dotted line ) of the fragment distributions ( projectile - like or target - like ) in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath7 mev and @xmath8 mev , respectively.,title="fig:",width=302 ] ( -35,135)(b ) we solve the coupled differential equations ( 10 - 12 ) for co - variances with the initial conditions @xmath110 .
the results are plotted in fig . 5 as a function time .
dashed and solid lines in the figure show proton and neutron dispersions @xmath111 , @xmath112 and the square - root of co - variances @xmath113 in @xmath0ca ( a ) and @xmath2ca ( b ) induced collisions , respectively .
the co - variances @xmath114 take negative values until @xmath115 seconds and @xmath116 seconds in @xmath0ca and @xmath2ca induced collisions , respectively . therefore , the square - root of co - variances are not shown in these time intervals .
we observe that at the separation instant in the @xmath0ca + @xmath1u collision , the dispersions of proton and neutron distributions are @xmath117 and @xmath118 , and the co - dispersion is @xmath119 .
the proton dispersion is nearly factor of 2.0 larger than the mean number of proton drift , while neutron dispersion is about 1.5 larger than the mean number of neutron drift .
target nucleus losses about 10 protons and 21 neutrons , but it can gain more protons and neutron by diffusion mechanism . on the other hand , at the separation instant in the @xmath2ca + @xmath1u collision , the dispersions of proton and neutron distributions are @xmath120 and @xmath121 , and the co - dispersion is @xmath122 .
the proton dispersion is nearly factor of 2.5 larger than the mean number of proton drift , while neutron dispersion is about 2.0 larger than the mean number of neutron drift .
target nucleus losses about 6 protons and 9 neutrons on the average , but it can gain more protons and neutron by diffusion mechanism . therefore , in these collisions , diffusion mechanism near quasi - fission regime can help to populate elements heavier than uranium target nucleus .
probability distributions of the projectile - like or target - like fragments at the exit channel are determined by the correlated gaussian of eq .
( [ eq12.1 ] ) , in which the magnitudes of co - variances and mean - values are taken at the separation instant of the collision .
6 shows equal probability lines for population of target - like fragments at the exit channel in the @xmath88 plane in the @xmath0ca ( a ) and @xmath2ca ( b ) induced collisions .
probability of populating a fragment with neutron and proton numbers @xmath123 relative to populating the fragment with mean neutron and proton numbers is determined by @xmath124 , where @xmath125 indicate numbers on the equal probability lines in fig .
6 . in this figure dots at the centers of ellipses
indicate the elements with the mean neutron and proton numbers at the exit channel .
the mean values of neutron and proton numbers at the exit channel are @xmath126 and @xmath127 in @xmath0ca and @xmath2ca induced collisions , respectively . as an example
, we can see from this figure , the probability of populating a heavy trans - uranium element with @xmath128 relative to the populating the element with mean neutron and proton numbers is about @xmath129 in the @xmath130ca + @xmath1u collision .
the relative population probability of the same element in the @xmath131ca + @xmath1u collision has about the same magnitude .
7 illustrates the dispersion @xmath132 of total mass number distributions of the projectile - like fragments or target - like fragments as a function of time in @xmath0ca ( a ) and @xmath2ca ( b ) induced collisions , respectively .
the total dispersion is calculated from @xmath133 . in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b).,title="fig:",width=340 ] ( -69,51)(a ) in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b).,title="fig:",width=340 ] ( -79,51)(b ) of the fragment distribution of the fragment distributions ( projectile - like or target - like ) in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath134 mev and @xmath135 mev , respectively.,title="fig:",width=302 ] ( -40,50)(a ) of the fragment distribution of the fragment distributions ( projectile - like or target - like ) in the central collisions of @xmath0ca + @xmath1u ( a ) and @xmath2ca + @xmath1u ( b ) systems at energies @xmath134 mev and @xmath135 mev , respectively.,title="fig:",width=302 ] ( -40,50)(b ) we note that the correlated gaussian function of eq .
( [ eq12.1 ] ) , which is specified by the first two moments , provides an approximate description of the fragment population .
the approximation is reasonable within the range of @xmath136 around the center points , but becomes gradually unreasonable as we move out from the center points near to the tail of the distribution functions .
for example , as seen from the upper ends of @xmath137 lines in fig .
6 , we observe finite but small probabilities for populating fragments even exceeding the total mass of the system .
therefore , in particular near the tail region , more accurate description of the fragment population probability is required . in the present work
, we do not discuss the energy dissipation and the excitation energy deposited in the populated fragments during nucleon diffusion process .
however , we can provide an estimate of the excitation energy deposited in the fragments with mean values of protons and neutrons in the exit channel .
it is possible to calculate the total excitation energy @xmath138 deposited in the mean fragments at the exit channel according to , @xmath139 where @xmath140 denotes the @xmath140-value and tke is the asymptotic value of the total kinetic energy at the exit channel .
we calculate the tke for the mean - fragment exit channel by employing the tdhf [ 11 ] .
since the interaction time is very long , the initial relative kinetic energy totally dissipates and the tke at the exit channel is essentially determined by the coulomb repulsion . because of very long interaction times , it is reasonable to consider the equilibration of the excitation energy . under this circumstance ,
the excitation energy between the mean fragments is shared in proportion to their mass numbers , @xmath141 and @xmath142 in @xmath0ca + @xmath1u system , @xmath143 and @xmath144 with @xmath145 . in @xmath2ca
+ @xmath1u system , @xmath146 and @xmath147 with @xmath148 .
calculation gives for the excitation energies @xmath149 mev , @xmath150 mev in the @xmath0ca + @xmath1u system , and @xmath151 mev , @xmath152 mev in the @xmath2ca + @xmath1u system .
we investigate nucleon exchange mechanism in the central collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems below the fusion barrier near the quasi - fission regime . sufficiently below the fusion barrier ,
colliding system maintains a di - nuclear structure . as a result ,
it is possible to describe nucleon exchange as a diffusion mechanism which is familiar from the description of deep - inelastic heavy - ion collisions .
the standard mean - field description based on the tdhf equations determines the mean drift path in the n - z plane . in order to describe fluctuations around the drift path
, we employ the microscopic basis of the smf approach , which incorporates the mean - field fluctuations beyond the average description of the standard tdhf .
we calculate diffusion coefficients for proton and neutron transfer mechanisms with the help of the smf approach in the semi - classical framework .
proton and neutron diffusion occurs in the n - z plane in a correlated manner according to the potential energy surface of the di - nuclear system .
the potential energy surface along the beta - stability line and perpendicular to the stability line are parameterized in terms of two parabolic forms . employing einstein relations
, we deduce simple analytical expressions for proton and neutron drift coefficients .
parameters of the drift coefficients are determined with help of the mean drift path obtained in the tdhf calculations .
we determine the co - variances of the neutron and proton distributions of the projectile - like fragments .
calculations show that after a fast charge equilibration , large amount of mean drift in the numbers of protons and neutrons toward symmetry .
we find large dispersions of the proton and neutron distributions of the projectile - fragments during very long interaction times . the mean numbers of proton and neutron of the target nucleus decrease due to drift toward symmetry .
on the other hand , large values of proton and neutron dispersions indicate that diffusion mechanism helps to populate heavy trans - uranium elements near the quasi - fission regime .
s.a . gratefully acknowledges tubitak and the middle east technical university for partial support and warm hospitality extended to him during his visits .
authors gratefully acknowledge useful discussions with d. lacroix and a. s. umar . this work is supported in part by us
doe grant no .
de - fg05 - 89er40530 , and in part by tubitak grant no .
113f061 .
99 y. t. oganessian et al .
lett . * 104 * , 142502 ( 2010 ) .
y. t. oganessian et al . , phys .
* 109 * , 162501 ( 2012 ) .
j. khuyagbaatar et al .
lett . * 112 * , 172501 ( 2014 ) .
s. hofmann and g. munzenberg , rev .
* 72 * , 733 ( 2000 ) .
s. hofmann et al . , eur .
j. a * 14 * , 147 ( 2002 ) .
sham et al . , z. phys .
a * 319 * , 113 ( 1984 ) . k .- h . schmidt and morawek ,
. phys . * 54 * , 949 ( 1991 ) .
g. g. adamian , n. v. antonenko , and w. scheid , phys .
c * 68 * , 034601 ( 2003 ) . v. zagrebaev and w. greiner .
g * 34 * , 2265 ( 2007 ) .
y. aritomo , phys . rev .
c * 80 * , 064604 ( 2009 ) .
v. e. oberacker , a. s. umar , and c. simenel , phys
c * 90 * , 054605 ( 2014 ) .
j. randrup , nucl .
phys . a * 327 * , 490 ( 1979 ) .
s. ayik , phys .
b * 658 * , 174 ( 2008 ) .
s. ayik , k. washiyama , and d. lacroix , phys .
c * 79 * , 054606 ( 2009 ) .
k. washiyama , s. ayik , and d. lacroix , phys .
c * 80 * , 031602(r ) ( 2009 ) . b. yilmaz , s. ayik , d. lacroix , and k. washiyama , phys . rev .
c * 83 * , 064615 ( 2011 ) . b. yilmaz , s. ayik , d. lacroix , and o. yilmaz , phys . rev .
c * 90 * , 024613 ( 2014 ) .
s. ayik et al . , phys .
c * 91 * , 054601 ( 2015 ) .
w. u. schroder , j. r. huizenga , and j. randrup , phys .
b * 98 * , 355 ( 1981 ) .
a. c. merchant and w. norenberg , phys .
b * 104 * , 15 ( 1981 ) .
a. c. merchant and w. norenberg , z. phys .
a * 308 * , 315 ( 1982 ) .
kim , t. otsuka , and p. bonche , j. phys .
g * 23 * , 1267 ( 1997 ) . | nucleon exchange mechanism is investigated in the central collisions of @xmath0ca + @xmath1u and @xmath2ca + @xmath1u systems near the quasi - fission regime in the framework of the stochastic mean - field ( smf ) approach . sufficiently below the fusion barrier , di - nuclear structure in the collisions
is maintained to a large extend .
consequently , it is possible to describe nucleon exchange as a diffusion process familiar from deep - inelastic collisions .
diffusion coefficients for proton and neutron exchange are determined from the microscopic basis of the smf approach in the semi - classical framework .
calculations show that after a fast charge equilibration the system drifts toward symmetry over a very long interaction time .
large dispersions of proton and neutron distributions of the produced fragments indicate that diffusion mechanism may help to populate heavy trans - uranium elements near the quasi - fission regime in these collisions . |
the behavior of elementary charged particles , like electrons and positrons is governed by quantum electrodynamics ( qed ) . in this theory
the quantum particles interact via the exchange of the quanta of light , i.e. the photons @xcite .
thus in principle we have to consider the quantum nature of the charged particles as well as of the light field .
however , in several important cases we can focus almost exclusively on either the charged particles or the photons , while employing crude approximations for the other degrees of freedom . in condensed matter physics and quantum chemistry the quantum nature of light can usually be ignored and the interaction between the charged quantum particles is approximated by the instantaneous coulomb interaction . however , even then the resulting quantum mechanical equations ( usually the many - body schrdinger equation ) , where the electromagnetic fields are treated classically through the solution of the maxwell equations , are solvable only for very simple systems .
this lies ultimately in our incapability of handling the huge number of degrees of freedom of many - particle systems and consequently in our inability to determine the many - body states .
this so - called many - body problem spawned a lot of interest into the question whether one can devise a closed set of equations for reduced quantities which do not involve the explicit solution of the full quantum mechanical equations and in which the many - body correlations can be approximated efficiently .
pursuits in this direction have led to various approaches such as , among others , many - body green s function theories @xcite , ( reduced ) density - matrix theories @xcite and density - functional theories @xcite .
these approaches differ in the complexity of the reduced quantity , which is used to calculate the various observables of interest
. especially density - functional theories , which are based on the simplest of those ( functional ) variables , the one - particle density ( current ) , have proven to be exceptionally successful @xcite .
their success can be attributed to the unprecedented balance between accuracy and numerical feasibility @xcite , which allows at present to treat several thousands of atoms @xcite .
although the different flavors of density - functional theories cover most of the traditional problems of physics and chemistry ( including approaches that combine classical maxwell dynamics with the quantum particles @xcite ) , by construction these theories can not treat problems involving the quantum nature of light . in quantum optics ,
on the other hand , the focus is on the photons , while usually simple approximations for the charged particles are employed , e.g. a few - level approximation .
however , even in this situation the solution of the resulting equations @xcite is only possible in simple cases ( again due to the large number of degrees of freedom ) and usually simplified model hamiltonians , e.g. the dicke model realized in a cavity @xcite , are employed to describe these physical situations .
already the validity of these effective hamiltonians and their properties can be a matter of debate @xcite and often further simplifications are adopted such as the jaynes - cummings model in the rotating - wave approximation .
the rapid progress in quantum - optical experiments on the other hand , especially in the field of cavity qed @xcite and circuit qed @xcite , allows to study and control multi - particle systems ultra - strongly coupled to photons @xcite , where such a simple approximative treatment is no longer valid @xcite .
this new regime of light - matter interaction is widely unexplored for , e.g. , molecular physics and material sciences @xcite .
possibilities like altering and strongly influencing the chemical reactions of a molecule in the presence of a cavity mode or setting the matter into new non - equilibrium states with novel properties , e.g. light - induced superconductivity @xcite , arise . specifically in such situations an oversimplified treatment of the charged particles may no longer be allowed and an approach that considers both , the quantum nature of the light field as well as of the charged particles is needed . in this work we give a comprehensive derivation of an exact and numerically feasible method that generalizes ideas of time - dependent density - functional theory ( tddft ) .
this method bridges the gap between the above two extreme cases and provides a scheme to perform ab - initio calculations of quantum particles coupled to photons .
the electron - photon generalization of tddft in describing non - relativistic many - electron systems coupled to photon modes of mesoscopic cavities was introduced in ref .
here we provide a general framework describing fully coupled electron - photon systems in most possible regimes / systems ranging from effective model hamiltonians to strongly relativistic cases , which has been introduced in refs . @xcite . for clarity
we divide the following presentation in two parts : we first demonstrate the basic ideas in a simple model system and then show how these concepts can be used in the case of general coupled matter - photon problems .
a summary of all findings of the present work for the time - dependent density - functional description of qed at different levels of approximations , namely the basic variables , initial conditions and fundamental kohn - sham multicomponent equations is given in appendix [ app : overview ] .
we start considering a simple model system for charged matter coupled to photons : the two - site hubbard model interacting with one photonic mode . by employing density - functional ideas
we show how one can solve this quantum - mechanical problem without the need to explicitly calculate the complex many - body wave function . instead
, we derive equations of motion for a pair of reduced quantities from which all physical observables can by determined .
we demonstrate that these equations have unique solutions and can be used to calculate the basic reduced quantities ( here the basic pair of reduced quantities is the charge density of the particle and the potential induced by the photons ) of the coupled problem . therefore we here reformulate the coupled matter - photon problem in terms of an effective theory , that we call in the following a model of quantum electrodynamical density - functional theory ( qedft ) .
since an explicit calculation of the coupled wave function is not needed , this approach allows to determine properties of the matter - photon system in a numerically feasible way .
we introduce an new kohn - sham scheme to approximate the unknown functionals in the basic equations of motion and present results for a simple approximation .
we compare these results to the exact kohn - sham functionals and identify shortcomings and indicate improvements . based on the ideas developed in the first part of this work we repeat the steps illustrated in our example but
now we construct a density - functional reformulation for the full theory of qed @xcite .
we show that a straightforward approach based on the current and the potential leads to problems and that a consistent density - functional reformulation of qed has to be based on the polarization and the potential which is generated by the photons .
this approach to the fully coupled qed problem we denote as relativistic qedft , and we present the corresponding kohn - sham construction and give the simplest approximation to the unknown functionals . in the following we
then demonstrate how relativistic qedft reduces in the non - relativistic limit to its non - relativistic version of the corresponding non - relativistic hamiltonian . by employing further approximations on the matter system or on the photon field
a family of different approximate qedfts is introduced , which are consistent with their respective approximate hamiltonians . at this level
we recover the theory of ref . @xcite . in lowest order
we rederive the model qedft of the first part of this work .
therefore , we demonstrate how all different flavours of qedft are just approximations to relativistic qedft in the same manner as different physical hamiltonians are merely approximations to the qed hamiltonian .
furthermore , by ignoring all photonic degrees of freedom , we find the standard formulations of tddft which are extensively used in the electronic - structure community @xcite .
* outline * in sec .
[ sec : model ] we investigate the qedft reformulation of a simple model of one particle coupled to one mode in great detail .
the developed ideas are then employed in sec .
[ sec : relqedft ] to derive a qedft reformulation of qed . in sec .
[ sec : nonrelqedft ] we show how all different qedft reformulations are approximations to relativistic qedft .
we conclude and give an outlook in sec . [
sec : conclusion ] .
in this section , we introduce the basic formulation and underlying ideas of qedft . by employing a model hamiltonian
, we can almost exclusively focus on the density - functional ideas that allow a reformulation of the wave - function problem in terms of simple effective quantities .
we first identify the pair of external and internal variables and then show that both are connected via a bijective mapping . as a consequence ,
all expectation values become functionals of the initial state and the internal pair .
this allows for a reformulation of the problem in terms of two coupled equations for the internal pair .
then we introduce the kohn - sham construction as a way to find approximations to the unknown functionals , and show first numerical results . to describe the dynamics of particles coupled to photons we solve an evolution equation of the form @xmath0 for a given initial state @xmath1 . here
@xmath2 with @xmath3 and the standard relativistic ( covariant ) notation @xmath4 ( see also appendix [ app : con ] for notational conventions ) .
the corresponding hermitean hamiltonian has the general form @xmath5 where the depedence of the total hamiltonian on @xmath6 indicates an explicit time - dependence . here
the ( time - independent ) hamiltonian @xmath7 describes the kinetic energy of the particles , i.e. how they would evolve without any perturbation , and @xmath8 is the energy of the photon - field .
the third term describes the coupling between the ( charged ) particles and the photons by the charge current @xmath9 and the maxwell - field operators @xmath10 ( where the einstein sum convention with the minkowski metric @xmath11 is implied and greek letters refer to four vectors , e.g. @xmath12 , while roman letters are restricted to spatial vectors only , e.g. @xmath13 ) .
this term is frequently called the minimal - coupling term and arises due to the requirement of a gauge - invariant coupling between the particles and the photon field .
the specific form of the operators @xmath9 and @xmath10 depends on the details of the physical situation . finally , the last term describes how the particles interact with a ( in general time - dependent ) classical external vector potential @xmath14 and how the photons couple to a ( in general time - dependent ) classical external current @xmath15 .
while we usually have no control over how the particles and photons evolve freely or interact , i.e. the first three terms of the hamiltonian ( [ basicham ] ) , we have control over the preparation of the initial state @xmath1 and the external fields @xmath16 .
therefore , all physical wave functions , i.e. found by solving eq .
( [ schrdinger ] ) , can be labeled by their initial state and _ external pair _
@xmath16 , @xmath17;t ) \ , \right\rangle}.\end{aligned}\ ] ] however , for any but the simplest systems the ( numerically exact ) solution of eq .
( [ schrdinger ] ) is not feasible .
even if we decouple the matter part from the photons by employing the coulomb - approximation ( i.e. describing the exchange of photons by the respective lowest - order propagator ) the resulting problem is far from trivial . in this subsection , we introduce a simple model of charged particles coupled to photons .
we discuss the basic concepts of a density - functional - type reformulation , identify the pair of conjugate variables and then deduce the fundamental equations of motion on which we base our qedft reformulation . in order to demonstrate the basic ideas of a qedft we employ the simplest yet non - trivial realization of one charged particle coupled to photons : a two - site hubbard model coupled to one photonic mode .
the resulting hamiltonian ( see appendix [ app : hubbard ] for a detailed derivation ) reads as @xmath18,\end{aligned}\ ] ] where the kinetic energy of the charged particle is given by @xmath19 and the energy of the photon mode reads @xmath20 here @xmath21 is the hopping parameter between the two sites , @xmath22 is the frequency of the photonic mode and @xmath23 are the pauli matrices that obey the usual fermionic anti - commutation relations . the photon creation and annihilation operators ( @xmath24 and @xmath25 , respectively ) obey the usual bosonic commutation relations .
the current operator is proportional to the dipole - moment operator , i.e. it is connected to the zero component @xmath26 of the general four - current operator @xmath9 . to highlight the analogy in structure to the general case discussed in the later sections , we give it the units of a current and denote it by @xmath27
is defined by @xmath28 where @xmath29 is a characteristic length - scale of the matter - part and @xmath30 is a dimensionless coupling constant .
+ the operator for the conjugate potential is actually proportional to the electric field as can be seen from the derivations in appendix [ app : hubbard ] .
this is , because in the course of approximations one employs the length - gauge and thus transforms from the potential to the electric field .
however , to highlight the analogy in structure to the general case discussed in the later sections , we give it the units of the potential and denote it by @xmath31 is given by @xmath32 where @xmath33 is the length of the cubic cavity .
further , the current operator couples to the external potential @xmath34 and the potential operator to the external current @xmath35 .
these are the two ( classical ) external fields that we can use to control the dynamics .
if we then fix an initial state @xmath1 and choose an external pair @xmath36 , we usually want to solve eq .
( [ schrdinger ] ) with the hamiltonian given by eq .
( [ hubbard ] ) .
the resulting wave function , given in a site basis @xmath37 for the charged particle and a fock number - state basis @xmath38 for the photons @xmath39;t ) \ , \right\rangle } = \sum_{x=1}^{2 } \sum_{n=0}^{\infty } c_{xn}(t ) { \left| \ , x \ , \right\rangle}\otimes { \left| \ , n \ , \right\rangle},\end{aligned}\ ] ] depends on the initial state and the external pair @xmath36 .
thus , by varying over all possible combinations of pairs @xmath36 , we scan through all physically allowed wave functions starting from a given initial state .
hence , we parametrize the relevant , i.e. physical , time - dependent wave functions by @xmath1 and @xmath36 .
since the wave functions have these dependencies , also all derived expressions , e.g. the expectation values for general operators @xmath40 @xmath41,t ) = { \left\langle \ , \psi(t)|\hat{o } \,|\ , \psi(t ) \ , \right\rangle},\end{aligned}\
] ] are determined by the initial state and the external pair @xmath36 . the idea of an _ exact _ effective theory like qedft is now , that we identify a different set of fundamental variables , which also allow us to label the physical wave functions ( and their respective observables ) , and that we have a closed set of equations for these new ( functional ) variables , which do not involve the full wave functions explicitly .
such a functional - variable change is similar to a coordinate transformation , say from cartesian coordinates to spherical coordinates .
this can only be done if every point in one coordinate system is mapped uniquely to a point in the other coordinate system . for a functional - variable change
we thus need to have a one - to - one correspondence , i.e. bijective mapping , between the set of ( allowed ) pairs @xmath36 and some other set of functions ( while we keep the initial state fixed ) . to identify the simplest new functional variables one usually employs arguments based on the legendre transformation @xcite .
that is why these new functional variables are often called conjugate variables .
we will consider this approach in the next sections where we investigate general qedft , and also show how one can determine the conjugate variables of this model system from more general formulations of qedft .
for this simple model we simply state that a possible pair of conjugate variables is @xmath42 . in the next subsection
we show that this functional variable - transformation is indeed allowed , i.e. @xmath43;t ) \ , \right\rangle}.\end{aligned}\ ] ] the main consequence of this result is , that from only knowing these three basic quantities we can ( in principle ) uniquely determine the full wave function .
accordingly , every expectation value becomes a unique functional of @xmath1 and @xmath42 .
thus , instead of trying to calculate the ( numerically expensive ) wave function , it is enough to determine the _ internal pair _
@xmath42 for a given initial state . an obvious route to then also find a closed set of equations for these new variables is via their respective equations of motion .
these equations will at the same time be used to prove the existence of the above change of variables , i.e. that the wave function is a unique functional of the initial state and the internal pair @xmath42 .
to find appropriate equations we first apply the heisenberg equation of motion once and find @xmath44 where @xmath45 .
yet these two equations are not sufficient for our purposes : we need equations that explicitly connect @xmath46 and @xmath42
. therefore we have to go to the second order in time , @xmath47 where @xmath48 @xmath49 and @xmath50 . here , eq .
( [ discerteforce ] ) is the discretized version of @xmath51 of standard tddft @xcite , and eq .
( [ discertemaxwell ] ) is the inhomogeneous maxwell equation for one photon mode @xcite . in the previous subsection
we have stated that @xmath42 and @xmath36 are the possible conjugate pair of the model hamiltonian ( [ hubbard ] ) . in this subsection
we want to demonstrate that indeed this holds true and that we can perform a variable transformation from the external pair @xmath36 to the internal pair @xmath42 .
what we need to show is , that for a fixed initial state @xmath1 the mapping @xmath52 is bijective , i.e. if @xmath53 then necessarily for the according expectation values @xmath54 .
to do so we first note , that in the above equations of motion every expectation value is by construction a functional of @xmath36 for a fixed initial state , @xmath55;t ) = - \frac{4 t^2_{\mathrm{kin } } } { \hbar^2 c^2 } j([a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) \\ & + \lambda \langle \hat{n } \a \rangle ( [ a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) + n ( [ a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) a_{{{\rm ext}}}(t ) , \nonumber \\ \label{discertemaxwellexp } \partial_0 ^ 2 a&([a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) = - k^2 a([a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) \nonumber \\ & + \frac{\mu_0 c}{l^3}\left ( \lambda j([a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) + j_{{{\rm ext}}}(t)\right),\end{aligned}\ ] ] i.e. they are generated by a time - propagation of @xmath1 with a given external pair @xmath36 . suppose now , that we fix the expectation values of the internal variables @xmath56 , i.e. we do not regard them as functionals but rather as functional variables .
then the above eqs .
( [ discerteforceexp ] ) and ( [ discertemaxwellexp ] ) become equations for the pair @xmath36 that produce the given internal pair @xmath42 via propagation of the initial state @xmath1 , i.e. @xmath57;t ) \nonumber \\ & + n ( [ a_{{{\rm ext}}},j_{{{\rm ext}}}];t ) a_{{{\rm ext}}}(t ) , \\
\label{discertemaxwellfun } \partial_0 ^ 2 a(t ) = & -k^2 a(t ) + \frac{\mu_0 c}{l^3}\left ( \lambda j(t ) + j_{{{\rm ext}}}(t)\right).\end{aligned}\ ] ] obviously these equations can only have a solution , if the given internal variables are consistent with the initial state , i.e. @xmath58 here we have used the definition @xmath59 and every internal pair @xmath42 that we consider is subject to these boundary conditions .
thus , the mapping ( [ pairmapping ] ) is bijective , if the corresponding eqs .
( [ discerteforcefun ] ) and ( [ discertemaxwellfun ] ) , which connect the internal pair @xmath42 with the external pair @xmath36 , allow for one and only one solution pair .
let us first note that for a given pair @xmath42 , eq .
( [ discertemaxwellfun ] ) uniquely determines the external current @xmath60 by @xmath61 thus , the original problem reduces to the question whether eq .
( [ discerteforcefun ] ) determines @xmath34 uniquely .
the most general approach to answer this question is via a fixed - point procedure similar to ref .
@xcite . in the case of a discretized schrdinger equation like eq .
( [ hubbard ] ) it should also be possible to apply a rigorous approach based on the well established theory of nonlinear ordinary differential equations @xcite . however , for simplicity we follow ref
. @xcite and employ the standard strategy of @xcite which restricts the allowed external potentials @xmath62 to being taylor - expandable in time , i.e. @xmath63 from eq .
( [ discerteforce ] ) we can find the taylor - coefficients of @xmath64 ( if they exist ) by @xmath65 where the terms @xmath66 and @xmath67 are given by their respective heisenberg equations at @xmath68 and only contain taylor coefficients of @xmath69 for @xmath70 .
now , assume that we have two different external potentials @xmath71 .
this implies , since we assumed taylor - expandability of @xmath72 and @xmath73 , that there is a lowest order @xmath74 for which @xmath75 for all orders @xmath70 ( even though the individual @xmath76 and @xmath77 might not exist ) it necessarily holds that @xmath78 but for @xmath74 we accordingly find that @xmath79 provided we choose the initial state such that ( @xmath80 )
. consequently , @xmath81 infinitesimally later for two different external potentials @xmath71 .
therefore , eq . ( [ discerteforcefun ] ) allows only one solution and the mapping @xmath82 is bijective . as a consequence , since every expectation value of the quantum system becomes a functional of the internal pair @xmath42 , in the above eqs .
( [ discerteforcefun ] ) and ( [ discertemaxwellfun ] ) we can perform a change of variables and find @xmath83;t ) + n ( [ j , a];t ) a_{{{\rm ext}}}(t ) , \\ \label{discertemaxwellintfun } \partial_0 ^2 a(t ) & = -k^2 a(t ) + \frac{\mu_0 c}{l^3}\left ( \lambda j(t ) + j_{{{\rm ext}}}(t)\right).\end{aligned}\ ] ] these coupled evolution equations have unique solutions @xmath56 for the above initial conditions ( [ initialcurrent ] ) and ( [ initialpotential ] ) .
therefore we can , instead of solving for the many - body wave function , solve these non - linear coupled evolution equations for a given initial state and external pair @xmath36 , and determine the current and the potential of the combined matter - photon system from which all observables could be computed .
this is an exact reformulation of the model in terms of the current and the potential of the combined system only . in the previous subsection
we have derived a qedft reformulation in terms of the current and the potential . while the equation that determines the potential @xmath84 is merely the classical maxwell equation , and every term
is known explicitly , the equation for the current contains implict terms .
therefore , to solve these coupled equations in practice , we need to give appropriate explicit approximations for the implict terms .
approximations based on @xmath42 directly would correspond to a thomas - fermi - type approach to the model . as known from standard density - functional theory ,
such approximations are in general very crude and hard to improve upon .
a more practical scheme is based on the kohn - sham construction , where an auxiliary quantum system is used to prescribe explicit approximations .
however , the numerical costs of a kohn - sham approach compared to a thomas - fermi - type approach are increased .
the details of the kohn - sham construction depend on the actual auxiliary quantum system one wants to employ .
the only restriction of the auxiliary system is that one can control the current and the potential by some external variables .
thus , one could even add further ( unphysical ) external fields to make approximations of the coupled quantum system easier .
however , here we only present the simplest and most natural kohn - sham scheme , which is to describe the coupled quantum system by an uncoupled quantum system . to this end
, we assume that we can find a factorized initial state @xmath85 that obeys the same initial conditions as the coupled problem ( [ initialcurrent ] ) and ( [ initialpotential ] ) . especially , if the initial state of the coupled system is the same as in the uncoupled problem , then this condition is trivially fulfilled . in a next step
we note , that for the uncoupled system subject to the external pair @xmath86 the equations of motion become ( since @xmath87 ) @xmath88;t ) & = - \frac{4 t^2_{\mathrm{kin } } } { \hbar^2 c^2 } j([a_{\mathrm{eff}},j_{\mathrm{eff}}];t ) \\ & + n ( [ a_{\mathrm{eff}},j_{\mathrm{eff } } ] ; t ) a_{\mathrm{eff}}(t ) , \nonumber \\ \label{discertemaxwellun } \partial_0 ^ 2 a([a_{\mathrm{eff}},j_{\mathrm{eff}}];t ) & = - k^2 a([a_{\mathrm{eff}},j_{\mathrm{eff}}];t ) + \frac{\mu_0 c}{l^3 } j_{\mathrm{eff}}(t),\end{aligned}\ ] ] now , obviously if one would choose @xmath89 , i.e. the external pair of the coupled problem , the uncoupled quantum system will in general lead to a different internal pair
. however , can we find an effective pair that _ reproduces _ the internal pair @xmath42 of the coupled system ?
the existence of such an effective pair can be based on equations for the uncoupled system similar to eqs .
( [ discerteforcefun ] ) and ( [ discertemaxwellfun ] ) .
note that before we were considering the question of uniqueness , i.e. can one have two external pairs leading to the same @xmath42 .
thus any internal pair @xmath42 was apriori associated with an external pair @xmath36 .
if on the other hand we are given some internal pair @xmath42 , say from a different ( coupled ) quantum system , we do not apriori know that this internal pair can be represented by propagation of an initial state with some @xmath86 .
thus this problem is equivalent to the existence of a solution to @xmath90;t ) a_{\mathrm{eff}}(t ) , \\
\label{discertemaxwellfunun } \partial_0 ^ 2 a(t ) & = -k^2 a(t ) + \frac{\mu_0 c}{l^3 } j_{\mathrm{eff}}(t).\end{aligned}\ ] ] for a given pair @xmath42 and @xmath91 . as before ,
@xmath92 is uniquely determined by simply rearranging eq .
( [ discertemaxwellfunun ] ) as @xmath93 while the existence of an @xmath94 that reproduces @xmath42 is less clear .
again , the most general approach to answer this question can rely on a fixed - point scheme similar to @xcite , or on mapping the problem to a special nonlinear schrdinger equation @xcite .
importantly , in the discretized case certain subtleties arise @xcite that have to be treated with care @xcite . disregarding these more subtle points
, we follow a simpler approach based on the assumption of taylor - expandability in time of @xmath64
. then one can successively construct the taylor coefficients of the effective potential from @xmath95 assuming that for the initial state @xmath91 the expectation value @xmath96 . further assuming that this taylor - series converges @xcite , we have constructed a pair @xmath97,j_{\mathrm{eff}}[a]),\end{aligned}\ ] ] that reproduces @xmath42 via propagation of @xmath91 .
the above defined pair @xmath98,j_{\mathrm{eff}}[a])$ ] actually describes the mapping @xmath99 now , in order to actually _ predict _ the physical pair @xmath42 via the kohn - sham system ( and thus solve eqs .
( [ discerteforceintfun ] ) and ( [ discertemaxwellintfun ] ) ) we have to connect the coupled and the auxiliary system . to do
so we make the composite mapping
@xmath100 i.e. we employ the fact that @xmath42 are unique functionals of the initial state @xmath1 and @xmath101 . the definition of the resulting kohn - sham potentials and currents
are then found by equalizing the functional eqs .
( [ discerteforceintfun ] ) and ( [ discertemaxwellintfun ] ) with the according equations of the uncoupled auxiliary system .
this leads to ( now also indicating the appropriate dependence on the initial states ) @xcite @xmath102;t ) a_{\mathrm{ks}}(t ) = &
\lambda \langle \hat{n } \a \rangle ( [ \psi_0,j , a];t ) \\ & + n ( [ \psi_0,j , a];t ) a_{{{\rm ext}}}(t ) \nonumber \\ \label{kscurrent } j_{\mathrm{ks}}(t ) = & j_{{{\rm ext}}}(t ) + j(t).\end{aligned}\ ] ] therefore , they are functionals of the two initial states , @xmath42 and @xmath101 , i.e. @xmath103 , j_{\mathrm{ks}}[j , j_{{{\rm ext } } } ] \right).\end{aligned}\ ] ] with these definitions the coupled problem , starting from @xmath1 and subject to the external pair @xmath101 , can be formally solved by the solution of an uncoupled , yet non - linear problem with initial state @xmath91 and the kohn - sham pair @xmath104 .
the resulting equations are @xmath105{\left| \ , \mathrm{m}(t ) \ , \right\rangle } , \label{modelkscurrent } \\ & \left ( \partial_0 ^ 2 + k^2\right ) a(t ) = \frac{\mu_0 c}{l^3}\left ( \lambda j(t ) + j(t)\right ) .
\label{modelksfield}\end{aligned}\ ] ] the self - consistent solutions of the kohn - sham eqs .
( [ modelkscurrent ] ) and ( [ modelksfield ] ) by construction obey eqs .
( [ kspotential ] ) and ( [ kscurrent ] ) , as well as equations of motion similar to eqs .
( [ discerteforcefunun ] ) and ( [ discertemaxwellfunun ] ) . by combining these equations
we see that the solutions to the kohn - sham equations generate the solutions to the coupled eqs .
( [ discerteforceintfun ] ) and ( [ discertemaxwellintfun ] ) .
we point out , that in the equation for the photonic mode we do not need any approximate functional . we merely need to solve a classical maxwell equation .
however , in practice it might be useful , especially when calculating non - trivial photonic expectation values , that one solves an actual ( uncoupled ) photon problem to have a first approximation to the photonic wave function . ) in the weak - coupling limit : ( a ) inversion @xmath106 , ( b ) density @xmath107 and ( c ) exact kohn - sham potential @xmath108 in the case of regular rabi oscillations.,scaledwidth=50.0% ] in this section , we show numerical examples for our model system .
we use the density - functional framework introduced in the previous sections and we explicitly construct the corresponding exact kohn - sham potentials . to illustrate our qedft approach , we focus mainly on two different examples : the first example treats a setup in resonance , where regular rabi oscillations occur .
we show results in a weak - coupling limit and in a strong - coupling limit .
the second example includes the photon field initially in a coherent state . for this case , we study collapses and revivals of the rabi oscillations . the hamiltonian in eq .
( [ hubbard ] ) is directly connected to the famous jaynes - cummings - hamiltonian and the rabi hamiltonian @xcite , which is heavily investigated in quantum optics .
it has been studied in the context of rabi oscillations , field fluctuations , oscillation collapses , revivals , coherences and entanglement ( see ref.@xcite and references therein ) . to directly see
the connection between the two - site hubbard model coupled to one photon mode and the rabi hamiltonian , we transform the hamiltonian in eq .
( [ hubbard ] ) by dividing with @xmath109 , where @xmath110 is an arbitary ( dimensionless ) scaling factor .
thus we make the hamiltonian and the according schrdinger equation dimensionless .
the hamiltonian of eq .
( [ hubbard ] ) can then be rewritten in a similar form as usually found in the literature @xmath111 where we transformed to the dimensionless external potential @xmath112 and the dimensionless external current @xmath113 .
further , we also transform to a dimensionless time variable @xmath114 . to actually perform numerical calculations
, we have to choose values for the free parameters . here , we choose typically used values from the literature : @xmath115 , @xmath116 , @xmath117 and external fields which are set to zero @xmath118 .
this set of parameters allow for a resonance situation , with no detuning between the transition energy of the atomic levels and the frequency of the field mode . as discussed above the basic variables ( densities )
are the current operator @xmath119 and the operator for the field potential @xmath120 . in this two - level example @xmath119
reduces to @xmath121 and @xmath120 reduces to @xmath122 . if the rotating - wave approximation is applied to the rabi hamiltonian in eq .
( [ rabi - hamiltonian ] ) , one recovers the jaynes - cummings hamiltonian .
this hamiltonian is then analytically solvable .
the rotating - wave approximation is only valid in the weak - coupling limit ( @xmath123 ) . in the strong - coupling limit ( @xmath124 )
, however , the rotating - wave approximation breaks down . only recently ,
analytic results without the rotating - wave approxmation have been published @xcite . here
we emphasize that the qedft approach presented in this paper is exact and does not rely on the rotating - wave approximation and hence also allows to treat strong - coupling situations . in our first example
we choose as initial state for both , the coupled many - body system and the uncoupled kohn - sham problem @xmath125 meaning the electron initially populates site one and the field is in the vacuum state .
therefore , no photon is present in the field initially . in fig .
[ fig : rabi_exact ] , we show the inversion @xmath106 , the density @xmath126 and the corresponding exact kohn - sham potential @xmath108 for the weak - coupling case .
the atomic inversion @xmath106 shows regular rabi oscillations .
rabi oscillations are also visible in @xmath127 , where we observe the typical neck - like features @xcite at @xmath128 and at later points in time . to determine the exact kohn - sham potential for this case , we follow a fixed - point construction similar to @xcite . as input for the fixed - point construction , we use the exact many - body densities .
in addition , we also compare to an analytic formula for the kohn - sham potential for a one - electron two - site hubbard model given in @xcite .
this expression gives an explicit formula for the dependence of the kohn - sham potential on the density .
such an explicit formula is only known in a few cases , while the fixed - point construction is generally valid . however , both methods yield in the present case the same results . a detailed discussion of the fixed - point construction for multicomponent systems of electrons and photons will be presented in a forthcoming work @xcite . we emphasize that a propagation of the uncoupled kohn - sham system with the exact kohn - sham potential @xmath108 obtained in fig . [
fig : rabi_exact ] reproduces by construction the exact many - body density ( @xmath126 in the present case ) .
however , as illustrated in sec .
iic , if a kohn - sham propagation is used , the numerical expenses can be drastically reduced , since the kohn - sham construction effectively decouples the quantum system . in practical calculations the exact kohn - sham potentials are normally not available and one has to rely on approximations . in the present case ,
the simplest approximation for @xmath129 $ ] is straightforward if we assume @xmath130 \approx n[\psi_0,j , a]$ ] and @xmath131 .
then , from eq .
( [ kspotential ] ) we find the mean - field approximation to the kohn - sham potential @xmath132;t ) = \lambda a(t ) + a_{{{\rm ext}}}(t ) .
\label{eq : meanfieldpotential}\end{aligned}\ ] ] the mean - field approximation is actually identical to the maxwell - schdinger approach , i.e. we treat the electromagnetic field as being essentially classical .
further , for @xmath133 and for @xmath134 the mean - field approximation becomes asymptotically exact . in fig .
[ fig : rabi_weak ] and fig .
[ fig : rabi_strong ] , we compare exact densities and exact kohn - sham potentials to densities and potentials , which were obtained by a self - consistend mean - field propagation . already in the weak - coupling limit ,
[ fig : rabi_weak ] , quite sizable differences between exact results and mean - field results become visible : already at @xmath68 the exact kohn - sham potential deviates from the mean - field potential . in the case of the densities , this leads to a frequency shift , where the mean - field density oscillates slower than the exact density . in the strong - coupling limit shown in fig .
[ fig : rabi_strong ] , effects beyond the rotating - wave approximation are visible . in the exact kohn - sham potential , we see a non - regular feature at @xmath135 , which is also not coverd by the mean - field approximation
. however , the mean - field approximation already covers at least some dynamical features of the propagation .
, ( b ) density @xmath107 and ( c ) exact kohn - sham potential @xmath108 in the case of coherent states ( in spirit of panel 3 in fig . 4 in ref .
@xcite),scaledwidth=50.0% ] for the second example in this section , we start with the field initially in a coherent state . for a single field mode , coherent states @xcite can be written as follows : @xmath136 in this example , we use as initial state for the many - body propagation and the kohn - sham propagation @xmath137 here , the atomic state @xmath138 is the ground state of the electronic hamiltonian ( @xmath139 . for the field state we choose @xmath140 .
this example is in the spirit of the calculation in panel 3 in ref .
@xcite . hence , as shown in fig .
[ fig : cs_exact ] , we obtain a similar time - evolution of the inversion @xmath106 as in ref .
we see the cummings collapse of rabi oscillations at @xmath141 followed by a quiescence up to @xmath142 occuring .
after @xmath142 , we see a revival of the rabi oscillations .
we also observe , as shown in @xcite , that the atomic dipole operator ( here the density @xmath107 ) continues to change during the interval of quiescence after the inversion collapse .
as before , we show in the lowest panel the corresponding exact kohn - sham potential obtained via fixed - point iterations . in fig .
[ fig : cs_weak ] , we show a comparison of the exact kohn - sham potentials and densities to the mean - field propagation . here
, we see that the mean - field approximation performs rather poorly . for this case the simple ansatz in eq .
( [ eq : meanfieldpotential ] ) is not sufficient and more sophisticated approximations to the exact kohn - sham potential are necessary to reach a better agreement @xcite . in summary
, we have shown in this section the exact kohn - sham potentials which reproduce the dynamics of the exact many - body densities .
in particular the coherent state example shows that there is a clear need for better approximations to the exact kohn - sham potential @xcite that go beyond the mean - field level and that include correlation contributions .
one possibility along these lines is provided by an approach based on the optimized effective potential ( oep ) method @xcite .
we have already implemented such an oep approach for the present model system and the corresponding results improve quite considerably over the mean - field approximation .
the details of this general oep approach to qedft are beyond the scope of the present paper and will be presented in a separate publication @xcite .
after having presented the basic concepts of a qedft reformulation of a coupled matter - photon problem in a model system , we apply the very same ideas to the full theory of qed .
while no new density - functional - type ideas have to be introduced , the intricacies of qed make the actual details more involved .
a first subtlety is the gauge freedom of the photon field . in this work
we choose coulomb gauge to fix the superfluous degrees of freedom .
this gauge has two distinct advantages over the other gauges : it reduces the independent components of the photon field to the two transversal ( physical ) polarizations , and it singles out the classical coulomb interaction between the charged particles . since we want to connect qedft to derived theories like cavity qed , where usually coulomb - gauged photons are employed , and condensed - matter theory , where coulomb interactions play a dominant role , the coulomb gauge is for the present purpose the natural gauge to work in .
however , we emphasize that also other gauges can be used as well @xcite .
we first present the standard approach to identify possible conjugate variables and introduce the basic equations of motions .
while in the usual non - relativistic setting this route works just fine , in the fully relativistic situation the internal structure of the `` dirac particles '' , i.e. the electronic and positronic degrees of freedom , give rise to certain subtleties when performing a density - functionalization .
therefore , instead of using the current , we employ the polarization as a basic fundamental variable in relativistic qedft . in the following we define the basic quantities of qed in coulomb gauge and
derive the equations of motion for the fundamental ( functional ) variables of the theory .
we employ si units throughout , since in the next section we perform the non - relativistic limit which is most easily done if we keep the physical constants explicit . a detailed discussion of quantizing qed in coulomb gauge is given in appendix [ app : coulomb ] . the full qed hamiltonian in coulomb gauge ( indicating explicit time - dependence of the hamiltonians by @xmath6 ) is given by @xmath143 here @xmath144 is the normal ordered ( : : ) free dirac hamiltonian in the schrdinger picture , where @xmath145 and @xmath146 denote the dirac - field operators and @xmath147 the dirac matrices ( see appendix [ app : coulomb ] for definitions ) . the energy of the free photon field is given by @xmath148 where @xmath149 and @xmath150 are the ( vector - valued ) electric and magnetic field operators defined as in appendix [ app : coulomb ] in terms of the maxwell - field operators @xmath151
we note that due to the coulomb - gauge condition @xmath152 only the spatial components of the maxwell field are quantized .
the time component @xmath153 is given by the classical coulomb field of the total charge density , which is the sum of the charge density of the dirac field and the classical external current , and gives rise to the coulomb term @xmath154 here @xmath155 is the zero component of the dirac current @xmath156 and @xmath157 is the zero component of a given external current @xmath158 . in the coulomb term the energy due to
the coulomb interaction of the external current with itself is elided .
since this term is purely multiplicative , i.e. it is equivalent to the identity operator times some real number , it does not influence the dynamics of the system and can be discarded . the rest of the coupling to the external fields
is given by @xmath159 finally , the coupling between the quantized fields in coulomb gauge reads as @xmath160 comparing to the lorentz - gauge qed hamiltonian @xcite the main difference lies in the coulomb term , that treats the zero component of the photon field explicitly . without further refinements the above qed hamiltonain is not well - defined , since it gives rise to infinities @xcite .
these infinities can be attributed , with the help of perturbation theory , to three divergent types of feynman diagrams : the self - energy of the fermions , the self - energy of the photons ( also called vacuum polarization ) and the vertex corrections .
these divergences vanish if we regularize the theory , e.g. , by introducing frequency cut - offs in the plane - wave expansions of the fermionic as well as the bosonic field operators or by dimensional regularization @xcite .
such procedures make the above hamiltonian self - adjoint @xcite , but we have introduced a dependence on parameters that change the theory at smallest and largest length scales .
perturbatively one can remove these dependencies by renormalizing the theory , i.e. we first identify and then subtract the part of each of these three terms that diverges due to these parameters .
the resulting three divergent counter - terms can be recast as a renormalization of the mass and the field - operators of the fermions ( due to the self - energy ) , as a renormalization of the photonic field - operators ( due to vacuum polarization ) and a renormalization of the charge ( due to the vertex corrections ) .
we can do this to any order in perturbation theory due to the ward - takahashi identities @xcite .
thus , the above qed hamiltonian is at least perturbatively renormalizable . for simplicity , we assume in the following that one can fully renormalize the qed hamiltonian ( as has been shown for certain limits @xcite ) and interpret it as a bare hamiltonian , i.e. we use the renormalized quantities . that a full renormalization is possible has been shown , e.g. , for the nelson model of qed @xcite , where the divergent self - energy term shifts the spectrum of the hamiltonian to infinity .
thus , subtracting this infinite shift , i.e. introducing a counter - term , makes the hamiltonian well - defined ( when removing the cut - offs ) , provided the energy of the system is below the pair - creation limit .
the same condition , i.e. a stable vacuum , we need to impose also on our qed considerations as discussed in @xcite . in a next step
we identify the _ possible _ conjugate ( functional ) variables of the above qed hamiltonian . here
the physical , time - dependent wave function @xmath161 depends on the initial state and the external pair @xmath162 , which is indicated by @xmath163;t ) \ , \right\rangle}.\end{aligned}\ ] ] thus , with @xmath164 , the ( negative ) qed action @xcite @xmath165 \!=\ ! -\!\ ! \int\!\ ! \mathcal{l}_{\mathrm{qed } } = \!-
\mathcal{b } \!+\ !
\frac{1}{c } \ ! \int \ !
\left ( j^{\mu}_{\mathrm{ext } } a_{\mu } + j_{\mu } a^{\mu}_{\mathrm{ext } } \right ) \end{aligned}\ ] ] becomes a functional of these variables ( @xmath166 corresponds to an arbitrary time ) .
here we employed the definition of the qed lagrangian of eq .
( [ qedlagrangian ] ) and defined the internal qed action with help of eq .
( [ qedham ] ) by @xmath167 eq .
( [ qedaction ] ) looks like a legendre transformation between @xmath168 and @xmath169 . since a legendre
transformation amounts to a change of variables , this indicates ( for a fixed initial state ) the possibility of transforming from @xmath16 to the conjugate variables @xmath170 . and
@xmath171 @xcite ]
if these variables would indeed be connected via a standard legendre transformation the functional derivative with respect to @xmath172 and @xmath173 should give the respective conjugate variables .
however , following derivations similar to @xcite we find the appearence of extra terms , i.e. @xmath174 these non - trivial boundary terms are due to the fact , that variations of the external fields give rise to non - zero variations of the wave function at the ( arbitrary ) upper boundary @xmath166 ( in contrast to direct variations of the wave function that are supposed to obey @xmath175 ) @xcite .
these boundary terms are necessary to guarantee the causality of @xmath176 and @xmath177 @xcite .
thus , eqs .
( [ conjugatea ] ) and ( [ conjugatej ] ) show that a straightforward approach to demonstrate a one - to - one correspondence between @xmath16 and @xmath170 based on a legendre transformation becomes difficult @xcite
. nevertheless , usually this legendre - transformation arguments work well to identify the possible conjugate variables .
however , in the relativistic situation a further problem arises : the current has an internal structure due to the electronic and positronic degrees of freedom .
the current @xmath176 describes the net - charge flow of the negatively charged electrons and the positively charged positrons @xcite .
therefore , the current expectation value can not differ between the situation of , e.g. , the movement of two electrons and one positron or three electrons and two positrons .
this fact , which is absent in the non - relativistic situation , will lead to problems when employing the ideas developed in subsection [ subsec : foundmodel ] .
for the moment , however , we follow the above identification scheme and derive the basic equations of motion for @xmath9 and @xmath10 . since @xmath178 f(\vec{r } ' ) = 0 $ ] , where @xmath179 is any testfunction , the term @xmath180 commutes with @xmath9 and the equation of motion for the four current is the same as in lorentz gauge @xcite @xmath181\spin(\vec{r } ) \\ & + e c \dspin(\vec{r})\left [ \gamma^{k}\gamma^{0}\left(- \imagi \vec{\gamma } \cdot
\vec{\nabla}\right ) + \left(- \imagi \vec{\gamma } \cdot \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\nabla } \right ) \gamma^0 \gamma^{k}\right ] \spin(\vec{r } ) \nonumber \\ & + \frac{e^2}{\hbar}\dspin(\vec{r})\left[\gamma^{k } \gamma^{0 } \gamma^{l}- \gamma^{l } \gamma^{0 } \gamma^{k } \right]\spin(\vec{r})\ !
\left ( \a_{l}(\vec{r } ) + a_{l}^{\mathrm{ext}}(x ) \right ) \nonumber,\end{aligned}\ ] ] where the zero component is given by @xmath182 , i.e. the current obeys the conservation of charge . a different equation that determines the charge current @xmath176
is found by the gordon - decomposition @xcite , which is the evolution equation of the polarization @xmath183 @xmath184 where @xmath185 is the levi - cevita symbol and @xmath186 with the definition of bigger and smaller components of the dirac - field operators @xmath187 we find that the current and the polarization are the real and imaginary part of the same operator @xmath188 the change of gauge only affects the equation for the photon - field operator which becomes @xmath189 and accordingly @xmath190 this is indeed the quantized maxwell equation in coulomb gauge . in this subsection
we first reexamine the previous approach to relativistic qedft @xcite and identify its shortcomings .
we then show why physically the polarization is better suited as fundamental variable of the matter part and reformulate qed in terms of @xmath191 .
already here we point out that both , a relativistic qedft based on the current or on the polarization , lead to the same density - functional - type theory in the non - relativistic limit .
a first restriction we impose is to fix a specific gauge for the external fields @xmath172 . since by construction external fields that only differ by a gauge transformation , i.e. @xmath192 , lead to the same current density ( and polarization),$ ] which determines the effect of a gauge on the equation of @xmath9 , i.e. eq .
( [ heisenbergcurrent ] ) . by partial integration ,
application of the continuity equation and the fact that @xmath193 \equiv 0 $ ] this term becomes zero and therefore has no effect on the current .
the same reasoning shows that also @xmath194 is gauge independent . ]
the desired one - to - one correspondence can only hold modulo these transformations .
thus in principle we consider a bijective mapping between equivalence classes , and by fixing a gauge we take a unique respresentative of each class . for simplicity we impose a gauge condition similar to @xcite @xmath195 in the following , any other gauge that keeps the initial state unchanged , i.e. the gauge function has to obey @xmath196 , is also allowed @xcite .
this condition is necessary for our further investigations , since we will employ that the initial state is fixed ( and thus the expectation values at @xmath68 ) , in accordance to the derivations of subsection [ subsec : foundmodel ] .
further , we assume that the external current obeys the continuity equation @xmath197 .
this leaves the choice of the initial charge configuration @xmath198 . since the photons only couple to moving charges this choice will not influence the dynamics of the system
therefore , also in the case of the external currents we have an equivalence class ( of possible zero components ) , and will therefore restrict to only prescribing @xmath199 .
in @xcite the one - to - one correspondence was based on the corresponding ehrenfest equations @xmath200 where @xmath201 \!\!\spin(\vec{r } ) \nonumber \\ & + \imagi \ ; \frac{e}{\hbar } m c^2 \dspin(\vec{r})\left [ \gamma^{0 } \gamma^{k } - \gamma^{k } \gamma^{0 } \right]\spin(\vec{r}),\end{aligned}\ ] ] @xmath202\spin(\vec{r}),\end{aligned}\ ] ] @xmath203 and the dalembert operator reads as @xmath204 .
we can then reexpress @xmath205 and therefore @xmath206 if we then want to show a possible one - to - one correspondence we can follow the reasoning of sec .
[ subsec : foundmodel ] and
consider the uniqueness of solutions of the functional equations @xmath207;x ) + q_{\mathrm{int}}^{k}([a_{{{\rm ext}}}^{m},j_{{{\rm ext}}}^{m}];x ) \nonumber \\ & + n^{k l}([a_{{{\rm ext}}}^{m},j_{{{\rm ext}}}^{m}];x ) a_{l}^{\mathrm{ext}}(x ) , \\ \label{intpotentialfun } \box a^{k}(x ) & + \partial^{k } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{\mathrm{ext}}(x ' ) + \vec{\nabla}'\cdot\vec{j}(x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) \nonumber \\ & = \mu_0 c \left(j^{k}_{{{\rm ext}}}(x ) + j^{k}(x ) \right),\end{aligned}\ ] ] for given @xmath208 and @xmath209 .
the freedom of the internal variable @xmath208 is also restricted , since @xmath210 is fixed by the initial state and the continuity equation for all times .
similarly the freedom of the external current @xmath211 is in correspondence to the freedom of the internal field @xmath209 . ] as before in sec .
[ subsec : foundmodel ] we can construct the external current uniquely . by defining the vector field @xmath212
we find from the helmholtz decomposition of @xmath213 and @xmath214 that @xmath215 thus , we need to show that for given @xmath216 there can only be a unique @xmath217 . and
@xmath218 , respectively .
since @xmath219 is given by eq .
( [ coulombpotential ] ) its first order time - derivative is , due to the continuity equation , given in terms of @xmath199 and @xmath220 .
thus , to determine @xmath219 we only have to choose an initial condition . however , due to eq .
( [ coulombpotential ] ) this choice also fixes automatically the initial condition for @xmath221 for the continuity equation @xmath222 . ] to show this we first define @xmath223 formally construct the respective taylor coefficients @xmath224 \\ & + \sum_{\beta=0}^{\alpha } { \alpha \choose \beta } \left ( a^{l \ , ( \beta)}_{\mathrm{ext}}(\vec{r } ) \right ) \left(n_{k l}^{(\alpha-\beta)}(\vec{r } ) \right),\nonumber \end{aligned}\ ] ] and consider two external potentials @xmath225 that differ at lowest order @xmath74 .
accordingly we find in this order @xmath226 where @xmath227 while before we could conclude that the difference between the currents is necessarily non - zero provided @xmath228 , here we find that this is not sufficient .
actually , we need to restrict the allowed potentials @xmath229 to those that are perpendicular to @xmath230 . if we do this , then eq . ( [ possiblecontradiction ] ) makes the currents necessarily different and we can conclude that we have a one - to - one correspondence .
this aspect was not taken into account in previous work @xcite , which is restricting effectively the one - to - one mapping to a smaller set of potentials and currents in these proofs .
still , it seems possible to find a different way to show the bijectivity of the complete mapping @xmath231 .
however , the true drawback of a relativistic qedft based on the current is found if we try to reproduce a given pair @xmath216 .
if we choose a current that obeys @xmath232 then the resulting equation that defines the taylor - coefficient of the external potential reads by employing eq .
( [ currenttaylor ] ) and following the same strategy as in subsection [ subsec : ksmodel ] @xmath233 this equation does not have a solution and therefore any current that obeys the above form can not be reproduced by the respective quantum system .
this does also call into doubt the possibility of exactly predicting the current of a coupled system by an uncoupled one , i.e. the kohn - sham construction of @xcite .
of course we can remedy this problem by adding terms to the qed hamiltonian that break the minimal - coupling prescription of the lagrangian .
such procedures could then be alternatively used to provide a kohn - sham scheme to describe the fully coupled qed problem .
the advantage of such an approach is , that still the equation for the vector potential is known explicitly in terms of the internal pair @xmath216 .
this is not the case , when we use a different basic variable for the matter part of the qed system , as we will do in the following . to avoid the problems with the relativistic current
, we will in the following base our considerations on the polarization @xmath234 .
while the current describes the flow of the charge of the system ( which is conserved ) , the polarization depends on the actual number of particles and anti - particles ( which is not conserved ) .
therefore , the polarization can actually differ between a local current produced by , e.g. , two electrons and one positron or three electrons and two positrons , in contrast to the current . to show
now that for a fixed initial state @xmath1 we actually have @xmath235 we demonstrate that for a given internal pair @xmath236 the two coupled equations @xmath237;x ) + \vec{q}_{\mathrm{int}}([a_{{{\rm ext}}}^{k},j_{{{\rm ext}}}^{k}];x ) \\ & + \frac{2emc}{\imagi \hbar } \vec{j}([a_{{{\rm ext}}}^{k},j_{{{\rm ext}}}^{k}];x ) + \frac{2e}{\imagi \hbar c } p_{0}([a_{{{\rm ext}}}^{k},j_{{{\rm ext}}}^{k}];x ) \vec{a}_{\mathrm{ext}}(x ) \nonumber \\
\label{intpotfun } \box & \vec{a}(x ) - \mu_0 c \left(\vec{j}_{{{\rm ext}}}(x ) + \vec{j}([a_{{{\rm ext}}}^{k},j_{{{\rm ext}}}^{k}];x ) \right ) \\ & = \vec{\nabla } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{{{\rm ext}}}(x ' ) + \vec{\nabla}'\cdot\vec{j}([a_{{{\rm ext}}}^{k},j_{{{\rm ext}}}^{k}];x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) , \nonumber\end{aligned}\ ] ] allow only for a unique solution @xmath238 . here
we used the definitions @xmath239 @xmath240 these coupled equations can only have a solution if the pair @xmath236 obeys the initial condition enforced by the fixed initial state @xmath1 , i.e. @xmath241 since the current @xmath208 is now a functional of @xmath238 the previous explicit construction of @xmath211 is no longer valid .
however , if we assume @xmath238 _ both _ to be taylor - expandable we find for the lowest order @xmath74 on the one hand that @xmath242 provided @xmath243 , which corresponds to the ( local ) total number of particles and anti - particles . on the other hand we have @xmath244 otherwise there would exist a current @xmath245 that fulfills @xmath246 however , due to the definition of the coulomb potential as the green s function of the laplacian ( see eq .
( [ coulombpotential ] ) and ( [ a0 ] ) ) we find from the divergence of eq .
( [ currentuniqueness ] ) that @xmath247 and thus the only possible current that fulfills eq .
( [ currentuniqueness ] ) is @xmath248 .
thus , the mapping ( [ relpairmapping ] ) is bijective ( at least for taylor - expandable external pairs @xmath249 ) .
therefore we can , instead of solving the fully coupled qed problem for the ( numerically infeasible ) wave function @xmath161 , determine the exact internal pair @xmath236 from the coupled non - linear equations @xmath250;x ) + \vec{q}_{\mathrm{int}}([p_{k},a_{k}];x ) \\ & + \frac{2emc}{\imagi \hbar } \vec{j}([p_{k},a_{k}];x ) + \frac{2e}{\imagi \hbar c } p_{0}([p_{k},a_{k}];x ) \vec{a}_{{{\rm ext}}}(x ) \nonumber \\ \label{intpotfunbasic } \box & \vec{a}(x ) - \vec{\nabla } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{{{\rm ext}}}(x ' ) + \vec{\nabla}'\cdot\vec{j}([p_{k},a_{k}];x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) \nonumber \\ & = \mu_0 c \left(\vec{j}_{{{\rm ext}}}(x ) + \vec{j}([p_{k},a_{k}];x ) \right),\end{aligned}\ ] ] for the initial conditions ( [ initialpolarization ] ) and ( [ initialfourpotential ] ) . in order to solve these equations simultaneously
we need to find approximations for the unknown functionals .
the only drawback in this more general approach than the ones used in @xcite , is that now we also have an unknown functional in the classical maxwell equation , i.e. @xmath251 $ ] . in this subsection
we provide the adopted kohn - sham construction based on the internal pair @xmath252 and give the simplest approximation for the kohn - sham potential and current . as in sec .
[ subsec : ksmodel ] , we choose our auxiliary kohn - sham system to be an uncoupled system . while different kohn - sham constructions are possible , this approach is the numerically least demanding . in a first step , in accordance to sec .
[ subsec : ksmodel ] , we first construct an uncoupled system that can _ reproduce _ a given internal pair @xmath236 of the fully coupled qed system .
to do so , we first need an initial state @xmath91 that fulfills the initial condition ( [ initialpolarization ] ) and ( [ initialfourpotential ] ) of the full qed system .
this allows that the coupled equations @xmath253;x ) + \frac{2emc}{\imagi \hbar } \vec{j}([a^{k}_{\mathrm{eff}},j^{k}_{\mathrm{eff}}];x ) \nonumber \\ & + \frac{2e}{\imagi \hbar c } p_{0}([a^{k}_{\mathrm{eff}},j^{k}_{\mathrm{eff}}];x ) \vec{a}_{\mathrm{eff}}(x ) \\ \label{intpotfuneff } \box \vec{a}(x ) & - \vec{\nabla } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{\mathrm{eff}}(x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) \nonumber \\ & = \mu_0 c \vec{j}_{\mathrm{eff}}(x ) , \end{aligned}\ ] ] can only have a unique solution . obviously , for the case of the uncoupled problem we can use a construction similar to eqs .
( [ currentconstruction ] ) to determine the unique @xmath254 . to show the existence of a solution to eq .
( [ intpolfuneff ] ) we perform the standard taylor - expansion construction and assume that the series converges @xcite
. a more general approach would be to follow a fixed - point procedure @xcite .
the respective taylor - coefficients of the effective potential are given by @xmath255 this construction makes plausible that there exists an uncoupled system subject to the effective external fields @xmath256 that reproduces a given pair of a fully coupled qed problem .
the above construction actually resembles the mapping @xmath257 for a given pair @xmath258 .
now , to _ predict _ the internal pair @xmath258 of the full qed problem we again introduce a composite mapping @xmath259 the resulting kohn - sham potential and kohn - sham current are then given by the functional equations @xmath260;x )
\vec{a}_{\mathrm{ks}}(x ) = \frac{\imagi \hbar c } { 2e}\left ( \vec{q}_{\mathrm{kin}}([\psi_0 , p_k , a_{k}];x ) \right .
\nonumber \\ & \left .
- \vec{q}_{\mathrm{kin}}([\phi_0 , p_k , a_{k}];x ) + \vec{q}_{\mathrm{int}}([\psi_0 , p_k , a_{k}];x ) \right ) \nonumber \\ & + mc^2 \left ( \vec{j}([\psi_0 , p_k , a_{k}];x ) - \vec{j}([\phi_0 , p_k , a_{k}];x ) \right ) \nonumber \\ & + p_{0}([\psi_0 , p_k , a_{k}];x ) \vec{a}_{\mathrm{ext}}(x ) \\ \label{intpotfunks } \vec{j}_{\mathrm{ks}}&(x ) = \vec{j}_{{{\rm ext}}}(x ) + \vec{j}([\psi_0 , p_k , a_{k}];x).\end{aligned}\ ] ] this allows to solve an uncoupled system instead of the fully coupled qed problem .
however , as also pointed out in @xcite , we can only fully decouple the matter from the photon part if also the initial state is of product form , i.e. @xmath261 . and
if we further assume that @xmath262 is given in terms of a slater - determinant we can actually map the whole problem to solving a dirac equation with the above kohn - sham potential @xmath263 and simultaneously a classical maxwell equation with @xmath264 .
the mean - field approximation recovers the approximation introduced in @xcite and reads as @xmath265 since for simplicity we used a gauge where @xmath266 while for the photon field we employed coulomb gauge , we have to perform an according gauge transformation to have the mean field @xmath267 in either the one or the other gauge completely .
this approximation is similar to the maxwell - schrdinger approach , that assumes the photon field to behave essentially classically .
while for the sake of generality we have been considering the full qed problem in the previous section , we are actually mainly interested in the behaviour of condensed - matter systems or atoms and molecules that interact with photons . in such situations
the external fields are usually small compared to the schwinger - limit , i.e. we do not have pair - production is such situations .
further , we want to investigate systems , where the quantum nature of the photons becomes important .
most prominently this happens for the case of a cavity , where different boundary conditions for the maxwell field have to be considered .
these quantum - optical situations also naturally restrict the available photonic modes .
such physical situations are then well described by models of non - relativistic particles interacting with a quantized electromagnetic field , such as the pauli - fierz hamiltonian ( see e.g. @xcite ) or the nelson model @xcite . in the lowest order of approximations
we find the situation of a two - level system interacting with one photonic mode , similar to the one presented in sec .
[ sec : model ] .
this simplest of models is the prime example of a quantum - optical problem .
we realize at this point , that all the conditions we had to impose in order to make our starting qed hamiltonian well - defined , are naturally met in the situations we aim at investigating .
actually , we even do not need to adopt a field - theoretical treatment for the particles in the first place and usually only need to take into account a few photonic modes .
such an approach would avoid a lot of unpleasent problems in connection with renormalization and regularization of these theories .
however , one would then need to introduce a new qedft approach for every new type of model hamiltonian .
therefore , in this section we want to demonstrate how naturally all lower lying qedft reformulations are just approximations to the fully relativistic qedft that we presented in the previous sections . in lowest order we then recover the two - site hubbard model coupled to one mode of sec . [
sec : model ] . in this subsection
we derive the non - relativistic limit of the basic equations of motion , on which the qedft reformulations are based .
we show how approximations in the hamiltonian correspond to approximations in the basic equations of the corresponding qedft approaches .
let us first start with the non - relativistic limit of the fully coupled qed hamiltonian in coulomb gauge . from the heisenberg equation of motion , defining @xmath268 and @xmath269 , we find the quantized dirac equation ( in the heisenberg picture ) @xmath270 \spin(x ) \nonumber \\ & \qquad + e^2 \int \diff^3 r ' \frac{:\spin^{\dagger}(x ' ) \spin(x'):}{4 \pi \epsilon_0 |\vec{r}-\vec{r}'| } \spin(x ) , \end{aligned}\ ] ] and accordingly for @xmath271 .
we see that the electronic components @xmath272 of the four - spinor are mixed with the positronic components @xmath273 .
of course , for small energies only the electronic component of the four - spinor is important , and therefore we would like to find an equation based solely on @xmath272 .
so naturally we would like to decouple the upper component @xmath272 from the lower component @xmath273 .
a possible way would be to find a unitary transformation of the dirac hamiltonian that does this , at least perturbatively . a possible expansion parameter for auch a perturbative transformation would be @xmath274 , since we know that the energies involved in non - relativistic processes are small compared to the rest - mass energy .
this energy also represents the spectral gap between the electronic and positronic degrees of freedom , which effectively decouples the dynamics of the particles and anti - particles for small enough energies .
the resulting unitary transformations are known as the foldy - wouthuysen transformations @xcite and are routinely used to generate the non - relativistic limits of the dirac equation to any order desired . here
, we employ an equivalent but different procedure to decouple the electronic from the positronic degrees of freedom . to do so , we first rewrite eq .
( [ quantizeddirac ] ) componentwise @xmath275 where we defined @xmath276 and thus we ( formally ) find that @xmath277^{-1 } \vec{\sigma } \cdot \left(\ !
-\imagi \hbar c \vec{\nabla}\ ! -\ !
e \hat{\vec{a}}_{\mathrm{tot}}(x)\!\right ) \p(x).\end{aligned}\ ] ] if we assume non - relativistic energies , the main contribution to the energy of the system stems from @xmath278 , i.e. @xmath279 .
accordingly , from the neumann series of the resulting operator we find the inverse operator to lowest order as @xmath280^{-1 } \approx 1/2mc^2 $ ] , and consequently @xmath281 at this level of approximation to the full qed problem we find the pauli - fierz hamiltonian ( already transformed back to the schrdinger picture ) , @xmath282 where the non - relativistic kinetic energy reads as @xmath283 the energy of the electromagnetic field is given as before , the coulomb energy is given by @xmath284 and the non - relativistic current is defined by @xmath285 here we used the definition of the paramagnetic current @xmath286,\end{aligned}\ ] ] the magnetization density @xmath287 and the zero component of the current @xmath288 by construction the current obeys the continuity equation @xmath289 .
we note here , that due to the non - relativistic limit the physical current defined in eq .
( [ nonrelcurrent ] ) becomes explicitly time - dependent @xcite .
further , we point out that the result of the above ( formal ) derivations is the same as the result obtained by first performing the non - relativistic limit of the classical hamiltonian @xmath290 ( constructed from the classical lagranian density of eq .
( [ qedlagrangian ] ) ) and then canonically quantizing the schrdinger field , as shown in fig .
[ fig : qed_mapping ] ( a ) . in the non - relativistic limit the resulting hamiltonian commutes with the particle - number operator @xmath291 , as can be seen directly from the continuity equation .
accordingly we do not need to employ a field - theoretical description for the electrons and all matter - operators can be expressed in first - quantized notation ( while still being a many - particle problem ) .
nevertheless , we can still encounter infinities due to the interaction between the non - relativistic particles and the quantized maxwell field @xcite .
while we do no longer have vacuum polarization ( no electron - positron pairs are possible ) and vertex corrections , we still have an infinite self - energy @xcite .
to first order in the coupling the ground - state energy ( for @xmath292 ) diverges as @xmath293 where @xmath294 is the ultra - violet cut - off for the photon modes . by subtracting the infinite self - energy of the ground - state , which is equivalent to introducing a renormalized mass , we can renormalize the hamiltonian perturbatively . in the following we assume ,
that the pauli - fierz hamiltonian can be fully renormalized . for instance , in the limit of only scalar photons ( the nelson model ) we know that we can perform a full renormalization of the hamiltonian by subtracting the self - energy ( provided that the kinetic energy of the problem is smaller than @xmath278 ) @xcite .
therefore , we interpret the electron mass in the hamiltonian as a bare mass , i.e. we subtract the infinite self - energy . now , the equation of motion for @xmath295 can be either found by the heisenberg equation with the pauli - fierz hamiltonian or by the non - relativistic limit of eq .
( [ heisenbergcurrent ] ) ( see appendix ( [ app : nonreleom ] ) ) . we have explicitly checked both ways of performing the non - relativistic limit as schematically indicated in fig .
[ fig : qed_mapping ] ( b ) . after some calculation
we find ( omitting the spatial and temporal dependences ) @xmath296 \j_{0}\right\ } \nonumber \\ & - \epsilon_{klj } \partial^{l } \left\ { -\frac{e \hbar^3}{4 m^2}\p^{\dagger } \left ( \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\partial}^{n } \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\partial}_{n } \sigma^{j } - \sigma^{j}\partial^{n } \partial_n \right)\p + \frac{i e}{mc^2 } \partial_n \a^n_{\mathrm{tot } } \hat{m}^{j } - \frac{i e}{2 mc^2 } \left[\left(\partial^{j } \a_n^{\mathrm{tot } } \right ) - \left(\partial_n{j } \a^j_{\mathrm{tot } } \right ) \right ] \hat{m}^{n } \right\ } \nonumber \\ & - \frac{1}{mc^2 } \left\{\left(\imagi \partial_0 \a_{k}^{\mathrm{tot } } \right)\j_{0 } + \a_k^{\mathrm{tot}}\left(\frac{\imagi e}{mc^2 } \partial_l \a^{l}_{\mathrm{tot } } \j_{0 } - \imagi \partial^{l } \j^{\mathrm{p}}_l \right ) \right\},\end{aligned}\ ] ] where @xmath297\end{aligned}\ ] ] is the usual momentum - stress tensor and @xmath298 is the interaction - stress force ( the divergence of the interaction - stress tensor ) @xcite .
if we would have started with an uncoupled problem , we would find a similar equation with the replacement @xmath299 and @xmath300 .
further , the equation for the electromagnetic field does not change , except that we now have to employ the non - relativistic current ( see appendix [ app : nonreleom ] ) . in a next step
we perform the non - relativistic limit for the equation of motion of the polarization , i.e. eq . ( [ gordondecomposition ] ) .
we find to order @xmath301 @xmath302 thus , at this level of approximation the polarization does not change in time .
in this subsection we derive the basic formulation of non - relativistic qedft for the full pauli - fierz hamiltonian .
we show how the gordon - decomposition , i.e. the equation of motion for the polarization @xmath303 , makes the current @xmath304 a unique functional of @xmath305 and thus becomes the basic variable for the matter part in this limit .
further we demonstrate how the non - relativistic limit of the above kohn - sham construction produces the kohn - sham construction for the pauli - fierz hamiltonian .
a comparison of this level of approximation with relativistic qedft and with other approximations is presented schematically in appendix [ app : overview ] .
we start by performing the non - relativistic limit of eq .
( [ polrelqedft ] ) .
irrespective of the difference between @xmath229 and @xmath306 ( note , that we again employ the @xmath307 gauge for the external potentials as explained in sec .
[ subsec : foundrelqedft ] ) the equation in this limit is always zero . however , by employing eq .
( [ nonrelgordon ] ) we can rearrange the non - relativistic limit to @xmath308 which is non - zero provided the density obeys @xmath309 .
the form of eq .
( [ potrelqedft ] ) does not change and thus we have in the non - relativistic limit that @xmath310 accordingly we can label all physical wave functions by the non - relativistic internal pair @xmath311 .
since @xmath208 does no longer have an internal structure ( no positronic degrees of freedom ) , our approach of sec .
[ subsec : eomqed ] to determine the conjugate pairs based on a legendre - transformation argument now works just fine .
the pauli - fierz lagrangian ( determined from the pauli - fierz hamiltonian of eq .
( [ nonrelham ] ) ) has a similar structure as the full qed lagrangian of eq .
( [ qedaction ] ) and thus allows to identify the conjugate pairs .
indeed , the legendre - transformation argument holds for all further non - relativistic approximations , especially for our model system of sec .
[ sec : model ] .
now we can , instead of solving for the wave function , solve the coupled equations @xmath312;x ) + \vec{q}_{\mathrm{m}}([j_k , a_k , a^{k}_{{{\rm ext}}}];x ) \nonumber \\ & + \vec{q}_{0}([j_k , a_k , a^{k}_{{{\rm ext}}}];x ) , \\
\label{nonrelpotfunbasic } \box \vec{a}(x ) & - \vec{\nabla } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{{{\rm ext}}}(x ' ) + \vec{\nabla}'\cdot\vec{j}(x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) \nonumber \\ & = \mu_0 c \left(\vec{j}_{{{\rm ext}}}(x ) + \vec{j}(x ) \right),\end{aligned}\ ] ] for a fixed initial state and external pair @xmath305 , where @xmath313 is the first term on the right hand side of eq .
( [ nonreleom ] ) , @xmath314 corresponds to the second term and @xmath315 corresponds to the third . the initial state and the fixed external pair @xmath305 determine the initial conditions for the above coupled equations .
the explicit appearance of the external potential in several terms in the equation of motion and in the initial condition is due to the non - relativistic limit .
the main advantage of this limit is , that we do no longer need an explicit approximation for functionals in the maxwell equation , since we now consider the current directly . in a next step
we can then perform the non - relativistic limit of the kohn - sham scheme of eqs .
( [ intpolfunks ] ) and ( [ intpotfunks ] ) which leads to @xmath316;x ) a_{\mathrm{ks}}^{k}(x ) \\ & = j_{0}([\psi_0,j_n , a_{n}];x ) a_{{{\rm ext}}}^{k}(x ) + \langle \a^{k } \j_0 \rangle ( [ \psi_0,j_n , a_{n}];x ) \nonumber \\ & + \frac{mc}{e}\left ( j_{\mathrm{p}}^{k}([\phi_0,j_n , a_{n}];x ) - j_{\mathrm{p}}^{k}([\psi_0,j_n , a_{n}];x ) \right ) \nonumber \\ & + \frac{mc}{e}\epsilon^{klj}\partial_l\left ( m_{j}([\psi_0,j_n , a_{n}];x ) -m_{j}([\phi_0,j_n , a_{n}];x ) \right ) \nonumber \\ & j_{\mathrm{ks}}^{k}(x ) = j^{k}_{{{\rm ext}}}(x ) + j^{k}(x ) .
\nonumber\end{aligned}\ ] ] if we then further assume that the different initial states fulfill @xmath317 ( due to the continuity equations the zero components stay equivalent ) we can define the so - called hartree - exchange - correlation ( hxc ) potential by @xmath318 = \vec{a}_{{{\rm ext } } } + \vec{a}_{\mathrm{hxc}}[\psi_0,\phi_0 , j_k , a_{k } ] , \end{aligned}\ ] ] and we end up with @xmath319;x ) \\ & + \frac{mc}{e}\left ( j_{\mathrm{p}}^{k}([\phi_0,j_n , a_{n}];x ) - j_{\mathrm{p}}^{k}([\psi_0,j_n , a_{n}];x ) \right ) \nonumber \\ & + \frac{mc}{e}\epsilon^{klj}\partial_l\left ( m_{j}([\psi_0,j_n , a_{n}];x ) -m_{j}([\phi_0,j_n , a_{n}];x ) \right ) .
\nonumber\end{aligned}\ ] ] thus , assuming that we have given an appropriate initial state of the form @xmath320 that has the same initial current , initial potential and electric field ( corresponding to the first time - derivative of the potential ) we can solve simultaneously @xmath321{\left| \ , \mathrm{m}(t ) \ , \right\rangle } , \nonumber \\
\label{nonrelksii } & \box a^{k}(x ) + \partial^{k } \left(\frac{1}{c } \int \diff^3 r ' \frac{\vec{\nabla}'\cdot\vec{j}_{{{\rm ext}}}(x ' ) + \vec{\nabla}'\cdot\vec{j}(x')}{4 \pi \epsilon_0 |\vec{r } - \vec{r}'| } \right ) \nonumber \\ & \qquad= \mu_0 c \left(j^{k}_{{{\rm ext}}}(x ) + j^{k}(x ) \right).\end{aligned}\ ] ] if we further assume that the initial state @xmath262 is given as a slater determinant of orbitals @xmath322 , we can solve single - orbital kohn - sham equations .
the simplest approximate hxc potential is just the non - relativistic limit of the mean - field approximation of eq .
( [ relmfapprox ] ) , i.e. @xmath323 note again , that without a further gauge transformation we now also have a scalar potential in the kohn - sham hamiltonian due to @xmath218 .
we point out , that we could alternatively use eq .
( [ nonreleom ] ) directly to show the one - to - one correspondence between the external pair @xmath305 and the non - relativistic internal pair @xmath311 @xcite .
however , besides being more involved , also the connection to relativistic qedft becomes less clear .
nevertheless , for the construction of approximations to the kohn - sham potential , eq .
( [ nonreleom ] ) seems better suited , since it is a more explicit equation . in this subsection
we show how , by introducing further approximations , we can find a family of non - relativistic qedfts , which in the lowest - order approximation leads to the model qedft of sec .
[ sec : model ] . as pointed out before , in the non - relativistic situation the initial guess for the conjugate pairs , i.e. by identifying a legendre - type transformation in the lagrangian of the problem , holds true .
thus we can now derive all sorts of approximate qedfts by investigating different conserved currents and restrictions to the photonic degrees of freedom . in the currents this holds since approximating
the conserved current @xmath304 implies approximating the hamiltonian in eq .
( [ nonrelham ] ) accordingly .
thus , e.g. , by assuming a negligible magnetic density @xmath324 , i.e. @xmath325 the according hamiltonian as well as the defining eqs .
( [ nonreleom ] ) and ( [ nonrelgordon ] ) change .
actually , all terms @xmath326 and @xmath327 vanish in these equations for this approximation .
we again find due to the corresponding eq .
( [ nonrelcontra ] ) that we have @xmath328 and we can consider the corresponding coupled eqs .
( [ nonrelcurrentfunbasic ] ) and ( [ nonrelpotfunbasic ] ) .
the kohn - sham current becomes accordingly @xmath329 and the hxc potential in this limit reduces to @xmath330;x ) \\ & + \frac{mc}{e}\left ( j_{\mathrm{p}}^{k}([\phi_0,j_k , a_{k}];x ) - j_{\mathrm{p}}^{k}([\psi_0,j_k , a_{k}];x ) \right ) .
\nonumber\end{aligned}\ ] ] on the other hand we can also restrict the allowed photonic modes . for instance we can assume a perfect cubic cavity ( zero - boundary conditions ) of length @xmath33 . then , with the allowed wave vectors @xmath331 and the corresponding dimensionless creation and annihilation operators @xmath332 and @xmath333 ( see appendix ( [ app : hubbard ] ) for more details ) we find @xmath334 \mathcal{s}(\vec{n}\cdot{\vec{r } } ) , \end{aligned}\ ] ] where the mode function @xmath335 is given in eq .
( [ modefunction ] ) . if we further restrict the modes by introducing a square - summable regularization function @xmath336 , e.g. @xmath337 for @xmath338 ( energy smaller than rest - mass energy ) and @xmath339 otherwise , the resulting regularized field @xmath340 \mathcal{s}(\vec{n}\cdot{\vec{r}})\end{aligned}\ ] ] makes the coupled pauli - fierz hamiltonian self - adjoint without any further renormalization procedure @xcite .
such a restriction is assumed in the following .
these approximations are then directly reflected in the hamiltonian and the derived equations of motion . while the basic eq .
( [ nonrelgordon ] ) does not change , and thus @xmath304 is the basic matter - variable , the basic equation of motion for the potential @xmath341 has to reflect the restriction to specific modes . by multiplying eq .
( [ nonrelpotfunbasic ] ) from the left by @xmath342 and integrating we find the mode expansions @xmath343 where @xmath344 and we use the definition @xmath345 the coulomb part vanishes since we employ a partial integration and the fact that @xmath346 .
of course , one finds the same equations by a straightforward calculation of the heisenberg equation of motion for the maxwell - field ( [ cavitypotential ] ) with the according pauli - fierz hamiltonian ( [ nonrelham ] ) . from the restriction to specific modes the field @xmath341 is restricted in its spatial form and therefore the photonic variable changes from @xmath341 to the set of mode expectation values @xmath347 this change in basic variable is also reflected in the conjugate external variable which is given via eq .
( [ modeexpansion ] ) as @xmath348 thus we accordingly find @xmath349 and the conjugate pairs become @xmath350 thus we have to solve the mode eqs .
( [ modeexpansion ] ) together with the according equation of motion for the current .
correspondingly also the kohn - sham scheme and the mean - field approximation for @xmath351 change to its mode - equivalents .
if we then also employ the dipole - approximation , i.e. we assume that the extension of our matter system is small compared to the wavelengths of the allowed photonic modes , we find @xmath352.\end{aligned}\ ] ] this only changes the definition of effective currents that couple to the modes , i.e. @xmath353 but leaves the structure of the qedft reformulation otherwise unchanged .
if we assume the magnetization density @xmath354 to be negligible , we have from first principles rederived the qedft formulation presented in @xcite . in this work
the situation of only scalar external potentials , i.e. @xmath355 and @xmath356 , has been considered as a second case . in this situation ,
the gauge freedom is only up to a spatial constant , which is usually fixed by choosing @xmath357 for @xmath358 .
since @xmath359 couples to the zero component of the current , i.e. the density @xmath360 , the conjugate pair becomes @xmath361 to demonstrate this mapping , the first time derivative of @xmath360 is obviously not enough , since this amounts to the continuity equation and no direct connection between the two conjugate variables of the matter part of the quantum system is found .
therefore , one has to go to the second time derivative of @xmath360 @xcite . if we then further simplify this physical situation ( see appendix ( [ app : hubbard ] ) for a detailed derivation ) we find the model hamiltonian of sec .
[ sec : model ] . in a similar manner , by imposing the restrictions on the corresponding equations of motion , we can rederive the model qedft of sec .
[ subsec : foundmodel ] . ) for the external potentials . by applying the dipole approximation also to the external vector potential
the conjugate variable becomes the density ( dipole moment ) .
for clarity of presentation , though , we have chosen to start from the scalar- potential case . ] finally , for a simple overview , we have collected the different qedfts that we have explicitly considered in this work in appendix [ app : overview ] .
in this work we have shown how one can extend the ideas of tddft to quantized coupled matter - photon systems .
we have first explained the basic ideas of qedft for a model system of a two - site hubbard model coupled to a single photonic mode . by rewriting the problem in terms of an effective theory for a pair of internal functional variables and proving the uniqueness of solutions for the resulting non - linear coupled equations ,
we have demonstrated how an explicit solution for the coupled photon - matter wave function can be avoided .
further we have discussed how an auxiliary quantum system , the so - called kohn - sham system , can be used to construct approximations for the implicit functionals appearing in the effective equations .
the kohn - sham construction gives rise to effective fields and effective currents , which are termed kohn - sham potential and kohn - sham current , respectively . by numerically constructing the exact kohn - sham potential and kohn - sham currents ,
we have illustrated the capability of this new approach to exactly describe the dynamics of coupled matter - photon systems and contrasted these exact fields with the mean - field approximation . in the following , instead of reformulating every possible approximate treatment of coupled matter - photon systems
seperately , we have shown how these qedfts for approximate hamiltonians are merely approximations to relativistic qedft , which itself is based on qed . to avoid problems with the kohn - sham construction ,
we have based relativistic qedft on the expectation value of the polarization and the vector potential of the quantum system . by then performing the non - relativistic limit of qedft
we have demonstrated that the resulting theory is the qedft reformulation of the pauli - fierz hamiltonian .
the non - relativistic limit automatically makes the ( non - relativistic ) current the basic variable for the matter system .
accordingly , the non - relativistic limit of the kohn - sham potentials and currents leads to the corresponding kohn - sham fields for the pauli - fierz hamiltonian . by performing further approximations for non - relativistic qedft , e.g. assuming the magnetic density negligible , we have shown how other qedfts ( that reformulate the corresponding approximate hamiltonians ) can be derived .
depending on the level of approximation , the basic internal functional variables change , e.g. if we confine the electromagntic field with a cavity , the ( allowed ) mode expectation values become the new internal variable of the photons . in a final step we restricted to a two - site model coupled to only one mode , recovering the model qedft of the beginning .
we point out , that at every level of qedft we recover the corresponding ( standard ) time - dependent density - functional reformulations @xcite if we assume the quantized nature of the photons negligible , i.e. the charged particles interact via the classical coulomb interaction only . this will be the case in most standard situation of condensed matter theory , e.g. when investigating dynamics of atoms or molecules in free space .
however , we expect that interesting effects happen when the boundary conditions for the maxwell field are changed , e.g. for atoms in a cavity .
thus we have a potential tool that can treat complex electronic systems in the setting of quantum optics .
also , we can investigate the explicit interplay of photons with molecules or nanostructures , e.g. in nano - plasmonics .
however , for this theory to be practical we are in need of reliable approximations to the basic functionals . in @xcite functionals based on an optimized effective potential approach @xcite
are constructed , which provide good results even in the situation of strongly - coupled systems .
although the currently available approximations have only been tested for simple model systems , the hierarchy of qedft approximations allows to simply scale up these functionals to more complex situations .
thus we can develop approximations for simple systems , e.g. only one mode couples to the matter system , and then extend these approximations to more involved problems , e.g. considering more modes . in this way we can easily control the validity of our approximations . in this respect , we are also working on a fixed - point approach in the spirit of @xcite , which allows us to construct the exact kohn - sham potentials and compare the approximate potentials to the ( numerically ) exact expressions .
details of this approach will be part of a forthcoming publication @xcite .
on the other hand , the fixed - point approach is also a way to extend the validity of qedft beyond taylor - expandable fields .
a different way , especially for discretized matter systems , is the non - linear schrdinger equation approach introduced in @xcite .
certain theoretical and mathematical details of the model qedft of sec .
[ sec : model ] , that are beyond the scope of the current manuscript , will be discussed in @xcite .
finally , since we are aiming at investigating quantum optical settings , we also need to discuss the cavity and the problem of open quantum systems . in this work we focused on closed systems and on a perfect ( cubic ) cavity .
it is straightforward ( but tedious ) to extend the current work to an arbitrary shape of the perfect cavity .
we have to use an expansion of the photon field in the according eigenfunctions of the cavity , such that these modes obey the coulomb - gauge condition .
however , in actual quantum optical experiments , the cavities are not perfect but rather an open quantum system , which allows for an exchange with the environment . to take care of this channel of decoherence there are several possible ways .
one can employ the current formulation of qedft and derive a master equation , as has also been done for standard tddft @xcite . also extensions to stochastic equations @xcite are possible . on the other hand
, one can couple further bosonic degrees of freedom to the system and prescribe a bath spectral density , making these degrees of freedom a bath for the system @xcite .
since the present framework allows for a consistent treatment of interacting fermionic and bosonic particles , the inclusion of a bath and coupling to other fields , e.g. phonons , will be the subject of future work .
mr acknowledges useful discussions with c. genes and h. ritsch and financial support by the austrian science fonds ( fwf projects j 3016-n16 and p 25739-n27 ) .
we further acknowledge financial support from the european research council advanced grant dynamo ( erc-2010-adg-267374 ) , spanish grant ( fis2010 - 21282-c02 - 01 ) , grupos consolidados upv / ehu del gobierno vasco ( it578 - 13 ) , ikerbasque and the european commission projects cronos ( grant number 280879 - 2 cronos cp - fp7 ) .
in this work we employ the standard covariant notation @xmath362 with greek letters indicating four vectors , e.g. @xmath12 , and roman letters indicating spatial vectors , i.e. @xmath363 . to lower ( or raise ) the indices , i.e. going from contravariant vectors to covariant vectors ( or vice versa ) , we adopt the convention @xmath364 for the minkwoski metric .
we denote spatial ( contravariant ) vectors with the vector - symbol , i.e. @xmath365 , and the derivatives with respect to the space - time vectors @xmath366 by @xmath367 . with these definitions
the divergence can be written as @xmath368 , where we also adopt the einstein summation convention .
further we note that @xmath369 . with the help of the levi - civita symbol @xmath370 we can write the curl as @xmath371 and the multiplication of pauli matrices becomes @xmath372 ) = -g^{kl}-\imagi \epsilon^{klm } \sigma_{m}$ ] .
further , for notational simplicity we only point out in the text ( when necessary ) , whether we are in the schrdinger or heisenberg picture , and do not explicitly indicate the picture used in the operators . in the schrdinger picture , operators which are not explicitly time - dependent only carry a purely spatial dependence , e.g. @xmath373 .
we indicate explicit time - dependence in the schrdinger picture by either carring the full space - time dependence , e.g. @xmath374 ( for the pauli - fierz current density ) of eq .
( [ nonrelcurrent ] ) , or by a dependence on @xmath6 , e.g. @xmath375 in eq .
( [ nonrelham ] ) . in the heisenberg picture
, every operator also depends on time , e.g. @xmath376 in eq .
( [ quantizeddirac ] ) .
in this appendix we give a detailed derivation of qed in coulomb gauge .
we start from the ( classical ) coupled qed lagrangian with external fields @xmath377 and @xmath378 given by @xcite @xmath379 here we use the standard definitions for the ( classical ) dirac fields , i.e. @xmath380 where @xmath381 is a dirac four - spinor with the two - component ( spin ) functions @xmath382 and @xmath383 , the gamma matrices are given by @xmath384 with @xmath385 the usual pauli matrices , @xmath386 and @xmath387 is the conserved ( noether ) current .
further we use the minkowski metric @xmath388 to raise and lower the indices .
for the ( classical ) maxwell field we have @xmath389 where @xmath390 is the electric field tensor and @xmath391 is the vector potential .
now we employ the coulomb gauge condition for the maxwell field , i.e. @xmath392 . then it holds that @xmath393 where @xmath394 is the laplacian . if we impose square - integrability on all of @xmath395 the green s function of the laplacian becomes @xmath396 and therefore @xmath397 since the zero component of the four potential @xmath391 is given in terms of the full current , it is not subject to quantization .
the conjugate momenta of the photon field ( that need to be quantized ) are the same as in the current - free theory and thus the usual canonical quantization - procedure applies @xcite , i.e. @xmath398 = -\imagi \hbar c \delta_{kl}^{\perp}(\vec{r } -\vec{r}'),\end{aligned}\ ] ] where @xmath399 is the electric field operator , @xmath400 is the transverse delta - function and @xmath401 are spatial coordinates only .
equivalently we can define these operators by their respective plane - wave expansions @xmath402 , \\ & \hat{\vec{e}}(\vec{r } ) \!\!=\!\ ! \sqrt{\frac{\hbar}{\epsilon_0}}\!\ ! \int \!\ ! \frac{\diff^3
, \imagi \omega_{k}}{\sqrt{2 \omega_k ( 2 \pi)^3 } } \!\!\sum_{\lambda=1}^2 \!\ ! \vec{\epsilon}(\vec{k } , \lambda ) \!\ !
} e^{\imagi \vec{k}\cdot \vec{r } } \!-\ !
e^{- \imagi \vec{k}\cdot \vec{r } } \right],\end{aligned}\ ] ] where @xmath403 , @xmath404 is the transverse - polarization vector @xcite , and the annihilation and creation operators obey @xmath405 = \delta^3 ( \vec{k}-\vec{k } ' ) \delta_{\lambda \lambda'}.\end{aligned}\ ] ] if we further define the magnetic field operator by @xmath406 , the hamiltonian corresponding to @xmath407 is given in eq .
( [ maxwellham ] ) .
we used normal ordering to get rid of the infinite zero - point energy in this expression . also for the dirac field
, the coupling does not change the conjugate momenta .
therefore we can perform the usual canonical quantization procedure for fermions which leads to the ( equal - time ) anti - commutation relations @xcite @xmath408 the hamiltonian corresponding to @xmath409 thus becomes the one of eq .
( [ diracham ] ) , where we used @xmath410 .
it is straightforward to give the missing terms of the qed hamiltonian due to the coupling to the external fields as well as due to the coupling between the quantized fields .
if we apply the definition of the quantized current @xmath411 of eq .
( [ 4current ] ) to eq .
( [ a0 ] ) we find ( using normal ordering , i.e. rearranging the annihilation parts of the operators to the right , to discard the corresponding zero - point energy ) @xmath412 here , if we disregard the purely multiplicative first term on the right - hand side , we arrive at eq .
( [ coulombham ] ) .
the rest is given in eqs .
( [ externalham ] ) and ( [ interactionham ] ) . alternatively , with the definition of @xmath413 in eq .
( [ a0 ] ) and @xmath414 the full qed hamiltonian can also be written as @xmath415
to find the non - relativistic limit of eq .
( [ heisenbergcurrent ] ) , we can not straightaway apply the decoupling to eq .
( [ decoupling ] ) . since we have to apply the decoupling consistently to the hamiltonian as well as the current we need to rewrite the equation of motion .
we start ( in the heisenberg picture ) by @xmath416 = \frac{2 e mc ^2}{\hbar } \left [ \c^{\dagger } \sigma^k \p - \p^{\dagger } \sigma^k \c \right ] \\ & - \imagi e c \left [ \p^{\dagger } \left ( \sigma^{k}\sigma^{l}\partial_l + \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\partial}_{l } \sigma^{l } \sigma^{k } \right)\p + \c^{\dagger } \left ( \sigma^{k}\sigma^{l}\partial_l + \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\partial}_{l } \sigma^{l } \sigma^{k } \right)\c \right ] \nonumber \\ & - \frac{2 \imagi e^2}{\hbar } \epsilon^{klj } \a_{l}^{\mathrm{tot } } \left[\p^{\dagger } \sigma_j \p + \c^{\dagger}\sigma_{j } \c \right ] .
\nonumber\end{aligned}\ ] ] this leads with @xmath417 and @xmath418 to @xmath419 \right . \\ & \left . - \imagi e \hbar c \c^{\dagger } \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\partial}_{l } \sigma^{l } \sigma^{k } \c - \imagi e \hbar c \p^{\dagger } \sigma^{k } \sigma^{l } \partial_{l } \p \right\}. \nonumber\end{aligned}\ ] ] adding and subtracting on the right hand side the term @xmath420 and employing eq .
( [ chicomponent ] ) we find @xmath421 \sigma^{k } \c \nonumber \\ & \left .
+ c \p^{\dagger } \sigma^{k } \imagi \hbar c
\partial_0 \c \right\}. \end{aligned}\ ] ] with the help of the definition @xmath422 = [ \hat{d } + mc^2]^{-1}$ ] this can be rewritten as @xmath423^{\dagger } \left(\imagi \hbar c \overset{\smash{\raisebox{-1.5pt}{\tiny$\leftarrow$}}}{\nabla } -e \hat{\vec{a}}^{\mathrm{tot}}\right)\cdot \vec{\sigma } - \p^{\dagger } e a_0^{\mathrm{tot } } - \p^{\dagger } e^2 \ ! \!\ ! \int \ ! \ !
\diff^3 r ' \frac{:\!\p^{\dagger}(x')\p(x ' ) \ ! + \ ! \c^{\dagger}(x')\c(x')\!:}{4 \pi \epsilon_0 |\vec{r}-\vec{r}'| } - mc^2 \p^{\dagger}\right ] \right .
\nonumber \\ & \sigma^{k } [ ... ] \vec{\sigma } \cdot \left(- \imagi \hbar c \vec{\nabla } - e \hat{\vec{a}}^{\mathrm{tot } } \right ) \p + \p^{\dagger } \sigma^{k}\left[\imagi \hbar c \partial_0 [ ... ] \vec{\sigma } \cdot \left(- \imagi \hbar c \vec{\nabla } - e \hat{\vec{a}}^{\mathrm{tot } } \right ) \right ] \p + \p^{\dagger } \sigma^{k } [ ... ] \vec{\sigma } \cdot \left(- \imagi \hbar c \vec{\nabla } - e \hat{\vec{a}}^{\mathrm{tot } } \right ) \nonumber \\ & \left . \left
[ \vec{\sigma } \cdot \left(- \imagi \hbar c \vec{\nabla } - e \hat{\vec{a}}^{\mathrm{tot } } \right ) [ ... ] \vec{\sigma } \cdot \left(- \imagi \hbar c \vec{\nabla } - e \hat{\vec{a}}^{\mathrm{tot } } \right ) + e a_0^{\mathrm{tot } } - e^2 \ ! \!\
! \int \ ! \ !
\diff^3 r ' \frac{:\!\p^{\dagger}(x')\p(x ' ) \ ! + \ ! \c^{\dagger}(x')\c(x')\!:}{4
\pi \epsilon_0 |\vec{r}-\vec{r}'| } - mc^2 \right ] \p \right\ } \nonumber\end{aligned}\ ] ] now , if we employ the approximation @xmath422\approx 1/2 mc^2 $ ] ( also in the coulomb terms ) we end up with @xmath424 which is just the equation of motion for the non - relativistic current ( [ nonrelcurrent ] ) with the pauli - fierz hamiltonian . for the maxwell field the non - relativistic limit of eq .
( [ intpotential ] ) is with the help of eq .
( [ nonrelcurrent ] ) straightforward .
it is only important to see , that this does agree with the equation of motion for @xmath425 due to the pauli - fierz hamiltonian ( [ nonrelham ] ) .
the main difference to the fully relativistic derivation is , that now we have a term of the form @xmath426 this term does not change anything in the first order equation , i.e. @xmath427 . in the second order we find due to eq .
( [ perpdelta ] ) that @xmath428 \j_{0}(x ' ) \\ & = 2 \frac{\imagi \hbar c}{\epsilon_0 } \a^{l}(x ) \j_{0}(x ) - 2 \frac{\imagi \hbar c}{\epsilon_0 } \partial^k \delta^{-1}\partial^l \a_l(x ) \j_{0}(x ) \nonumber\end{aligned}\ ] ] and @xmath429 a_{l}^{{{\rm ext}}}(x')\j_{0}(x ' ) \\ & = 2 \frac{\imagi \hbar c}{\epsilon_0 } a^{l}_{{{\rm ext}}}(x ) \j_{0}(x ) - 2 \frac{\imagi \hbar c}{\epsilon_0 } \partial^k \delta^{-1 } \partial^l a_l^{{{\rm ext}}}(x ) \j_{0}(x ) \nonumber\end{aligned}\ ] ] now , with the above definition for @xmath430 used in eq .
( [ a0 ] ) we find that these commutators lead to the terms @xmath431 of the equation of motion for the maxwell field in the non - relativistic limit .
the rest of the derivation is similar to the relativistic situation .
if we restrict the allowed space for the photonic modes we also need to impose according boundary conditions .
let us first start with a cubic cavity of length @xmath33 with periodic boundary condition .
we then find with the allowed wave vectors @xmath432 and the corresponding dimensionless creation and annihilation operators @xmath332 and @xmath333 , which are connected to their continuous counterparts by @xmath433 that @xmath434.\end{aligned}\ ] ] here @xmath435 .
if we change the conditions at the boundaries to zero - boundary conditions , then the allowed wave - vectors change to @xmath436 and the discrete operators obey @xmath437 with the normalized mode functions @xmath438 the field operator therefore reads as @xmath439 \mathcal{s}(\vec{n}\cdot \vec{r}).\end{aligned}\ ] ] here @xmath440 .
we start with the non - relativistic hamiltonian of eq .
( [ nonrelham ] ) where we assume that the magnetization density @xmath441 is negligible .
we further assume a perfect cubic cavity of length @xmath33 and employ the dipole approximation , i.e. @xmath442 . thus we find a maxwell field defined by eq .
( [ dipolepotential ] ) . at this level of approximation
our starting point coincides with that adopted in @xcite . in a next step
we allow only scalar external potentials . in the following
we present a detailed derivation of the length - gauge hamiltonian employed in @xcite for the formulation of the electron - photon tddft . for simplicity
we restrict our derivations to the case of one mode and one particle .
the case of several modes and particles works analogously and leads to the hamiltonian ( 13 ) of ref .
@xcite . with the definition of the dimensionless photon
coordinate @xmath443 and the conjugate momentum @xmath444 , the single - mode vector potential is given by @xmath445 where we use the definition @xmath446 and assume @xmath447 .
the resulting hamiltonian in first quantized notation reads @xmath448 since at this level of approximation @xmath449 due to the expansion in coulomb - gauged eigenmodes . in eq .
( [ startingham ] ) we introduced the notation @xmath450 in a next step we transform the hamiltonian into its length - gauge form @xcite by the unitary transformation @xmath451.\end{aligned}\ ] ] if we then perform a canonical variable transformation of the photon - coordinate @xmath452 and @xmath453 ( leaving the commutation relations unchanged ) we find @xmath454 then we perform yet another time - dependent gauge transformation @xmath455\end{aligned}\ ] ] where @xmath456 is the projection of the external current on the direction of the photon polarization .
the above transformation is aimed at eliminating the linear in @xmath457-derivative term in eq .
( [ first_stepham ] ) . using the general transformation rule
@xmath458 we obtain @xmath459 here we see , that the photonic variable @xmath457 is coupled to the dipole moment @xmath460 , which indicates that @xmath457 is actually proportional to the electric field . in a last step
we then discretize the matter - part of the problem and employ a two - site approximation such that @xmath461 where @xmath21 is the kinetic ( hopping ) energy , @xmath462 is the vector connecting two sites , and @xmath463 corresponds to the potential difference between the sites . to highlight the general structure of the photon - matter hamiltonian ( and bring it to the form used in sec .
[ sec : model ] ) we also redefine the external current , the external potential , and the photon field as follows @xmath464 after implementing the above redefinitions in eq .
( [ 2nd_stepham ] ) and neglecting irrelevant constant terms we arrive at the following hamiltonian @xmath465 with the choice of an appropriate dimensionless coupling constant @xmath30 eq .
( [ final2site - ham ] ) reduces to the simple model hamiltonian of eq .
( [ hubbard ] ) .
we note , that the same model hamiltonian could have been derived by assuming an external vector potential in a gauge such that @xmath266 and @xmath466 . then by the dipole approximation
the corresponding hamiltonian to eq .
( [ startingham ] ) we would have terms of the form @xmath467 , @xmath468 and mixed terms of internal and external vector potential . by going into length gauge also for the external potential and performing the same steps as above , one ends up with the same two - site one - mode hamiltonian .
here we give an overview of the different qedfts that we have discussed explicitly .
we employ for the kohn - sham scheme an uncoupled auxiliary quantum system with an initial state @xmath261 . for the different levels of approximation the prerequisites for this initial state change ,
i.e. we might have different initial conditions that have to be fulfilled .
further we use the notational convention that the super - index s refers to the ( uncoupled ) kohn - sham quantity , e.g. @xmath469 = p_{0}^{\mathrm{s}}$ ] .
we point out , that due to the change of the physical current @xmath220 through out the hierachy of qedfts also the inhomogeneity in the according maxwell equations change .
this inhomogeneity describes how the photons are coupled to the charged quantum particles , which effectively also leads to a coupling between the photons .
this can be most easily seen in the non - relativistic limit , where the inhomogeneity contains terms like @xmath470 . since the current of the auxiliary kohn - sham system is by construction equal to the exact current ( at least for the non - relativistic limit ) , this coupling between the photons is also present in the kohn - sham maxwell equation .
the term @xmath471 of the kohn - sham current contains these non - trivial couplings as functionals of the initial states and internal pair . when restricting the photons to a cavity ,
the kohn - sham current is then responsible to couple the different photon modes .
the coupling terms in the kohn - sham current are specifically relevant in the context of , e.g. nano - plasmonics , where the electromagnetic fields are enhanced due to the presence of the plasmons , or in the optical control of currents in solids @xcite . | in this work we give a comprehensive derivation of an exact and numerically feasible method to perform ab - initio calculations of quantum particles interacting with a quantized electromagnetic field .
we present a hierachy of density - functional - type theories that describe the interaction of charged particles with photons and introduce the appropriate kohn - sham schemes .
we show how the evolution of a system described by quantum electrodynamics in coulomb gauge is uniquely determined by its initial state and two reduced quantities .
these two fundamental observables , the polarization of the dirac field and the vector potential of the photon field , can be calculated by solving two coupled , non - linear evolution equations without the need to explicitly determine the ( numerically infeasible ) many - body wave function of the coupled quantum system . to find reliable approximations to the implicit functionals we present the according kohn - sham construction . in the non - relativistic limit this density - functional - type theory of quantum electrodynamics
reduces to the density - functional reformulation of the pauli - fierz hamiltonian , which is based on the current density of the electrons and the vector potential of the photon field . by making further approximations , e.g. restricting the allowed modes of the photon field , we derive further density - functional - type theories of coupled matter - photon systems for the corresponding approximate hamiltonians . in the limit of only two sites and one mode we deduce the according effective theory for the two - site hubbard model coupled to one photonic mode
this model system is used to illustrate the basic ideas of a density - functional reformulation in great detail and we present the exact kohn - sham potentials for our coupled matter - photon model system . |
massive stars play a key role in the evolution of the universe .
they are the principal source of heavy elements and uv radiation . through a combination of winds , massive outflows , expanding hii regions , and supernova explosions they provide an important source of mixing and turbulence in the interstellar medium ( ism ) of galaxies .
turbulence in combination with differential rotation drives galactic dynamos .
galactic magnetic fields are generated , interacting with supernova shock fronts that accelerate cosmic rays .
cosmic rays , uv radiation , and dissipation of turbulence are the principal sources of heating in the ism , whereas heavy elements found in dust , molecules , and in atomic / ionic form ultimately are responsible for its cooling .
massive stars thus profoundly affect the star- and planet - formation process ( bally , moeckel & throop 2005 ) as well as the physical , chemical , and morphological structure of galaxies ( e.g. , kennicutt 1998 , 2005 ) . in spite of the dominant role that massive stars play in shaping galactic structure and evolution , our understanding of their formation and early evolution is still sketchy .
there are many reasons .
high dust extinction makes it difficult to observe high - mass stars during critical early formation phases .
they are rare .
they evolve quickly and important evolutionary phases are short - lived .
the theoretical problem is extremely complex .
finally , massive stars are seldomly ( if at all ) formed in isolation ; the proximity of other high - mass stars compounds the complex influence of the forming star on its local environment via gravitational interactions , powerful outflows and winds , ionizing radiation , and supernovae . the low number statistics of young or forming high - mass stars is only partially offset by their higher luminosities , which allow us to study them at greater distances than their low - mass counterparts .
however , insufficient spatial resolution is an issue an entire ob - star cluster is often contained in a single observing pixel ( e.g. , henning & stecklum 2002 ) .
star formation typically starts with a collapsing gas condensation ( core ) inside a larger subunit ( clump ) of a molecular cloud ( cf .
williams , blitz & mckee 2000 ) .
a protostar forms that increases its mass by accretion ( accumulation ) of neighboring gas , while at the same time some mass loss occurs through a bipolar outflow and/or a collimated jet .
let us define some of the terminology adopted here .
one of the most misused terms in papers dealing with star formation is protostar , which is considered the holy grail ( wynn - williams 1982 ) of ir astronomy . here , we reserve the term protostar or protostellar object for a gaseous object in hydrostatic equilibrium ( gas pressure forces balance the gas self - gravity ) , which has not yet begun hydrogen burning but which will , given time , burn hydrogen . at
the point hydrogen burning commences , we shall speak of a zero - age main - sequence ( zams ) star ; as long as hydrogen burning occurs in the center , we shall speak of a main - sequence star .
note that the size scale of a protostar is at most a few tens of solar radii .
we use the terms massive star and high - mass star interchangeably to denote an ob star sufficiently massive to produce a type ii supernova ( @xmath08 for solar abundances ) . with these definitions in mind ,
the term high - mass protostar denotes a @xmath18m@xmath2 hydrostatic object that has not yet begun hydrogen burning .
as we shall see in the following , such objects exist only briefly during a transitory stage between `` accreting intermediate - mass protostar '' and `` accreting high - mass star . '' because it will be impossible to distinguish observationally when an accreting object begins burning hydrogen , we suggest that the terms massive protostar and high - mass protostar generally be avoided . in * table 1 * , we give a crude classification of massive stars in terms of logarithmic mass intervals and the corresponding main sequence spectral types .
.main sequence massive star definition ( logarithmic mass ranges ) [ cols="^,^,^,^,^,^,^",options="header " , ] @xmath3emb stands for embedded , i.e. , a very young phase ( @xmath41myr ) ; sources optically invisible + @xmath5obsc stands for obscured , i.e. , a young phase ( @xmath61myr ) ; sources optically visible but heavily dust extincted + note : afgl 2591 may be at a distance of 1700pc ( schneider et al .
2006 ) and part of the cyg ob2 association , one of the richest regions of massive stars in the galaxy with @xmath6100 o stars , including a hidden very massive ob cluster , shown in * figure 18*. note that this massive star forming region is a source of significant diffuse @xmath7 ray emission , i.e. the 1.8mev line from the radioactive decay of @xmath8al ( plschke et al .
* accretion + gas accumulation of a star or protostar , increasing the mass of the object * competitive accretion + gas initially unbound to a star or protostar and moving with relative speed , v@xmath9 , w.r.t . to the object
is added to it , i.e. , to its gravitational sphere or column of influence ( defined by an impact parameter 2gm / v@xmath10 ) * coalescence + growth of the mass of a stellar or protostellar object by a physical collision and merger with another object * merger + inelastic collision between two stars or protostars leading to the amalgamation of the two bodies * massive star + star more massive than about 8m@xmath11 that ends its life with a type ii supernova ( unless it is in a close binary system with mass transfer ) ; a massive star on the main sequence has spectral type b3 or earlier . *
ob star + star of spectral type o ( @xmath1216m@xmath11 ) or type b ( b@xmath124m@xmath11 ) * ultracompact hii region + small blob ( typical density @xmath13@xmath14 , size 0.1pc ) of ionized gas emitting radio continuum radiation of high emission measure ( ln@xmath15=10@xmath16pc@xmath17 ) * hypercompact hii region + very small blob ( density 10@xmath18@xmath14 , typical size 0.01pc ) of ionized gas emitting radio continuum radiation of very high emission measure ( ln@xmath15=10@xmath19pc@xmath17 ) * protostar + object on its way to become a star , with more than half of its final mass still to be accumulated * infrared dark cloud ( irdc ) + dense interstellar cloud ( often filamentary ) seen in absorption against background thermal ir emission ; detected with _ iso _ , _ msx _ , and _ spitzer _
* core + basic molecular cloud unit ( small dense gas fragment , mass 10100m@xmath2 , size @xmath60.1pc ) to form one or a few stars * clump + basic molecular cloud unit ( large dense gas fragment , mass @xmath610005000m@xmath2 , size @xmath60.5pc ) to form a young cluster of stars * hot molecular core ( hmc ) + dense warm ( t@xmath12100k ) molecular gas in emission ( co , nh@xmath20 , ch@xmath20oh ) heated from a protostar inside ; a compact region ( size @xmath40.1pc ) in a star forming molecular clump such as the bn / kl region in orion * orion nebula cluster ( onc ) + elongated 1@xmath212pc cluster of young stars , centered on the high - mass trapezium stars in orion ; about 4000 cluster members with a total stellar mass of @xmath61800m@xmath11 * trapezium cluster + the 45 stars in the center of the orion nebula , @xmath22 c , a , b , d , e in order of decreasing luminosity , the main exciting / ionising source being @xmath22 c ( o5.5v , ca .
35m@xmath11 ) * 30 doradus ( 30 dor ) + prime example of a giant ( 200pc ) extragalactic hii region , powered by @xmath6100 massive o stars stars in the ngc 2070 cluster ( cluster mass some 10@xmath23m@xmath11 , age @xmath43myr , cluster radius about 1arcmin or 15pc ) * r136 + nominally the 1arcsec core of the 30 dor nebula and hii region ( radcliffe object no .
136 ) , but r136 also often designates the central dense core of 1pc ( 4arcsec ) radius of the exciting star cluster ngc 2070 * jeans mass + critical mass that must be exceeded for a gas cloud to collapse dynamically , with the self - gravity of the gas cloud being opposed only by thermal pressure ; if squeezed by an external pressure , the critical mass for the onset of collapse is called the bonnor - ebert mass .
other anisotropic forces such as rotation or magnetic fields can oppose the cloud s self - gravity and thus impede star formation .
* free - fall timescale + timescale it takes for a pressure - free cloud to collapse dynamically to a very small size ( nominally a point ) under its own self - gravity ; the timescale depends on the average cloud gas density only ( for number density n=10@xmath24@xmath14 , t@xmath25=10@xmath24yr ) .
* kelvin - helmholtz ( kh ) timescale + time it takes for a young stellar object or stellar core to radiate its thermal energy content , typically 10@xmath23yr for young high - mass stars but 10@xmath16yr for solar - type stars
( t@xmath26=gm@xmath27/rl ) .
nuclear energy generation eventually offsets the thermal energy loss .
* premain - sequence evolution + while high - mass stars are born on the main - sequence ( i.e. they reach hydrogen burning while still accreting matter onto their stellar surface during their deeply embedded phase ) , the collapse of a low - mass star does not immediately lead to a hydrogen - burning star , rather to a quasi - static object that keeps contracting in radius by factors of a few before hydrogen burning ignites in the center . during this contraction ,
the young stellar object has a time - dependent radius , effective temperature , and luminosity that can be calculated by pre - ms evolutionary models .
* zams + zero - age main sequence : a star that has contracted enough to start hydrogen burning in its central region * initial mass function ( imf ) + mass distribution of ( single ) stars at birth introduced by salpeter in 1955 * _ spitzer _ + nasa mid - infrared space telescope of 85 cm diameter launched in 2003 * _ msx _ + mid - infrared satellite explorer with an aperture of 20 cm diameter * _ chandra _
+ 0.510kev nasa x - ray telescope in space with particularly good imaging quality comparable to optical seeing ( angular resolution @xmath61arcsec ) * _ hst _ + _ hubble space telescope _ , 2.4 m aperture , workhorse for high angular resolution space astronomy in uv , optical , and near - ir ; in orbit since 1990 the present review benefited greatly from insightful discussions with john bally , henrik beuther , adriaan blaauw , ian bonnell , bernhard brandl , wolfgang brandner , anthony brown , ed churchwell , paul crowther , melvyn davies , lise deharveng , bruce elmegreen , neal evans , simon glover , wolf - rainer hamann , thomas henning , george herbig , david hollenbach , jim jackson , eric keto , spyros kitsionas , ralf klessen , pavel kroupa , mark krumholz , richard larson , hendrik linz , andre maeder , eric mamajek , jorge melnick , karl menten , vincent minier , tony moffat , dieter nrnberger , tolya piskunov , simon portegies zwart , thomas preibisch , gregor rauw , hugues sana , fernando selman , joe silk , nathan smith , bringfried stecklum , jonathan tan , neal turner , virpi niemela ( deceased ) , nolan walborn , and gerd weigelt .
we thank geoffrey burbidge and allan sandage for their vote of confidence , and ewine van dishoeck and particularly john kormendy for their critical reading of the manuscript .
portions of this research was conducted at the jet propulsion laboratory , california institute of technology , which is supported by the national aeronautics and space administration ( nasa ) .
the assistance of u. hanschur at aip potsdam was indispensible in completing this review and is gratefully acknowledged .
would like to thank andrea lagarini and the magalhaes family in rio de janeiro and eso in santiago de chile for their hospitality during the very final stages of writing this review .
this collaboration has a long history , dating back to 19901995 when the authors were colocated at the university of wrzburg .
klein ri , fisher r , mckee cf . 2004 . in _ gravitational collapse : from massive stars to planets , _ ed .
g garcia - segura , g tenorio - tagle , j franco , hw yorke .
mex . astron .
22:37 zinnecker h. 1996 .
in _ the interplay between massive star formation , the ism and galaxy evolution .
11th iap astrophys . meet .
d kunth , b guiderdoni , m heydari - malayeri , tx thuan , pp . 24958 .
gif - sur - yvette : editions frontieres .
30arcsec@xmath2130arcsec or [email protected] ) .
dozens of massive o stars are found crowded within the half - light radius of 2pc ( brandl et al .
( _ a _ ) a vlt image of ngc 3603 ( brandl et al .
1999 ) and ( _ b _ ) a vlt image of the trapezium cluster in orion ( mccaughrean 2001 ) are shown , as these two galactic clusters would be seen if they were located at the distance of r136 in the large magellanic cloud ( 50kpc ) and imaged with similar angular resolution ( see zinnecker 2002).,width=453 ] m@xmath2 and diameter 100pc , somewhat centrally condensed ( factor of 20 , with a gaussian radial profile ) .
the cloud s turbulent kinetic energy was equal to its gravitational energy .
note the filamentary structure and the associated dense cores ( blue dots ) .
these cores technically are represented by sink particles ( @xmath62400 at this stage of evolution ) . with typical masses of 10 to 100m@xmath2 and typical sizes of
0.1pc they could be the initial fragments for high - mass star formation .
a total of 2.5 million sph particles was used . some similarity
to cosmological simulations of structure formation is noted.,width=453 ] /m@xmath2@xmath2812 with mass loss , by pietrinferni et al .
( 2004 ) for stars 0.5@xmath29m@xmath30/m@xmath2@xmath2910 , and for completeness from unpublished tracks by yorke for masses m@xmath30=0.1 , 0.2 , and 0.3m@xmath2 , using the computer code provided in the book by bodenheimer et al .
zams radii are well represented by two power laws with a break at m@xmath30/m@xmath2=1.5 .
the power law slope of the luminosity - mass relation for massive stars varies from about 3.7 to 1.6 in the range 8@xmath4m@xmath30/m@xmath2@xmath4120.,width=453 ] t=2.0 , 2.5 , and 3.0 and for @xmath31=-18 , -16 , -14 , and -12 . at this evolutionary time t
= 65,000 yr after formation of the protostellar core , 7.0m@xmath2 of material have accreted onto the protostar , 2.8m@xmath2 are in the disk , and 0.2m@xmath2 are in the infalling envelope.,width=453 ] pre - main - sequence star .
non - accreting tracks up to h burning were calculated by yorke for m@xmath30/m@xmath2=0.1 , 0.3 , 1 , 3 , 10 , and 30 , using the computer code supplied with the book by bodenheimer et al .
[ adapted from yorke ( 2002)],width=453 ] interface by dust , molecular hydrogen , and co. the dust is destroyed at r@xmath32 .
interior to r@xmath33 , the radius where the sound speed exceeds the escape velocity , even the ionized gas is gravitationally bound .
sizes are not to scale.,width=453 ] cloud showing the formation of sheets , filaments , and cores , the latter of which become gravitationally unstable .
newly formed stars are shown in dark blue , the gas is shown in blue - green .
stars tend to cluster , they continue to accrete material in competition with other stars .
times shown are 1.0 , 1.4 , 1.8 , and 2.4 initial free - fall times ( t@xmath34 ) , from left to right and top to bottom .
[ adapted from bonnell , bate & vine ( 2003)].,width=453 ] premain - sequence stars that results in the formation of a binary ( r@xmath35=25.8 solar radii ) .
( _ b _ ) a detached encounter between a 3m@xmath2 premain - sequence star and a 10m@xmath2 zero age main - sequence star with the same minimum periastron distance as in the ( a ) encounter .
the stars have radii of 12.9 and 3.92 solar radii , respectively
. the greater density of the 10m@xmath2 star results in tidal disruption of the low - density 3m@xmath2 star to form a disk around the massive star .
the encounters have zero relative velocity at infinity ( i.e. , they are parabolic encounters ) .
[ adapted from zinnecker & bate ( 2002 ) , davies et al .
( 2006 ) ] .
the cross section for a subsequent collision with ( say ) another 10m@xmath2 star is significantly increased , and a runaway collisional growth in mass is possible .
this is the so - called shred and add process.,width=453 ] 2arcmin , seeing 0.35arcsec).the broad - band ks filter includes the ro - vibrational molecular hydrogen v=10 s(1 ) emission line at 2.12@xmath36 m , which is indicative of shock - excited gas .
however , much of the nebulosity is likely seen in reflection because it is also detected in j and h. note the multiple poorly - collimated flows , loops , and cavities emanating to the west from this massive ( ca .
20m@xmath2 ) star ( see also the speckle image in preibisch et al .
2003 from the 6 m sao telescope ) .
nothing of this kind is seen to the east of the source , except for a few faint bow - shocks , as the ( redshifted ) counterflow is deeply embedded and probably partly obscured by a dense circumstellar disk ( trinidad et al .
2003 ) around this very young star .
the fact that the flow and counterflow both reach the same distance of @xmath640arcsec ( 0.2pc ) from the central source indicates that this is intrinsically a bipolar outflow , confirmed by high - velocity co radio observations ( lada et al .
1984 ; see also the h@xmath37 and optical spectroscopic observations of the associated herbig - haro objects by tamura & yamashita 1992 and poetzel , mundt & ray 1992 , respectively ) .
it seems that the bipolar outflow from this late o - type massive star has both a well - collimated and a wide - angle component!,width=453 ] 4myr old galactic starburst cluster westerlund 1 ( wd 1 ) in cyg ob2 obtained with the eso ntt telescope ( courtesy w. brandner ) .
this obscured object is held to be the most massive young cluster in the galaxy , with ca .
100 o - type stars and a mass of at least 20,000m@xmath2 ( see kndelseder 2000 , clark et al .
2005 , mengel & tacconi - garman 2007 , brandner et al .
a number of massive eclipsing binary systems have been identified ( bonanos 2007 ) as well as a very massive runaway star , an of - type supergiant , whose origin can be traced back to the cluster ( comern & pasquali 2007 ) .
like the orion nebula cluster , or the ngc 3603 and 30 dor central clusters , wd 1 is strongly mass segregated with the high mass stars being more centrally concentrated than the low mass stars .
the field of view is 4@xmath214 , corresponding to [email protected] at the distance of wd 1.,width=453 ] | although fundamental for astrophysics , the processes that produce massive stars are not well understood .
large distances , high extinction , and short timescales of critical evolutionary phases make observations of these processes challenging . lacking good observational guidance , theoretical models have remained controversial .
this review offers a basic description of the collapse of a massive molecular core and a critical discussion of the three competing concepts of massive star formation : * monolithic collapse in isolated cores * competitive accretion in a protocluster environment * stellar collisions and mergers in very dense systems we also review the observed outflows , multiplicity , and clustering properties of massive stars , the upper initial mass function and the upper mass limit .
we conclude that high - mass star formation is not merely a scaled - up version of low - mass star formation with higher accretion rates , but partly a mechanism of its own , primarily owing to the role of stellar mass and radiation pressure in controlling the dynamics . to appear in annual reviews of astronomy and astrophysics , vol .
45 , 2007 * toward understanding + massive star formation * hans zinnecker harold w. yorke * key words * accretion , circumstellar disks , hii regions , massive stars , protostars , star formation |
@xmath1 scuti stars ( hereinafter , @xmath1 sct stars ) are pulsating variables inside the classical instability strip and on or close to the main - sequence
. they are typically placed at the lower on the instability strip than rr lyrae stars or cepheids and thus they are fainter than rr lyrae stars or cepheids .
their spectral types are between a and late f. their periods are between @xmath20.02 days and @xmath20.25 days , which is relatively shorter than other types of variables ( e.g. @xmath3 dor ; @xcite ) . based on these characteristics , @xmath1 sct stars can be separated from other types of variable stars such as rr lyrae , @xmath4 cepheid , @xmath3 dor etc @xcite .
the majority of the @xmath1 sct stars are low - amplitude @xmath1 sct stars ( lads ) with amplitudes from a milli - magnitude to a few tens of milli - magnitude .
lads are mainly non - radial p - mode pulsators @xcite .
another subgroup of @xmath1 sct stars is the high - amplitude @xmath1 sct stars ( hads ) , whose amplitudes are bigger than @xmath20.3 magnitude . hads are radial pulsators @xcite .
in addition to lads and hads , there is another interesting type pulsation star called sx phe variable stars , which exhibit a type of pulsation similar to the @xmath1 sct stars .
they are relatively old and evolved population ii stars , whereas most of the @xmath1 sct stars are population i stars .
stellar evolutionary theory is not yet successful at explaining these sx phe variable stars @xcite .
most of the sx phe show similar properties with hads such as high amplitude and short period .
more detailed review of @xmath1 sct stars is presented in @xcite @xcite and references therein .
because of their great number of radial and non - radial modes , it is known that @xmath1 sct stars are suitable for asteroseismology research , which enables study of stellar interior structures @xcite . for a better understanding of pulsating @xmath1 sct stars and
thus stellar structure , several authors studied @xmath1 sct stars and detected their multiple frequencies of pulsation using either ground - based observations or space - based observations @xcite . due to
the better photometric precision , space - based observations data show better results on the analysis of multiple frequencies than ground - based observation data @xcite . however , some authors have pointed out that ground - based observations using multiple - site telescopes are still valuable because , with a baseline longer than space - based observations , they are useful for detecting long - period pulsation ( for more details , see @xcite and references therein ) . moreover , by parameterizing the amplitude ratio and the phase differences in different filters ( e.g. @xmath5 ) , it is possible to derive the spherical harmonic degree , @xmath6 @xcite , which is an important parameter for the asteroseismology studies .
therefore ground - based telescopes that are more feasible for multiple - site and multiple - filter observations ( e.g. delta scuti network , @xcite ) are nonetheless useful for the identification of pulsation modes and thus for the study of interior structures .
another interesting feature of @xmath1 sct stars is that some of them show period and amplitude variations @xcite .
the period variation ( 1/p ) dp / dt , based on observations , is about 10@xmath7 per year for both period increases and decreases with equal distribution . on the other hand ,
theoretical models give ten times smaller period variation than observed ; they also predict that period increases should be dominant over period decreases @xcite .
amplitude variations and time scales of the variations are different from star to star , ranging from a few milli - magnitudes to several tens of milli - magintudes and from a few tens of days to a few hundreds of days @xcite . these period and amplitude variations
are thought to be caused not by evolutionary effects but by some other mechanism ( e.g. light - time effect because of the orbital motion in binaries or nonlinear mode interactions ) .
however , the true origin of the variations is still unknown . for more details , see @xcite and references therein .
@xcite investigated hads in the large magellanic cloud ( lmc ) and their period - luminosity (
p - l ) relation to test if they can be used as the standard distance candles .
they found that the distance modulus for lmc derived using @xmath1 sct stars is consistent with the distance moduli for lmc derived using rr lyrae and cepheids , which implies the p - l relation of @xmath1 sct stars can help to determine distances of _ long - distance _ objects such as objects in the lmc . in this paper , we present the detection of 41 @xmath1 sct candidate stars from the taiwan - american occultation survey ( taos ) data accumulated during 2005 and 2006 observation ( hereinafter , taos @xmath1 sct stars ) . among the 41 detections ,
there is one previously known ` suspected variable ' star , nsv 3816 , from the suspected variable stars and supplement @xcite ( no period or type is provided in that catalog ) .
the rest of the 40 taos @xmath1 sct stars are newly detected by this study .
only 14 of the detected taos @xmath1 sct stars have spectral types .
twelve of those have spectral types from a to f , which are typical for @xmath1 sct stars .
the remaining two have b8 and g5 spectral types , which are peculiar spectral types .
using spectroscopic instruments -boes @xcite and fast @xcite- , we obtained spectra for those two stars . as a result we found that the b8 star is an a5 star and
the g5 star is an f0 star .
even though the rest of the detected stars do not have spectral type information , their low amplitudes , short periods and morphologies of light - curves strongly suggest that they are lads . in section [ sec : taos_delta_scuti ] , we present a taos overview , data reduction processes and the detection algorithm we used to detect @xmath1 sct stars in taos 2-year data .
we provide a list of the detected taos @xmath1 sct stars and their physical parameters ( e.g. magnitude , period , amplitude , spectral type , etc ) in section [ sec : detectionresults ] . in section [ sec : conclusion ] , we present summaries .
taos aims to detect stellar occultations caused by small - sized kuiper belt objects ( kbos ) at a distance of neptune s orbit or beyond @xcite . because of the short duration ( @xmath8 sec ) and the rareness of occultation events , taos monitors several hundreds of stars in a wide field of view ( 3 deg@xmath9 ) with a high sampling rate . to reduce false positives , taos uses four @xmath10 cm telescopes which simultaneously monitor the same patch of the sky . due to the high sampling rate ,
taos data is also useful for detecting short - period variable stars such as @xmath1 sct stars .
moreover , taos telescopes keep monitoring the same field up to 1.5 hours and can thus obtain full - phase light - curves of variable stars whose periods are shorter than 1.5 hours .
to detect such short - period variable stars , we analyzed taos data accumulated during 2005 and 2006 .
the dataset consists of 117 taos observation fields , which cover 351 @xmath11 of the sky .
it consists of @xmath2200 _ runs _ , where a run is a set of multiple ( two or three ) telescope observations for a given field and a given date .
note that the taos telescopes occasionally visit the same observation field multiple times according to the telescopes observation schedules , which enables detecting the same variable stars multiple times . in such case
, we are able to derive multiple frequencies of the stars as explained in section [ sec : listdeltascuti ] . to detect periodic signals , we analyzed the light - curves generated by the taos photometry pipeline @xcite .
the pipeline was developed by the collaboration to extract light - curves of each star from _
zipper images_. the _ zipper images _ are generated by the unique telescope operation mode called _ zipper mode _ which was developed in order to achieve high - speed photometry @xcite . after obtaining the light - curves using the taos photometry pipeline , we applied further cuts to the light - curves .
some of the individual measurements are flagged as invalid .
this happens when the star moves out of the field of view because of temporary telescope vibrations or tracking error , thus yielding no photometrical measurements .
we therefore applied a b - spline @xcite and replaced the flagged measurements with values interpolated from the spline fit .
after the interpolation process , in order to increase the signal - to - noise ratio ( snr ) , we binned each light - curve using a 50 point window ( 10 sec ) . during the binning process
, we used the average time of the 50 data points as the time of the binned data .
we then removed the systematic variations that are common across light - curves of the same run .
such systematic variations , which we call _ trends _ , could be caused by airmass , temporary telescope vibrations , noise in ccd images , etc .
to remove such trends , we applied the photometric detrending algorithm ( pdt , @xcite ) to each individual run .
pdt first calculates the correlation between whole light - curves as a measure of similarity between light - curves .
pdt then uses the hierarchical clustering algorithm @xcite to group similar light - curves together and determines one _ master - trend _ per group by summing weighted light - curves in the group . using the determined master - trends
, pdt finally removes trends from each individual light - curve by minimizing the residual between the master - trends and the light - curve . for more details about pdt ,
see @xcite . figure [ fig : lc_example ] shows an example of a taos @xmath1 sct star s light - curve before and after detrending process .
x - axis is time in minutes and y - axis is flux . as the figure shows , periodic signals are clearly recovered after detrending .
we show the errors for each photometric measurement before detrending , propagated from the errors estimated by the taos photometry pipeline @xcite .
x - axis is time in minutes and y - axis is flux .
the top panel is the light - curve before detrending and the bottom panel is the light - curve after detrending .
the periodic signals contaminated by unstable weather ( e.g. moving clouds ) are successfully recovered after detrending .
we show the errors for each photometric measurement of the raw light - curve.,scaledwidth=45.0% ] after we finished the preprocessing , explained in the previous section , we applied the fast fourier transform algorithm ( fft ; @xcite ) to each light - curve in order to detect periodic signals .
note that the individual measurements of taos light - curves are evenly spaced with a 5hz sampling rate .
thus fft is appropriate for the detection of periodic signals .
we focused on the detection of short - period variable stars whose periods are @xmath12 hour because taos monitors a given field for a maximum of 1.5 hours .
we describe the basic steps of the detection process below : * we apply fft to each detrended light - curve and derive the power spectrum of the light - curve .
[ item : fft1 ] we then examine if there exists a frequency ( or frequencies ) whose power is bigger than five times the standard deviation of powers of the background frequencies .
the standard deviation of powers is calculated after removing outliers using 3-sigma clipping .
we identify the star as a variable candidate if there is a frequency higher than five times the standard deviation .
[ item : fft2 ] * for each candidate variable we check if the periodic signal is detected on the other telescopes light - curves of the same run .
if it is not detected by the other telescopes , we remove the star from the candidate list .
[ item : fft5 ] * we visually inspect all raw zipper images for the candidates and remove false positives caused by moving asteroids , photometry defects or other contamination due to various noise sources . for instance
, the flux of stars in the neighborhood of fast moving objects could be increased and decreased within an hour , which resembles periodic signals .
[ item : fft6 ] * we cross - match all of the candidates with simbad @xcite and remove the false positives that are confirmed to be other types of variable stars ( e.g. eclipsing binary stars ) . *
finally we remove the variable stars whose periods are longer than 1.5 hours .
with the detection algorithm described in the previous section , we found 41 @xmath1 sct candidate stars whose periods are shorter than 1.5 hours and whose amplitudes are within a few hundredth of a magnitude ( hereinafter , taos @xmath1 sct stars ) . among those 41 taos @xmath1 sct stars ,
one of them is a previously _ suspected _ variable star , nsv 3816 , from the suspected variable stars and supplement @xcite .
however , the period and amplitude of nsv 3816 have never been published before .
the remaining 40 taos @xmath1 sct stars are newly detected by this study .
after the taos @xmath1 sct stars were identified , we extracted the physical parameters of each star by cross - matching them with various astronomical catalogs .
we show catalogs we used in table [ tab : catalogs_parameters ] .
we found that 12 of the detected 41 taos @xmath1 sct stars have spectral types from a0 to f5 , which are typical spectral types for @xmath1 sct stars .
unfortunately the rest of them , except for two peculiar @xmath1 sct stars -discuss in section [ sec : spectroscopy]- , do not have spectral information .
nevertheless , their short period and low amplitude strongly suggest that they are lads rather than other types of variables , such as rr lyrae or cepheids , whose periods and amplitudes are relatively longer and larger than those of @xmath1 sct stars .
cc catalog & reference + gcvs & @xcite + all - sky compiled catalogue of 2.5 million stars & @xcite + hd & @xcite + catalogue of stellar spectral classifications & @xcite + tycho-2 catalogue of the 2.5 million brightest stars & @xcite + guide star catalog ( gsc ) & @xcite + usno - b 1.0 & @xcite + sao star catalog j2000 & @xcite + catalog of projected rotational velocities & @xcite + rotational velocity determinations for 118 @xmath1 sct variables & @xcite + as a byproduct of our analysis , we detected a previously known variable star with @xmath1 sct pulsation , gm leo , which is actually a @xmath13 bootis star @xcite .
some @xmath13 bootis stars show @xmath1 sct pulsations @xcite and can have spectral types from late b to early f , which makes it difficult to distinguish them from @xmath1 sct stars . in such cases ,
there are no clear differences between @xmath1 sct stars and @xmath13 bootis stars except the metal abundance @xcite ; @xmath13 bootis stars show weak metal lines such as mg @xmath134481 line @xcite .
we also checked several preexisting catalogs of @xmath1 sct stars to see if there are previously known @xmath1 sct stars in the taos observation fields .
table [ tab : catalogs ] shows the preexisting catalogs we checked . as a result of this search
, we found only one previously known @xmath1 sct pulsation star to be in the taos observation fields . that turns out to be gm leo , which as we mentioned above , we successfully detected .
although gcvs classified gm leo as @xmath1 sct star based on the work by @xcite , @xcite in their paper claimed gm leo is not a @xmath1 sct star but a @xmath13 bootis star .
thus we removed gm leo from our detection list .
cccc catalog & source surveys & number of @xmath1 sct stars & reference + r2000@xmath14 & macho , ogle , hipparcos , etc & @xmath2600 & @xcite + rotse@xmath15 & rotse & 6 & @xcite + asas@xmath16 & asas & @xmath2500 & @xcite + gcvs@xmath17 & various surveys & @xmath2500 & @xcite + taos & taos & 41 & this paper + & & & + others & & & + & 5 new @xmath3 doradus and 5 new @xmath1 sct survey & 5 & @xcite + & case study for hd 173977 & 1 & @xcite + & case study for hd 8801 & 1 & @xcite + & the first hads in an eclipsing binary star & 1 & @xcite + & variable stars in ngc 2099 & 9 & @xcite + & transit survey of m37 & 2 & @xcite + & asas variable stars in the kepler field of view & 4 & @xcite + @xmath14 the catalog compiled by @xcite + @xmath15 robotic optical transient search experiment .
+ @xmath16 all - sky automated survey .
+ @xmath17 general catalog of the variable stars .
table [ tab : delta_scuti ] shows the 41 taos @xmath1 sct stars physical information such as positions , magnitudes , frequencies , amplitudes , spectral types , etc .
we used the fft algorithm to detect periodic signals however we used period04 to derive their physical parameters such as period and amplitude .
this is because period04 improves the frequency by fitting the light - curve with a combination of sine curves .
moreover period04 also provides errors for the derived frequencies .
figure [ fig : comparespectrum ] shows a comparison result of power spectra derived from an fft method and period04 for a single taos @xmath1 sct star .
x - axis is frequency in counts / day , y - axis is scaled power .
solid line is power spectrum derived from an fft method and dashed line is power spectrum derived from period04
. as the figure shows , the two spectra are almost consistent .
as we mentioned in the previous section , taos operates multiple telescopes simultaneously monitoring the same patch of the sky .
thus we have a maximum of three simultaneous light - curves for all taos @xmath1 sct stars for a given zipper run . to derive more precise frequencies and amplitudes , for each identified @xmath1 sct star we summed the light - curves from each of the telescopes .
moreover , the taos telescopes occasionally visit same fields multiple times . in such cases ,
we merge all corresponding normalized light - curves of each identified @xmath1 sct star into a single but longer light - curve . having longer light - curves we were able to extract multiple frequencies using period04 . among the extracted frequencies
, we selected the frequencies whose snr is bigger than five .
the snr of each frequency was calculated using period04 as well .
@xcite , we empirically found that a threshold of snr @xmath18 4 produces false positives and thus we set the snr threshold to five . ] as a result we found 16 taos @xmath1 sct stars having multiple frequencies ( see table [ tab : delta_scuti ] ) .
note that we did not attempt to extract multiple frequencies if the star is detected only once ( i.e. detected in only single zipper run ) .
it is also worth mentioning that we lose detectability on relatively long period pulsations because we normalized the light - curves while merging them .
all detected @xmath1 sct stars are relatively bright as shown in the table ( the faintest star s @xmath19 is around 12 ) .
this is because the limiting magnitude of taos _ zipper mode _ is relatively bright at @xmath213.5 @xcite and also because high snr is needed to detect low - amplitude variations of a few milli - magnitude . to find @xmath19 , @xmath20 and spectral types , we used centre de donnees astronomiques de strasbourg ( cds ) web service @xcite . in table
[ tab : delta_scuti ] , we also provide the number of the total observations ( i.e. the number of the total zipper runs ) and the number of identifications by the fft analysis for each taos @xmath1 sct star .
since the taos telescopes observe same fields multiple times , we have multiple light - curves for stars in the fields .
note that we applied the fft algorithm to each light - curve to detect periodic signals .
nevertheless , as the table shows , not every light - curve of the taos @xmath1 sct stars were confirmed to have periodic signals .
this is mainly because of the poor quality of some of the light - curves caused by trends and noise ( i.e. unstable weather , telescope vibration , etc ) .
although we removed most of the trends using pdt , it is nearly impossible to recover the intrinsic periodic signal of a few milli - magnitude in the presence of large systematic errors . in figure
[ fig : nyquist_sw ] we show an example of a spectral window of a single zipper run . since the observational times are almost equally spaced , peaks with regular intervals ( two times the nyquist frequency ) are present in the spectral window @xcite .
the nyquist frequency is 4320hz since the gap between each consecutive binned data is 10 sec .
figure [ fig : worst_ps ] , [ fig : inter_ps ] and [ fig : best_ps ] show the spectral windows along with the power spectra of three taos @xmath1 sct stars .
stars included in the figures are 020.00141 ( figure [ fig : worst_ps ] ) , 121.00043 ( figure [ fig : inter_ps ] ) and 054.00014 ( figure [ fig : best_ps ] ) .
the top panels in each figure show the spectral windows and the bottom panels ( and the middle panel in figure [ fig : best_ps ] ) show the power spectra of the stars .
dashed lines indicate the detected frequencies .
note that we improved the detected frequencies by fitting a combination of sine waves using period04 .
thus the improved frequencies could be slightly shifted from the original peaks in the power spectra ( e.g. see the bottom left panel in figure [ fig : worst_ps ] ) after the fitting . as table [ tab : delta_scuti ] shows , the star with i d 020.00141 was identified only once and is relatively fainter ( @xmath21 11.40 ) than other taos @xmath1 sct stars .
its amplitude is one of the smallest amplitudes and its detected frequency snr is the lowest among taos @xmath1 sct stars .
thus the power spectrum of this star represents one of the `` worst case scenario '' .
we detected one single frequency for this star .
the star with i d 121.00043 was identified three times and is relatively bright ( @xmath19 = 9.13 ) . the spectral window and the power spectrum of the star could represent a `` moderate - level scenario '' .
we detected one frequency for this star .
finally , the star with i d 054.00014 was identified nine times and is relatively bright ( @xmath19 = 9.50 ) .
thus its spectral window and power spectrum represents one of the `` best case scenario '' . using period04 , we detected three frequencies for this star .
as can be seen from the figures , there are no significant peaks in the spectral windows at the detected frequencies .
as we mentioned in the previous section , we found two peculiar spectral type @xmath1 sct stars which have b8 and g5 spectral types .
these are the bluest and reddest spectral types of @xmath1 sct stars ever detected .
the spectral type of the b8 star was extracted from the henry draper catalogue and extension ( hd , @xcite ) .
unfortunately we could not find any spectroscopic literature for the g5 star so we suspect that the spectral type is most likely derived from its color information .
the spectral type of the g5 star was extracted from simbad . to confirm their spectral types , we observed the two stars with spectroscopic instruments . for the b8 star
, we used the boes of the 1.8-m telescope at the bohyunsan optical astronomy observatory ( boao ) , south korea @xcite .
we used iraf @xcite for the reduction of the obtained spectroscopic data .
figure [ fig : spectrum ] shows the normalized spectrum of the b8 candidate star .
we indicate several important spectral lines in the figure .
as the figure shows , ca k line is very strong which is typical for a type stars @xcite .
b type stars do not show such strong ca k line .
the spectrum also shows weak metallic lines ( e.g. ca and mg line ) which are usually presented in a type stars . based on the strength of ca k line , hydrogen lines and metallic lines ,
the star is likely an a5 type star although the classification of sub - class is rather uncertain due to the low snr of the spectrum .
in addition , to observe the g5 star , we used the fast instrument mounted at the fred lawrence whipple observatory ( flwo ) 1.5 m telescope , mount hopkins in arizona @xcite . after comparing
the observed data with standard spectral libraries @xcite , we found that the spectral type of the star is not a g5 but an f0 .
therefore the star is likely a typical @xmath1 sct star .
we analyzed the taos 2-year data accumulated during 2005 and 2006 observations in order to find short - period variables . using the taos photometry pipeline we created photometric light - curves .
we removed systematic trends commonly appeared in the light - curves using pdt . to detect periodic signals in the detrended light - curves
, we applied the fast fourier transform ( fft ) to each light - curve .
fft is a simple but _ powerful _
algorithm for detection of periodic signals when data points are evenly spaced .
we then chose light - curves which possess a frequency ( or frequencies ) whose power is five times larger than the standard deviation of powers of all background frequencies in the power spectrum derived using fft .
we visually checked the light - curves and raw images of all candidates to remove false positives caused by moving asteroids , photometry defects , etc .
we also removed candidates which were detected by only one of the three telescopes .
all remaining 41 variable candidates have periods about an hour and amplitudes less than a few hundredth of a magnitude , which strongly suggests that they are low - amplitude @xmath1 sct stars ( lads ) .
we cross - matched the detected @xmath1 sct candidate stars with many astronomical catalogs to extract additional information ( e.g. magnitude , spectral type , variability type , etc ) . as a result we found that 14 stars have spectral types from a to f , which are typical spectral types for @xmath1 sct stars .
the rest of the detected @xmath1 sct stars do not have spectral information .
the light - curves of taos @xmath1 sct stars are accessible at the time series center ( tsc , http://timemachine.iic.harvard.edu ) , initiative in innovative computing ( iic ) at harvard .
period04 project files of each star is also provided .
the project files contain complete light - curve data , power spectrum , frequency and amplitude information .
byun acknowledges the support of national research foundation of korea through grant 2009 - 0075376 .
the work at national central university was supported by the grant nsc 96 - 2112-m-008 - 024-my3 .
work at academia sinica was supported in part by the thematic research program as-88-tp - a02 .
work at the harvard college observatory was supported in part by the national science foundation under grant ast-0501681 and by nasa under grant nng04g113 g .
slm s work was performed under the auspices of the u.s .
department of energy by lawrence livermore national laboratory in part under contract w-7405-eng-48 and by stanford linear accelerator center under contract de - ac02 - 76sf00515 .
k. h. cook s work was performed under the auspices of the u.s .
department of energy by lawrence livermore national laboratory in part under contract w-7405-eng-48 and in part under contract de - ac52 - 07na27344 .
we also thanks j. d. hartman at harvard - smithsonian center for astrophysics for useful discussion .
the detrending and the analysis of datasets in this paper were run on the odyssey cluster supported by the fas research computing group at the harvard .
this research has made use of the simbad database , operated at cds , strasbourg , france .
iraf is distributed by the national optical astronomy observatories , which are operated by the association of universities for research in astronomy , inc .
, under cooperative agreement with the national science foundation . | we analyzed data accumulated during 2005 and 2006 by the taiwan - american occultation survey ( taos ) in order to detect short - period variable stars ( periods of @xmath0 1 hour ) such as @xmath1 scuti .
taos is designed for the detection of stellar occultation by small - size kuiper belt objects ( kbos ) and is operating four 50 cm telescopes at an effective cadence of 5hz .
the four telescopes simultaneously monitor the same patch of the sky in order to reduce false positives . to detect short - period variables , we used the fast fourier transform algorithm ( fft ) inasmuch as the data points in taos light - curves are evenly spaced . using fft
, we found 41 short - period variables with amplitudes smaller than a few hundredths of a magnitude and periods of about an hour , which suggest that they are low - amplitude @xmath1 scuti stars ( lads ) .
the light - curves of taos @xmath1 scuti stars are accessible online at the time series center website ( http://timemachine.iic.harvard.edu ) . |
coupled oscillations are encountered in many contexts in nature and technology .
their prevalence spans over such diverse fields as pacemaker cells in the heart muscle @xcite , neural activity in the brain @xcite , long term climate patterns @xcite or the electrical power grid @xcite .
all such systems have in common that their dynamical behavior results from the interaction between a local oscillation and the coupling between the individual oscillators .
thus , the dynamics depends on generic properties of the oscillations such as their harmonical or relaxational character and the way they interact with each other , which might be local ( diffusive ) , long - range or global @xcite . therefore , the study of a specific system yields also universal information of spatio - temporal organization of other oscillatory systems @xcite . in this context
, we study an electrochemical oscillatory medium , the photoelectrodissolution of n - doped silicon , with a spatial coupling that produces novel types of synchronization patterns . + + our work ties in with previous studies of oscillatory media in chemical and physical systems .
best known are the belousov - zhabotinsky ( bz ) reaction and the oxidation of co on pt in uhv @xcite , which proved to be prototypical systems with diffusive coupling , exhibiting a variety of dynamically different states , such as sprial waves and chemical turbulence .
in addition to the diffusive , i.e. local , coupling , there are many physical mechanisms leading to a long - range or even global coupling of the oscillators .
this applies , e.g. , to electrically controlled systems , such as semiconductor devices @xcite , gas discharge tubes @xcite or electrochemical systems @xcite .
the most prominent patterns emerging from a global coupling are scale free cluster patterns where the medium separates into a few domains , oscillating with a certain , constant phase lag @xcite .
+ + in parallel to these experimental works , theoretical studies employing the complex ginzburg - landau equation ( cgle ) or phase oscillator models as universal descriptions of oscillatory systems were carried out @xcite .
the former is a normal form equation of an oscillatory medium close to the hopf bifurcation where the limit cycles are only weakly attractive , while the latter captures the dynamics of strongly attractive , coupled limit cycles . among the vast literature on the solutions of these equations , cluster patterns are most relevant in the context of our work @xcite .
typically , one distinguishes between phase clusters , where all oscillators follow the same limit cycle but form distinct groups with a constant phase difference , and amplitude clusters , where the different groups occupy different limit cycles with distinct amplitudes and phases .
both types of clusters emerge in globally coupled systems , however , phase clusters require coupling functions with higher harmonics . thus ,
in contrast to amplitude clusters , they can not be described in the framework of the cgle with a linear global coupling . in our work
we encounter cluster patterns with novel features , most remarkably self - organized chimera states and other types of coexistence of different dynamical states on the electrode surface , and we discuss that a nonlinear coupling leads to their emergence . furthermore , a collection of other peculiar patterns and unusual synchronization phenomena is presented .
+ + our experimental system is the photoelectrochemical oxidation of n - doped silicon to sio@xmath0 under potentiostatic conditions and the concurrent dissolution of this oxide mediated by fluoride species . owing to this interplay of oxidation and etching process
the system exhibits oscillations in current and oxide layer thickness in a broad parameter regime @xcite .
the electrochemical oxidation equation reads : @xmath1 where @xmath2 denotes the number of charge transfer processes via the valence band of the silicon , i.e. , processes involving minority charge carriers in n - doped silicon .
these holes drive the initial charge transfer step in equation [ eq : oxreac ] , i.e. , the inequality @xmath3 holds and there is no oxidation current without illumination @xcite . a limited illumination and , therefore , limited generation rate of holes in the valence band of the working electrode thus introduces a cut - off for the total current .
it constitutes a global constraint as well as a nonlinear coupling mechanism between the oscillating electrode positions .
+ an additional , global coupling mechanism is introduced by connecting a resistance @xmath4 in series to the working electrode in the potentiostatically controlled experiments .
this external resistance couples the voltage drop @xmath5 across the electrode@xmath6oxide@xmath6electrolyte interface at each individual point of the surface of the working electrode to the total current through the working electrode and the introduced coupling consequently has a linear characteristic .
@xmath7 here , @xmath8 is the externally applied voltage , @xmath9 the spatial average of the current density and @xmath10 the electrode area .
+ + in order to get a comprehensive picture of the dependence of the patterns on the experimental parameters , four of these parameters are varied over a wide range where current oscillations occur .
two of these parameters characterize the electrolyte , the first one being the total concentration of fluoride species @xmath11 and the second one the ph value which determines the distribution of the fluoride on the three species f@xmath12 , hf and hf@xmath13 according to the law of mass action .
both parameters together determine the total silicon oxide etching rate as the different fluoride species follow different etching pathways with distinct rates @xcite .
the other two parameters represent the strengths of the coupling mechanisms given by the illumination intensity @xmath14 for the nonlinear coupling and @xmath15 for the purely linear global coupling , respectively .
+ + after a description of the experimental setup for the spatially resolved recording of the oxide layer with an ellipsometric imaging technique and the data processing in section [ sec : exp ] , an overview of the parameter dependence of the pattern forming regime is given and the patterns occurring in the experiments are classified by spatially averaged quantities .
subsequently , in sections [ subsec : p1]-[subsec : irreg ] for each such class the different prototypic patterns encountered in the experiments are shown in a systematic way .
additionally , the requirements on a generic model exhibiting such spatio - temporal dynamics are discussed .
the working electrodes are the ( 111 ) faces of monocrystalline , n - type doped silicon samples with a resistivity of @xmath16 cm .
they are ohmically back - contacted with a thermally evaporated 200 nm aluminum layer annealed at @xmath17c for @xmath18 minutes .
the silicon is attached to a custom made ptfe holder via silicone paste ( scrintex 901 , ralicks gmbh , rees - haldern , germany ) which also serves as a sealing for the back - contact against the electrolyte .
the typical electrode size is @xmath19 mm@xmath20 . as a counter electrode ,
a circular shaped platinum wire is placed symmetrically in front of the working electrode and a saturated hg@xmath6hg@xmath0so@xmath21 reference electrode is used which is placed behind the working electrode at a distance of several centimeters . during the experiments the electrolyte
is constantly stirred with a magnetic stirrer rotating at about 10 hz . the sample illumination is realized by a he - ne laser and the illumination intensity is adjusted by a polarizer plate .
the laser beam is optically expanded to a diameter of ca .
1.5 cm and passes an iris diaphragm so that only its central part is incident on the sample to guarantee a spatially uniform illumination .
+ + for the ellipsometric imaging of the optical path through the oxide , light from a blue led ( @xmath22 nm ) is first elliptically polarized by passing a linear polarizer and a zeroth order @xmath23/4-plate .
it then hits the sample surface at an angle close to the brewster angle of the silicon@xmath6electrolyte system ( @xmath24 ) .
the reflected light passes another polarizer which converts changes in the polarization upon reflection caused by the optical path through the oxide into an intensity signal . using a suitable lens , an image of this ellipsometric intensity distribution across the surface
is then created on the ccd chip of a camera ( jai cv - a50 ) .
the data is then digitized using a frame grabber card ( pci-1405 , national instruments ) and preprocessed and recorded by a labview program .
a sketch of the setup ( with the optical components of the ellipsometric imaging path omitted ) is shown in figure [ fig : sketch ] .
+ + before the experiments the silicon working electrodes are cleaned by first rubbing them carefully with an acetone soaked tissue and subsequently immersing them in acetone ( p.a . ) , ethanol ( p.a . ) , methanol ( p.a . ) and ultrapure water for 5 min each .
the electrolyte is always an aqueous solution of nh@xmath21f ( merck , p.a . ) as a fluoride source and h@xmath0so@xmath21 ( merck , suprapur ) to adjust the ph value .
prior to the experiments the electrolyte is bubbled with argon and an argon atmosphere is retained during the experiments .
+ + all experiments are potentiostatically controlled ( potentiostat : fhi-2740 , electronics laboratory of the fritz - haber- institut , berlin , germany ) .
a constant voltage of 8.65 v vs. standard hydrogen electrode is applied between the reference and the working electrode with a tunable ohmic resistance connected in series to the working electrode . + + for the analysis of the patterns the spatial distribution of the ellipsometric intensity at specified points in time is considered as well as the ellipsometric signal of one dimensional subsets of the electrode over the entire time window .
furthermore , the time series of the ellipsometric intensity at each individual point is fourier transformed ( fft , matlab ) and the spatial distribution of absolute value and phase of relevant fourier coefficients is then used to gain insight into phase clustering behavior for these fourier modes .
lastly , the data from a selected area on the electrode is hilbert transformed ( matlab ) and the resulting analytical signal , i.e. the complex signal with the original data as the real part and the hilbert transform of the data as the imaginary part , is used to visualize the different dynamics on the electrode .
the occurrence of oscillations and the accompanying pattern formation for a given electrolyte depends on the values of the scaled external resistance @xmath15 and the illumination intensity @xmath14 .
an exemplary 2d parameter space is shown in figure [ fig : parameter ] . and the illumination intensity @xmath14 for an electrolyte with @xmath11=50 mm and ph 2.3 .
uniform oscillations ( red ) and oscillations accompanied by spatial pattern formation ( blue ) are shown together with stable fixed points with ( circle ) and without ( cross ) an oscillatory transient outside of the oscillatory regime .
the borders of the oscillatory and the pattern forming regime are indicated by lines.,scaledwidth=35.0% ] at high @xmath14 and low @xmath15 , the system settles to a stable focus .
conversely , if the external resistance is set at a too high value , the system relaxes monotonically to a stable node . in between , within the oscillatory regime ,
both parameters have an impact on the local oscillators , but as explained above they mediate also a coupling between the individual positions on the electrode . the impact on local dynamics and spatial coupling
can not be separated in the experiments . however , as we elaborate below , we obtain a consistent picture if we assume that the principle trend of the respective locations of patterns and synchronized oscillations in the parameter plane is dominated by the two spatial coupling mechanisms rather than the change in the properties of the local oscillators .
+ + for relatively high illumination intensities , i.e at a low nonlinear coupling strength , the entire electrode oscillates uniformly reflecting a strong synchronizing spatial coupling .
an example for such a uniform oscillation is shown in figure [ fig : nurhom1 ] .
mm , @xmath25 , @xmath26 , @xmath27 mw / cm@xmath20 ) . *
a ) * time series of the global quantities @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] uniform oscillations are also found in all oscillatory measurements at non - illuminated p - type silicon samples , as discussed in detailed studies of such samples conducted previously in our group @xcite .
in addition to the uniformity , the shape of the oscillations for highly illuminated n - type doped silicon and p - type doped silicon samples are identical at otherwise matched experimental parameters and in both cases a minimal external resistance has to be introduced .
this similarity in behavior can be fully attributed to the sufficiency of the amount of holes in the valence band to maintain the current present in both cases , strongly corroborating the notion that the nonlinear coupling introduced by the restriction of holes is solely responsible for the emergence of patterns .
consequently , in the parameter space shown in figure [ fig : parameter ] only at illumination intensities below an electrolyte specific threshold value of @xmath31 mw / cm@xmath20 pattern formation can be observed .
if the linear global coupling is too strong , the patterns vanish and uniform oscillations are observed although the current is still restricted by the illumination and an effect of the nonlinear coupling is thus still to be expected .
this behavior is in accordance with literature where the global coupling through an external resistor tends to synchronize the oscillations , at least when the electrode potential acts as the activator @xcite . as a consequence patterns are only found in the parameter region where the nonlinear coupling dominates the linear global coupling , i.e. , in the lower left hand corner of figure [ fig : parameter ] .
most notably , here , sustained oscillations are observed even without an external resistance . compared to patterns classically found in oscillatory media
the silicon oscillator exhibits two characteristic and peculiar phenomena : first , in a large part of the pattern forming parameter region the global quantities , i.e. , the total current @xmath29 and the spatially averaged ellipsometric intensity @xmath30 oscillate in a simple periodic fashion although local time series exhibit complex oscillation forms up to completely irregular ones . besides , @xmath30 may oscillate with a period-2 or irregularly .
second , typically different dynamic behaviors as for example cluster patterns and spatio - temporal chaos coexist on the electrode .
we call a part of the surface with one distinct dynamics a region . + + in the following we classify the patterns , in a first step , according to the spatially averaged ellipsometric intensity @xmath30 into three categories : * simple periodic oscillations * period-2 oscillations * oscillations with an irregular amplitude exemplary time series for the three types of @xmath30 oscillations are shown in figure [ fig : classification ] . for three different measurements showing simple periodic oscillations ( top ) , period-2 oscillations ( middle ) and oscillations with an irregular amplitude ( bottom).,scaledwidth=35.0% ] in the case of simple periodic oscillations in @xmath30 , 2d simulations using the suitably modified complex ginzburg - landau equation with a nonlinear global coupling show striking similarities with the patterns observed in the experiments @xcite .
this opens a pathway to a possible deeper understanding of both the origin of the pattern formation and the coupling mechanism .
consequently , in the following the emphasis is first put on the analysis of the first class of patterns and the second and third class are treated as modifications of this first class and are discussed later .
+ + in a first step we now consider the patterns forming during simple periodic @xmath30 oscillations depicted in the top of figure [ fig : classification ] .
the regularity of the spatially averaged signal implies that the spatial variations of the patterns cancel out on average .
there are two dominant types of patterns forming .
first , subharmonic phase clusters are found , where the ellipsometric intensity at each point shows oscillations with a strong frequency component at an integer fraction of the spatially averaged oscillation frequency .
these points are arranged in phase domains where all points within a domain are synchronized and the distinct domains differ in their phase @xcite . typically but not exclusively , 2-phase clusters are found .
the second type of dynamical behavior found are desynchronized oscillations with a broad distribution of oscillation frequencies .
hence , at any given point the oscillation amplitude is irregular . + + as already mentioned , typically the electrode forms different regions . in a region
either one of the two patterns mentioned or a synchronized oscillation with the frequency of the spatially averaged signal exists .
the occurrence of only one type of dynamical behavior on the entire electrode surface and any pairwise coexistence have been found in this study and are subsequently presented .
first , we take a look at the patterns with only one region covering the entire electrode .
we start by discussing 2-phase clusters .
an example is shown in figure [ fig : nursh1 ] .
mm , @xmath25 , @xmath32 , @xmath33 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity ; * c ) * 1d cut as in b ) where @xmath30 is subtracted from every local time series .
the color scale is not identical to b).,scaledwidth=35.0% ] as apparent from figure [ fig : nursh1 ] a ) the spatially averaged ellipsometric intensity @xmath30 as well as the spatially averaged current density @xmath29 oscillate in a simple periodic fashion . at the same time this oscillation is not uniform as can be seen in figure [ fig : nursh1 ] b ) where two snapshots of the ellipsometric images and the temporal evolution of a 1d cross section are depicted .
clearly , at every point , an amplitude variation can be seen which inverts every other oscillation . for a better visibility of the spatial distribution of the amplitude variations
the underlying uniform oscillation is subtracted in subfigure [ fig : nursh1 ] c ) , uncovering subharmonic local oscillations .
the characteristics of this structure are further elucidated by the results of a fourier analysis of the local timeseries at each point of the surface shown in figure [ fig : nursh2 ] . .
* a ) * spatial average of the absolute values of the pointwise fourier spectra ( blue ) and fourier spectrum of the spatially averaged signal @xmath30 ( red ) ; * b ) * spatial distribution of the absolute value ( left ) and phase ( right ) of the fourier coefficient corresponding to the fourier mode indicated by the vertical line in the fourier spectra in a).,scaledwidth=35.0% ] in subfigure [ fig : nursh2 ] a ) , the spatial average of the absolute values of the pointwise fourier spectra is compared to the fourier spectrum of the spatially averaged ellipsometric intensity .
the occurrence of an additional peak at half the dominant frequency can clearly be seen .
the spatial distribution of absolute value and phase of the fourier coefficient corresponding to this additional , subharmonic frequency are shown in figure [ fig : nursh2 ] b ) .
the latter is composed of two values which are about @xmath34 apart , confirming that a 2-phase cluster spreading across the entire electrode area is indeed present .
the two clustered phase domains are separated by ising - type walls , i.e. , by a zero crossing of the absolute value of the fourier coefficient where the phase jump occurs .
this manifests in the black lines in the left hand picture of figure [ fig : nursh2 ] b ) . + + more insight into the details of the cluster dynamics
can be gained by considering the analytical signal of points from the individual phase domains as shown in figure [ fig : nursh3 ] .
b ) at four consecutive extrema of the @xmath30 oscillation shown in figure [ fig : nursh1 ] a ) together with the traces of the analytical signal of @xmath30 ( solid ) and a point located in the center of the blue rectangle in figure [ fig : nursh1 ] ( dash).,scaledwidth=35.0% ] the analytical signal of the points in the two areas indicated in figure [ fig : nursh1 ] are arranged on a bar ( blue and green points , respectively ) whose center follows the analytical signal of @xmath30 ( black solid line ) .
the end points of the bar , i.e. , the points from the individual phase domains , trace a moebius strip projected on the complex plane .
this suggests that the subharmonic clusters result from a period doubling bifurcation of the uniform oscillation .
the four snapshots in figure [ fig : nursh3 ] are taken at consecutive extrema of @xmath30 and the sequence of pictures thus shows that a full rotation of the bar is accompanied by two full rotations of its center . + + the second type of dynamical pattern covering the entire active area observed is a desynchronized oscillation spreading across the entire electrode as shown in figure [ fig : nurturb1 ] .
mm , @xmath25 , @xmath35 , @xmath36 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] here , the remarkable case is shown where the ensemble of pointwise time series of the ellipsometric intensity across the surface is oscillating in a desynchronized manner , while in the spatially averaged signal these differences cancel out almost perfectly leaving a simple periodic oscillation in @xmath30 and @xmath29 .
the corresponding full analytical signal of two sets of points on the electrode further illustrates this behavior .
it is shown in figure [ fig : nurturb2 ] .
b ) at four points during one oscillation period of the @xmath30 oscillation shown in figure [ fig : nurturb1 ] a ) together with the trace of the analytical signal of @xmath30 ( solid).,scaledwidth=35.0% ] the trajectories of the individual points are uncorrelated , both within the two areas shown and between them , leading to broad point clouds circling the origin .
a region of one distinct dynamical behavior is typically not covering the entire electrode .
more often two such regions are observed to coexist on the electrode .
in figure [ fig : hom+sh1 ] an example for the coexistence of a 2-phase cluster with an approximately synchronous oscillation with the frequency of the spatially averaged signal is depicted .
mm , @xmath25 , @xmath37 , @xmath38 mw / cm@xmath20 ) . * a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] the two regions are clearly distinguishable in figure [ fig : hom+sh1 ] b ) with the 2-phase cluster showing a remarkably fine structured border between its two phase domains .
note that the spatially averaged signals still oscillate in a simple periodic manner .
+ + the fourier analysis of this pattern is shown in figure [ fig : hom+sh2 ] . . *
a ) * spatial average of the absolute values of the pointwise fourier spectra ( blue ) and fourier spectrum of the spatially averaged signal @xmath30 ( red ) ; * b ) * spatial distribution of the absolute value ( left ) and phase ( right ) of the fourier coefficient corresponding to the fourier mode indicated by the vertical line in the fourier spectra in a).,scaledwidth=35.0% ] outside the clustered domain the subharmonic frequency is inactive as can be seen in figure [ fig : hom+sh2 ] b ) . as in the experiment where the 2-phase cluster covers the entire electrode shown in figures [ fig : nursh1]-[fig : nursh3 ] , the domain walls are ising - type walls . the analytical signal of points from three areas on the electrode is shown in figure [ fig : hom+sh3 ] .
b ) at four points during two oscillation periods of the @xmath30 oscillation shown in figure [ fig : hom+sh1 ] a ) together with the trace of the analytical signal of @xmath30 ( solid ) .
the dashed line represents points in the ising wall between the two phase domains.,scaledwidth=35.0% ] two main results can be read from the analytical signal in figure [ fig : hom+sh3 ] .
first , the center of the rotating bar formed by the two domains of the 2-phase cluster and their separating wall is now approximately stationary in the origin .
this means that the underlying harmonic frequency of the spatially averaged signal is strongly suppressed in the region of the 2-phase cluster .
second , the trajectory of the points taken from the approximately synchronized region does not follow the trace of the spatially averaged analytical signal perfectly .
this can also be seen in the one dimensional cut in figure [ fig : hom+sh1 ] b ) where the active phase seems to propagate across the surface .
we treat this , however , as a minor variation of a perfectly uniform oscillating region .
actually , the region oscillating with the frequency of the spatially averaged signal in a synchronous manner tends to this wave - like disturbance easily .
+ + an interesting point arises comparing the measurement where the 2-phase cluster covers the entire electrode shown in figures [ fig : nursh1]-[fig : nursh3 ] to the measurement showing coexistence of the 2-phase cluster and synchronized , harmonic oscillation shown in figures [ fig : hom+sh1]-[fig : hom+sh3 ] .
the transition from the former to the latter dynamical state is seemingly achieved by spatially decomposing the two active frequencies .
this is best seen by the corresponding deconvolution of the movement of the bar in the analytical signal shown in figure [ fig : nursh3 ] into a pure rotation of the bar around the origin and a pure center of mass rotation around the origin in the different regions on the electrode as shown in figure [ fig : hom+sh3 ] .
the frequency ratio between the center of mass movement and the rotation of the bar remains unaffectedly 1:0.5 .
a further investigation of this possible deconvolution process would be desirable .
+ + the next case frequently found in the experiments is the coexistence of a synchronized and a desynchronized region on the electrode .
this dynamical state is typically called chimera state and has attracted a lot of research interest recently @xcite . while in most experiments in literature these states have to be initialized in a special way , their spontaneous occurrence in our system is especially interesting @xcite .
an exemplary chimera state is shown in figure [ fig : hom+turb1 ] .
mm , @xmath39 , @xmath40 , @xmath41 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] note that while a region of desynchronized local oscillations can easily be seen in figure [ fig : hom+turb1 ] b ) , the spatial average is again unaffected .
furthermore , the region of synchronized oscillations seems to be less perturbed by the presence of the region of desynchronized oscillations than by the presence of the subharmonic 2-phase cluster as discussed for figure [ fig : hom+sh1 ] .
the coexistence of the two dynamical regions is further emphasized in figure [ fig : hom+turb2 ] where the analytical signal of points from three areas on the surface shows the qualitative difference between the synchronized and the desynchronized oscillations .
b ) at four points during one oscillation period of the @xmath30 oscillation shown in figure [ fig : hom+turb1 ] a ) together with the trace of the analytical signal of @xmath30 ( solid).,scaledwidth=35.0% ] + + another type of coexistence of synchronized and desynchronized patches on the electrode arises in the case of the coexistence of a 2-phase cluster with a desynchronized region as shown in figure [ fig : sh+turb1 ] .
mm , @xmath39 , @xmath42 , @xmath43 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] here , the individual domains in the 2-phase cluster are again synchronized , show a phase difference of @xmath34 and are separated by ising - type walls .
this is further elucidated in the analytical signal obtained from points spreading across the ising - type wall in figure [ fig : sh+turb2 ] ( blue and green points ) .
comparable to figure [ fig : nursh3 ] these points form a bar that rotates at half the angular frequency of the @xmath30 oscillation .
in contrast , the points from the desynchronized region , displayed in red , spread out in a cloud .
b ) at four points during two oscillation periods of the @xmath30 oscillation shown in figure [ fig : sh+turb1 ] a ) together with the trace of the analytical signal of @xmath30 ( solid).,scaledwidth=35.0% ] as this pattern state is again showing the coexistence of synchronized and desynchronized regions one might argue that it could also be called a chimera state. it has to be noted that a slight period-2 perturbation is visible in the oscillation of @xmath30 in figures [ fig : sh+turb1 ] and [ fig : sh+turb2 ] .
this point will be further discussed in section [ subsec : p2 ] .
+ + the coexistence of 2-phase clusters with a synchronous simple periodic oscillation as well as chimera states discussed above have also been observed in two other contexts .
tinsley et al .
observed them experimentally in a network of coupled chemical oscillators and reproduced the dynamics with a reduced two group model @xcite . here
the two groups were obtained by the introduction of a non - local feedback .
furthermore , the uniform oscillation was always stable and thus special initial conditions had to be prepared .
this is in contrast to the work described here as well as to the predictions of the modified complex ginzburg - landau equation where the patterns occur spontaneously in a purely globally coupled system @xcite .
a first deviation from the simple periodic oscillations in @xmath30 is the occurrence of period-2 oscillations in @xmath30 as shown in the middle of figure [ fig : classification ] . in all experiments where
this behavior is found it is accompanied by the formation of a 2-phase cluster region where one phase domain covers a larger area or has a higher amplitude in the subharmonic mode than the other .
an example is shown in figure [ fig : shinav ] .
mm , @xmath25 , @xmath44 , @xmath38 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity ; * c ) * fourier analysis of the pattern shown in a ) .
spatial average of the absolute values of the pointwise fourier spectra ( blue ) and fourier spectrum of the spatially averaged signal @xmath30 ( red ) and the spatial distribution of the absolute value ( left ) and phase ( right ) of the fourier coefficient corresponding to the fourier mode indicated by the vertical line in the fourier spectra.,scaledwidth=35.0% ] in the distribution of amplitude and phase of the fourier coefficient of the subharmonic mode in the lower images of figure [ fig : shinav ] c ) the lack of balance between the phase domains of the 2-phase cluster becomes apparent .
the difference between the unbalanced and the balanced case is visualized in figure [ fig : vergleich ] by an overlay of the amplitude and phase plots of the fourier coefficients of the subharmonic mode for the patterns shown in figures [ fig : shinav ] ( unbalanced ) and [ fig : hom+sh1 ] ( balanced ) , respectively .
while the two phase domains on the left hand side of the unbalanced electrode in figure [ fig : vergleich ] are of similar size and the amplitude of the fourier coefficient in these domains is similar , there is no counterbalancing phase domain for the domain in the upper right corner .
for this reason the phase color coded as purple in figure [ fig : shinav ] c ) is also visible as an amplitude variation in the spatially averaged ellipsometric intensity signal .
whenever the average signal @xmath30 of a patterned electrode oscillates with a period-2 , unbalanced 2-phase clusters are present .
the last type of patterns discussed here occurs accompanied by an irregular oscillation of @xmath30 as shown in the bottom of figure [ fig : classification ] . in this case three
different pattern categories are distinguished .
the first category , shown in figure [ fig : irreg1 ] , can again be seen as a deviation from the case of simple periodic spatial average oscillations .
oscillation ( @xmath45 mm , @xmath25 , @xmath26 , @xmath46 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] here the irregular oscillation of the spatial average is imposed on the entire electrode and the relative intensity of the maxima and minima of @xmath30 in figure [ fig : irreg1 ] a ) can be traced in the synchronized region in the lower part of the one dimensional cut in figure [ fig : irreg1 ] b ) .
a striking similarity between this type of pattern and the patterns forming accompanied by a simple periodic oscillation in @xmath30 is that the regions on the electrode retain their respective dynamical behaviors and sizes until the measurement is finished .
spatially uniform oscillations with an irregular amplitude are also encountered with p - type @xcite and highly illuminated n - type silicon samples , i.e. , in the absence of the nonlinear coupling . for this reason the measurement shown in figure [ fig : irreg1 ] can be seen as a deviation from the chimera state shown in figure [ fig : hom+turb1 ] .
the former is caused by the nonlinear coupling imposed on a spatially uniform oscillation with irregular , most likely chaotic , dynamics , while the latter is caused by the nonlinear coupling acting on a simple periodic , spatially uniform oscillation .
+ + the other two categories of patterns forming during oscillations of @xmath30 with an irregular amplitude are qualitatively different from all other types of patterns considered so far .
the most important difference is that the regions spontaneously forming on the electrode , while still well distinguishable , show dynamical behavior varying with time .
a first example of such spatio - temporal pattern formation is shown in figure [ fig : irreg2 ] .
mm , @xmath47 , @xmath48 , @xmath49 mw / cm@xmath20 ) . *
a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] one important feature the spatio - temporal pattern shown in figure [ fig : irreg2 ] shares with all spatio - temporal patterns shown so far , is that the regions with distinct dynamical behavior do not change their shape as is well visible in the temporal evolution of the cross section in figure [ fig : irreg2 ] b ) .
conversely , the dynamical behavior in the region in the middle of the cross section changes significantly twice on a long time scale from a completely desynchronized oscillation pattern to a nearly synchronized oscillation pattern with a significantly lower amplitude and back .
these profound changes in the overall behavior are also visible in the time series of @xmath30 . + another qualitatively different behavior where the walls of the regions themselves are moving is shown in figure [ fig : irreg3 ] .
mm , @xmath39 , @xmath50 , @xmath43 mw / cm@xmath20 ) . * a ) * time series of @xmath29 and @xmath30 ; * b ) * ellipsometric intensity distribution on the electrode for the two times indicated by the vertical lines in a ) and the temporal evolution of a 1d cut along the line indicated in the left electrode picture .
red indicates a relatively high and blue a relatively low ellipsometric intensity.,scaledwidth=35.0% ] in this example the respective sizes of the two regions distinguishable on the electrode surface change as one region first grows at the expense of the other to completely vanish once it fills up the entire electrode .
it is remarkable in this example how long the memory of the surface seems to be as the period of the growing process is about 30 times larger than the base oscillation period .
in this work , we presented a variety of self - organized spatio - temporal patterns forming in the ellipsometric intensity during the oscillatory photoelectrodissolution of n - type silicon under limited illumination .
remarkably , different regions of distinct dynamical behavior formed spontaneously under spatially uniform experimental parameters which were kept constant in time .
we classified these patterns according to the behavior of the spatially averaged signal of the ellipsometric intensity @xmath30 , the dominant case being a simple periodic oscillation of @xmath30 . in this case , the patterns included synchronized oscillations , subharmonic 2-phase clusters emerging from the uniform oscillation by a period doubling bifurcation and desynchronized oscillations .
any pairwise coexistence of such dynamical regions was found including so - called chimera states. a short overview of the patterns is given in table 1 .
.summary of the patterns and of some important properties shown in this work .
the lines are sorted by the types of patterns encountered in the respective regions and the columns by the behavior of the spatially averaged signal of the ellipsometric intensity @xmath30 .
( sync.=synchonized , 2-ph.cl.=2-phase clusters , desync.=desynchronized , stat.=spatially stationary , reg.=region , cl.=cluster , ph.=phase ) [ cols="^,^,^,^",options="header " , ] the conservation of the simple periodic oscillation of @xmath30 and @xmath29 suggest the presence of a nonlinear global coupling . indeed
, the patterns only occur when the total coupling was found to be dominated by the nonlinear coupling imposed by the limited illumination .
however , whether the coupling induced by the limited illumination of the sample is indeed global , remains an open question at this point .
in addition to the case of simple periodic oscillations in @xmath30 , patterns accompanied by period-2 and irregular oscillations in @xmath30 have been presented . in the former case ,
unbalanced 2-phase clusters were always present on the electrode , linking the period-2 oscillation of @xmath30 to the lack of phase balance .
the latter case opens a wide variety of possible patterns which were classified by the stability of the region walls and their long term behavior but not yet studied in detail .
+ + the presented occurrence of patterns where different dynamical states coexist under completely uniform experimental parameters broadens the spectrum of self - organized patterns in oscillatory media .
so far , mainly one particular state amongst these coexistence states , the chimera state , has received immense interest . here , we demonstrate that this state is embedded in a series of related coexistence patterns , arguably showing a similar level of seemingly contradictory behavior in the coexisting regions .
an example is the coexistence of a cluster and a desynchronized region shown in figure [ fig : sh+turb1 ] .
the wide variety of such patterns in the presence of a simple periodic global oscillation is remarkable .
this conserved oscillation of the spatially averaged quantities seems to strongly promote the spontaneous emergence of the coexistence of different dynamical states on the electrode . in a wider sense , this can be seen as diversification of the dynamical behavior of a system with uniform parameters .
the essential dynamical features responsible for this behavior remain unclear at this point . considering our experiments ,
the assumption of a nonlinear global coupling as a necessary ingredient seems to be justified when considering the minority charge carrier dynamics in the silicon electrode .
the lifetime of these charge carriers is several orders of magnitude shorter than any characteristic time scale of the observed oscillatory dynamics while the diffusion length is comparable to the system size within one order of magnitude .
this should bring about a long - range if not effectively global coupling .
+ in conclusion , the novel type of spatio - temporal patterns presented here sheds light on a number of fundamental questions concerning universal laws of pattern formation in oscillatory media , inviting future studies in both experiment and theory .
we thank our bachelor students martin wiegand and elmar mitterreiter for carrying out some of the experiments shown and iljana miethe and lennart schmidt for fruitful discussions . financial support from dfg ( grant kr1189/12 - 1 ) and the cluster of excellence nanosystems initiative munich ( nim ) is gratefully acknowledged .
10 s. jakubith , h. h. rotermund , w. engel , a. von oertzen , and g. ertl .
spatiotemporal concentration patterns in a surface reaction : propagating and standing waves , rotating spirals , and turbulence .
, 65(24):3013 , 1990 . | we report and classify the rich variety of patterns forming spontaneously in the oxide layer during the oscillatory photoelectrodissolution of n - type doped silicon electrodes under limited illumination .
remarkably , these patterns are often comprised of several dynamical states coexisting on the electrode , such as subharmonic phase clusters and spatio - temporal chaos , and include so - called chimera states. the experiments suggest that the subharmonic phase clusters emerge from a period doubling bifurcation which , upon further parameter changes , evolve into classical phase clusters .
experimentally the occurrence of the patterns is controlled via two coupling mechanisms : a linear global coupling by an external resistor and a nonlinear coupling imposed on the system by the limitation of the illumination . |
binary alloys in stoichiometric compositions invariably exhibit ordered structures at low temperatures .
as we depart from perfect stoichiometric compositions , it is not possible to populate the lattice in the given compositions so as to produce a perfectly ordered structure .
take for example , a b@xmath2a@xmath3 binary alloy on an fcc lattice .
one of the possible stable ordered phases is the l12 arrangement as shown in figure 1 .
an example of this is cu@xmath4au . in a cubic unit cell
the corner is occupied by an a atom , while the three face centres by b atoms .
since there are n corners and 3n face centres ( n being the total number of unit cells in the solid ) and exactly as many a and b atoms , at this composition the l12 ordered arrangement exactly populates all the lattice sites . ordered arrangement becomes impossible , say for b@xmath5a@xmath6 .
however , the following arrangement becomes possible : since there are now 1.2n a atoms , n of them may occupy the n corners .
the 3n face centres may be occupied randomly by the 2.8n b atoms and 0.2n remaining a atoms .
the original a sublattice remains ordered , while the b sublattice becomes disordered .
since there are on the whole 3n face centres , the occupation probability of the a and b atoms in this sublattice are 0.93 and 0.06 respectively .
this arrangement is quite different from the completely random alloy , where all sites are randomly occupied by either the a or b atom with probabilities 0.3 and 0.7 respectively .
it is also rather different from the _ partial disorder _ defined by @xcite . in this communication
we shall define _ partial disorder _ in the manner described above .
= 2.5 in = 4.5 in the recursion method was introduced by @xcite as a convenient and numerically efficient method for calculating green functions and realistic physical properties like the local density of states ( ldos ) , the fermi energy and the band energy .
the method comes to its own in situations where translation symmetry of the potential in an effective one - electron hamiltonian is lost .
the bloch theorem is violated , and we can not employ the standard reciprocal space techniques unless we carry out a homogeneity restoring mean - field approximation ( like the coherent potential approximation , cpa ) .
this can happen , for instance , at a surface or in a random alloy .
in particular , if the ion - cores sit on a topologically distorted network or at a rough surface or interface , where the recursion method seems the only natural way of dealing with electronic property calculations .
there exists a large body of literature on the application of the recursion method to various situations ( @xcite ) . for substitutional disorder ,
the recursion method has been employed in tandem with the augmented space formalism ( @xcite ) ( augmented space recursion or asr ) for dealing with random alloys , taking into account local configuration fluctuations like chemical clustering , local lattice distortions due to size effects and inhomogeneous randomness at a surface because of surface segregation . in spite of this extensive body of literature , the recursion method does not seem to have gained wide acceptability in the electronic structure community , and one often has had to face skepticism regarding its accuracy and feasibility when applied with sophisticated electronic structure techniques to realistic systems
. in this communication we shall first present the asr in detail for the most general case of systems with many atoms per unit cell . our subsequent aim will be to study partially disordered systems in alloys with off - stoichiometric compositions .
we shall carry out recursion calculations based on the tight - binding linearized muffin - tin orbitals ( tb - lmto ) method ( @xcite ) for various metals and alloys , to start with for perfectly ordered systems , compare them with standard reciprocal space methods also based on the tb - lmto and then carry out a thorough analysis of the convergence and accuracy of the technique .
the aim will be to have a convincing handle on the possible errors in the method with a view to minimizing them within our tolerance limits .
having convinced ourselves about the feasibility of the method , we shall first apply it to a toy model for partial disorder before carrying on to realistic systems .
in earlier works ( @xcite ) it has been established that for a disordered system the augmented space theorem maps a disordered hamiltonian described in a hilbert space @xmath7 onto an ordered hamiltonian in an enlarged space @xmath8 , where the space @xmath8 is constructed by augmenting the configuration space @xmath9 of the random variables of the disordered hamiltonian together with the hilbert space @xmath7 of the disordered hamiltonian . thus @xmath10 .
the configuration average of the green function reduces to the evaluation of a particular matrix element of the resolvent of this enlarged hamiltonian in the augmented space . hence ,
if one performs a recursion in the augmented space , one can obtain the matrix element necessary to calculate the configuration average of a green function directly .
the advantage of this method is that it does not involve a single site approximation or the solution of any self - consistent equation ( which is a prerequisite for the cpa or its generalizations ) .
furthermore , one can treat both diagonal and off - diagonal disorder on an equal footing . in spite of
its immense potential , the method could not be implemented in a real alloy systems successfully earlier because of the large rank of the augmented hamiltonian . for a binary alloy with n sites and with only _
s_-orbitals it is @xmath112@xmath12 ) .
however , if we exploit the symmetry of the augmented space ( which arises due to the homogeneity of the disorder ) the rank of the irreducible part of the augmented space in which the recursion is effectively carried out , is considerably reduced and recursion becomes tractable . unlike other methods like the embedded - cluster method , where the configuration averaging is done by explicit averaging over all the distinct configurations of the embedded cluster ( 144 for a fcc lattice with first shell of neighbours ) , this generalized augmented - space recursion technique yields , in a single recursion , the configuration average directly . in a recent paper , @xcite has rederived the augmented space technique and related it to the idea of spatial ergodicity which connects volume averaging to configuration averaging in the infinite solid limit .
as has been mentioned earlier , since the recursion method needs a localized , short - ranged basis for its operation , one can implement augmented space recursion in the framework of the tb - lmto formalism .
we now describe the methodology of generalized augmented space recursion in the framework of the tb - lmto .
the second order tb - lmto hamiltonian is written in terms of potential parameters and screened structure matrix . for a random binary alloy a@xmath13b@xmath14 ,
the lmto hamiltonian in the most localized representation is given by , @xmath15 where ,
+ @xmath16 c , o and @xmath17 are potential parameters of the tb - lmto method , these are diagonal matrices in the angular momentum indices .
also @xmath18 has the dimension of energy and is a measure of the energy window around @xmath19 around which the approximate hamiltonian @xmath20 is reliable .
@xmath21 @xmath22 denotes a _ cell position _ label associated with a tb - lmto basis and @xmath23 = ( @xmath24 ) is the composite angular momentum index .
@xmath25 denotes an atom in the @xmath22-th cell whose position is @xmath26 .
@xmath27 is the site - occupation variable which takes values 0 or 1 depending upon whether site @xmath25 in the @xmath22-th cell is occupied by an a or a b atom . for _ partial disorder
this is a random variable whose probability density depends upon which sublattice it belongs to , hence the label @xmath25 associated with it .
the structure matrix s@xmath28 is non - random in case of substitutional alloys with negligible size mismatch .
now one can obtain the full @xmath20 by inserting @xmath29 in expression ( 1 ) .
@xmath30 and @xmath31 are the projection and transfer operators in hilbert space @xmath7 spanned by tight - binding basis @xmath32 .
@xmath33 the expanded hamiltonian @xmath34 in the augmented space is constructed by replacing the random site occupation variable @xmath35 by its corresponding operator representation @xmath36 in configuration space .
@xmath37 where , @xmath38 is given by @xmath39 now inserting equation(4 ) in equation(1 ) one can obtain the full hamiltonian @xmath34 this hamiltonian is now an operator in a much enlarged space @xmath40 ( the augmented space ) .
the hilbert space @xmath8 is spanned by the basis set @xmath41 , where @xmath42 is a pattern ( @xmath43 ) spanning the configuration space and described in terms of the _ cardinality sequence _ @xmath42 of positions of the @xmath44 in the pattern .
the enlarged hamiltonian does not involve any random variables but incorporates within itself the full statistical information about the random occupation variables . in the presence of off - diagonal disorder , which is invariably present in the form of the tb - lmto hamiltonian , even the reduction of the rank of the invariant subspace on which recursion acts ,
using the symmetries of the augmented space , does not allow us to sample as many configuration states as we would like .
this is because , as recursion proceeds , the number of configuration states sampled at the n - th step of recursion becomes unmanageably large when n @xmath45 5 ( for example ) . to do away with this problem ,
the working equations are transformed so as to put the hamiltonian for the recursion in an effective diagonal disorder form .
this allows one to sample further shells in augmented - space and to confirm the shell - convergence of the recursion . to do this ,
we first suppress all the indeces and write the expression for the resolvent as follows : @xmath46 @xmath47^{-1 } { \mathbf{\tilde \delta}}^{1/2}\ ] ] using the augmented space theorem , we can write the expression of configuration averaged green function as , @xmath48 where , @xmath49 we define @xmath50^{1/2}\vert 1\}$ ] as the normalized initial recursion vector @xmath51 .
finally , we arrive at the convenient expression : @xmath52^{-1}\vert 1\rangle\ ] ] where , @xmath53 @xmath54 here , @xmath55 and @xmath56 , where , @xmath57 where , @xmath58 z is any single site parameter .
though the computational burden is considerably reduced due to diagonal formulation , the recursion now becomes energy dependent as is clear from the expressions of @xmath59 and @xmath60 above .
the old formalism was free of this constraint .
this energy dependence makes the recursion technique rather unsuitable because now we have to carry out one recursion per energy point of interest .
this problem is tackled using _ seed recursion technique _ ( @xcite ) .
the idea is to choose a few seed points across the energy spectrum uniformly , carry out recursion on those points and then fit the coefficients of recursion throughout the whole spectrum .
this way one can save huge computation time . for example , if one is interested in an energy spectrum of 100 points , in the bare diagonal formulation recursion has to be carried out at all the 100 points but in the seed recursion technique one needs to perform recursions only at 15 - 20 points .
the whole idea stems from the fact that in most of the cases of binary alloys , it is seen that the recursion coefficients @xmath61 and @xmath62 vary quite weakly across the energy spectrum .
so , one can easily pick up a few of them and fit throughout the whole range of energy by a suitable function .
before we apply the above methodology to study realistic alloy systems , we must first ensure that the technique is stable and convergent .
we shall illustrate the nature of convergence of various calculated physical quantities which are basic for any electronic structure calculation .
this study of convergence is essential because otherwise we will not be able to set the cut - offs of various parameters like number of shells in real and configuration spaces , recursion levels and so on , for further calculations and ensure that our numerical results are within our tolerance window . before we set out to analyze the errors in the method ,
let us first compare the results for the density of states for cu , cu@xmath4au , v and ni as obtained from a forty - step recursion carried out on an exact twenty shell real - space map .
the choice of the systems is deliberate .
cu is a full @xmath63-shell noble metal , v is a half - filled metal in the lower end of the transition series while ni is a magnetic transition metal .
cu@xmath4au is an ordered alloy which is stable in the l12 arrangement on a fcc lattice .
figures [ fig2 ] and [ fig3 ] show this comparison .
= 5.5 in = 5 in for @xmath64 and @xmath65 most of the detailed features of the density of states calculated from the recursion calculations match with the k - space results .
as expected , the k - space results show sharper structures .
the fermi energies in both cases match to within a few hundredths of a rydberg .
both the majority and minority band density of states for ferromagnetic @xmath66 are reproduced excellently well in the recursion calculations .
the agreement is equally good for @xmath67 upto the fermi energy .
the unoccupied part of the band shows disagreement .
however , it must be understood that the recursion is done with the second order hamiltonian in the most localized representation of the tb - lmto , while the k - space calculations are done with the orthogonal representation .
a higher order recursion calculation should improve the unoccupied part of the band considerably .
= 5.5 in = 5 in how good are our results ? when we talk of convergence of the recursion method , we have to be careful in stating precisely what we mean .
finite space approximants to green functions do not converge for real energy values .
this problem arises in every computational method , as noted by @xcite .
the problem definitely arises in the usual k - space integration techniques , where methods using either complex energies or complex k - s have been attempted .
the cause of this non - convergence is that an arbitrary small perturbation , like adding a single atom to a large but finite system , can shift all eigenvalues of the system .
this causes an infinite change in the green function near its corresponding poles .
thus , the precise meaning of the convergence of the recursion should imply rather the convergence of physical quantities built out of it .
most physical quantities are averages over the spectrum of the type : @xmath68 @xmath69 is the lower band edge , and @xmath70 is any smooth , well - behaved function of @xmath71 .
it is the convergence of these quantities which will decide whether the recursion is convergent or not .
for example , the fermi energy is defined by @xmath72 where @xmath73 is the total number of electrons . while the band energy is @xmath74 we shall study , in general , the convergence of indefinite integrals of the kind @xmath75 the integrand @xmath76 is monotonic and well behaved within the integration range .
a measure of the root - mean - square error in the moments is @xmath77 errors can arise in the recursion procedure because of two distinct sources : ( i ) the error that arises because we carry out a finite number of recursion steps and then terminate the continued fraction using one of the available terminators ; ( ii ) the error that arises because we choose a large but finite part of the nearest - neighbour map and ignore the part of the augmented space very _ far _ from the starting state .
haydock has justified both these approximations by stating that ( i ) if the terminator is so chosen so as to reflect the asymptotic behaviour of the continued fraction , errors should be small , and ( ii ) since we can write expressions for the continued fraction in terms of self - avoiding _ walks _ in augmented space , long walks are dominated by those that wind round the starting state and do not go far away from it .
we shall first carry out a simple error analysis of the continued fraction expression for the green function because of errors created on the continued fraction coefficients .
the recursion is a two - term recurrence relation .
we may therefore generate from this , a pair of linearly independent set of polynomials through the relations : @xmath78 where , @xmath79 is either @xmath80 or @xmath81 according to the initial conditions : + [ cols="^,^,^,^,^,^ " , ] the approximated green function in terms of the terminator @xmath82 is given by : @xmath83}\ ] ] the terminator determines entirely the essential singularities of the the spectrum .
@xcite showed that a finite composition of fractional linear transformations like the one above can at most add a finite number of poles to the spectrum .
the essential singularities of the exact @xmath84 and @xmath82 coincide .
the fractional linear transformation redistributes the spectral weights over the spectrum .
let us now assume that we make errors @xmath85 in the corresponding continued fraction coefficients . if we now start generating the orthogonal polynomials , starting from the exact initial conditions , but with the errors in the continued fraction coefficients , we shall obtain a pair of sets @xmath86 and @xmath87 . in general we shall have , @xmath88 if we substitute this back into the recurrence relation and keep only the first order terms in the errors , = 5.5 in = 5 in @xmath89\right.\nonumber\\ & & \left .
+ \delta b_{n } \left [ p_{n } ( e)q_{n+1 } ( e)+ p_{n+1 } ( e)q_{n } ( e)\right]\right\}/b_{1}\nonumber\\ b_{n } ( e ) & \eq & \left\ { - \delta a_{n } p_{n+1 } ( e)^{2 } \mns \delta b_{n } \left [ 2\ ; p_{n } ( e)p_{n+1 } ( e)\right]\right\}/b_{1 } \nonumber\\\end{aligned}\ ] ] using the above and the expression for the local density of states , we find that the first order relative error produced in the local density of states @xmath90\ ] ] where @xmath91 @xmath92e @xmath84 .
if we define the weighted hilbert transforms of @xmath93 as the so - called associated functions :
@xmath94 these associated functions are also solutions of the three - term recursion .
they are not polynomials , but are nevertheless orthogonal to the set @xmath80 . in terms of these ,
the error in the density of states is : @xmath95 \right\}\nonumber\\\end{aligned}\ ] ] if the first n continued fraction coefficients are exact , that is in case we carry out our recursion on a n - shell neighbour map upto n steps and then terminate , the error in the various moment functions are : @xmath96 where , = 5.5 in = 5 in @xmath97 from this expression and equation(6 ) we can obtain an expression for the overall error in the moments .
numerical results for the errors in the moment functions are shown in figure [ fig4 ] .
the convergence in fermi and band energies with number of recursion steps are shown in figures [ fig5 ] and [ fig6 ] .
= 5.5 in = 5 in = 4.5 in = 4 in these are calculated on a 20 shell neighbour map and upto 40 steps in recursion .
the figures clearly illustrate the convergence of the procedure with increasing recursion steps , as surmised by haydock .
as expected the higher moments converge more rapidly .
the rapid convergence of higher moments is at the basis of the reproduction of most of the density of states shape in an approximate recursion procedure .
it should be noted , however , that we carry out recursion beyond 40 steps , very soon the procedure becomes unstable .
this instability arises due to two sources : ( i ) we have carried out recursion on a 20 shell neighbour - map . as we go beyond 20 steps of recursions , finite size effects begin to show up .
numerically till 40 steps these errors are tolerable .
beyond 45 they lead to instability .
we can control this by increasing the size of the neighbour - map .
( ii ) numerical cumulative errors lead to the loss of orthogonality in the recursion generated basis .
this can be controlled by deliberately orthogonalizing after every 20 steps or so . in the presence of disorder
, the sharp structures in the density of states are smoothened by finite life - time effects.the convergence of the higher moments is even more rapid .
the above analysis in ordered systems is therefore in a _
worst scenario _ situation .
figure 7 shows the convergence of the magnetic moment of ni as a function of the number of recursion steps towards the value 0.6 @xmath98 obtained from @xmath0-space techniques .
magnetic moment is a physical quantity which is very sensitive to the errors in numerical approximations .
the figure shows that within 35 - 40 recursion steps on a 20 shell real - space map , the magnetic moment has converged within our error tolerance .
it is evident from the above analysis that the size of the neighbour map and the number of recursion steps required can vary from system to system .
for every situation , we have to carry out this error analysis before we can rely on our numerical results with a degree of satisfaction .
before we carry out calculations on a real alloy , let us first apply our formalism to a toy model in order to understand the effects of partial disordering .
we shall consider a 50 - 50 ab alloy ordered first on a square lattice as shown in figure [ toy1 ] and only @xmath99-states on this atomic arrangement .
the hamiltonian in a tight - binding basis set is then , = 2 in = 2 in @xmath100 where @xmath101 denotes that i and j are nearest neighbour cells on the lattice , @xmath25 is 1 or 2 according to whether we a referring to the corner atoms or central atoms in a square unit cell ( dark and light atoms in figure [ toy1 ] ) .
if we consider only the nearest neighbour overlaps at a distance @xmath102 where @xmath103 is the square lattice constant , the diagonal and off - diagonal terms of the hamiltonian are ( referring to figure [ toy2 ] ) @xmath104 = 2.5 in = 2.0 in the ordered lattice has a density of states which has a central band gap with integrable infinite van hove singularities at the two internal band edges , two kink singularities within the band and the usual square - root singularities at the external band edges .
as expected , the density of states is symmetric about the band centre @xmath105 = 0.5 .
this is shown in figure [ toy0 ] .
= 4 in = 5 in the density of states for the perfectly disordered lattice , where each lattice site is occupied by either a or b atom with probabilities proportional to their concentrations , is shown in the figure [ toy3 ] .
disorder washes away the van hove singularities and the central band gap is filled up .
the results are identical to a disordered square lattice with lattice vectors ( @xmath102,0 ) and ( 0,@xmath102 ) .
= 6 in = 4.5 in the next figure [ toy4 ] shows the density of states and fermi energies for the partially disordered half - filled alloys at just off - stoichiometric compositions ( 49.5 - 50.5 ) and ( 50.5 - 49.5 ) .
disorder washes out the van hove singularities , although vestiges of the kink singularities remain .
the signature of the internal infinite singularities show up as peaks , but disorder fills up the internal band gap .
the band - edge square - root singularities remain as artifacts of the termination procedure .
loss of stoichiometry weights the two ` bands ' differently leading to a loss of symmetry about the band centre .
= 5.5 in = 5 in the figure [ toy5 ] shows the band energy for the partially ordered alloys . there is a jump at the stoichiometric composition with the band energy for the ordered alloy falling half way between the two branches .
this jump can be understood by carefully examining the graphs on the left panels and right panels of the figure [ toy5 ] .
the band energy represents the average energy ( centre of gravity ) of the portion of the graph to the left of the fermi - energy .
the density of states on the left panel have greater weight below the fermi energy as compared to those on the right panel , ensuring the band energy to be more negative for the compositions sown on the left panels .
the sudden jump is the result of the fermi energy shifting across the pseudo band - gap as we cross the 50 - 50 composition and including the high peak at e=0 , which shifts the average energy to higher values .
the specific behaviour of the band energy depends sensitively on the features of the density of states . the specific behaviour for our toy model may not obtain for realistic alloy systems .
in contrast , the behaviour of the band energy with composition for the fully disordered alloy is smooth , reflecting the smooth behaviour of the density of states without any internal band gaps .
the figure indicates that for compositions to the left of the 50 - 50 stoichiometric one , the alloy prefers to be partially ordered . at 50 -
50 the ordered alloy is stable for our toy model . while for compositions to the right of 50 - 50 the alloy stabilizes in the completely disordered phase .
= 6 in = 5 in
in this communication we have first proposed a generalization of the asr for partially ordered binary alloy systems .
partial ordering of the type where disorder in different sublattices are different has been studied in particular .
we have analyzed the convergence of the recursion technique in the worst scenario case of no disorder and have argued that disorder smoothens out structure in the density of states , so that for disordered alloys the convergence of the moments is even faster .
finally we have studied a model system , in order to understand the effects of partial disordering .
we are now in a position to apply our formalism to realistic alloy systems . this will be the subject matter for a subsequent communication .
99 andersen o.k . in _ computational methods in band theory _ , eds .
p.m. marcus , j.f .
janak and a.r . williams ( plenum , new york , 1971 ) , p.178 .
andersen o.k . and jepsen o. , 2571 ( 1984 ) andersen o.k . , jepsen o. and krier g. in _ lectures on methods of electronic structure calculations _ , eds . v. kumar , o.k .
andersen and a. mookerjee ( world scientific , singapore , 1994 ) , p.63 .
ghosh s. , das n. , mookerjee a. , 10701(1997 ) haydock r. 2120 ( 1972 ) haydock r. , heine v. and kelly m. j. , 2845 ( 1972 ) haydock r. , heine v. and kelly m. j. , _ surf .
* 38 * l39 ( 1973 ) haydock r. _ solid state physics _
( academic press , new york ) * 35 * ( 1988 ) arnold w. t. , haydock r. ( private communication ) ( 2001 ) kudrnovsk . , bose s. k. , anderson o. k. , 4613 ( 1991 ) mookerjee a. , l205 ( 1973 ) mookerjee a. , 1973 1340 ( 1973 ) saha t. , dasgupta i. and mookerjee a. , l245 ( 1994 ) saha t. , dasgupta i. and mookerjee a. , 1979 ( 1996 )
saha t. and mookerjee a. , 2179 ( 1997 ) | off - stoichiometric alloys exhibit partial disorder , in the sense that only some of the sublattices of the stoichiometric ordered alloy become disordered .
this paper puts forward a generalization of the augmented space recursion ( asr ) introduced earlier by one of us ( mookerjee 1997 ( * ) ) for systems with many atoms per unit cell . in order to justify the convergence properties of asr
we have studied the convergence of various moments of local density of states and other physical quantities like fermi energy and band energy .
we have also looked at the convergence of the magnetic moment of ni , which is very sensitive to numerical approximations towards the @xmath0-space value 0.6 @xmath1 with the number of recursion steps prior to termination . -0.5 |
in 2014 costantino , geer and patureau introduced a new class of non - semisimple witten - reshetikhin - turaev type quantum invariants of closed 3-manifolds .
the theory is indexed on some rather complicated algebraic structures : if @xmath1 and @xmath2 are abelian groups called the _ structure group _ and the _ periodicity group _ respectively and if @xmath3 is a `` small '' subset of @xmath1 called the _
critical set _ then a very general construction provides an invariant of decorated closed 3-manifolds for every _ pre - modular @xmath1-category @xmath4 relative to @xmath5 _ which satisfies some non - degeneracy condition .
such relative pre - modular categories , introduced here in definition [ relative_pre - modular_graded_categories_def ] , are ribbon categories with three important new features : they carry a @xmath1-structure , they have finiteness properties only up to the action of @xmath2 and , more importantly , they are not necessarily semisimple , with the deviation from semisimplicity being measured by @xmath6 .
the costantino - geer - patureau invariants have to be evaluated against closed 3-manifolds @xmath7 equipped with @xmath4-colored ribbon graphs @xmath8 and cohomology classes @xmath9 with @xmath1-coefficients , but not arbitrary ones : @xmath8 and @xmath9 should satisfy a certain compatibility condition and , more importantly , a certain admissibility condition and [ admissibility ] ] , the lack of which makes it impossible to even lay down the definition .
the construction extends the multivariable alexander polynomial , the kashaev invariant and the akutsu - deguchi - othsuki invariant to framed colored links in arbitrary 3-manifolds . a small survey of these results is provided by @xcite .
in @xcite blanchet , costantino , geer and patureau extended these quantum invariants to @xmath10-graded tqfts in the case of a non - degenerate relative pre - modular category @xmath11 of finite - dimensional complex - weight representations of the so called _ unrolled _ version of quantum @xmath0 at a root of unity .
this category features @xmath12 as its structure group with critical set @xmath13 , and the periodicity group @xmath14 appears as the grading group for vector spaces .
when the order of @xmath15 is a multiple of 4 then the braiding on @xmath10-graded vector spaces is the super - symmetric one .
the representations for mapping class groups of surfaces issued from these graded tqfts feature some new interesting phenomena , as the actions of dehn twists along separating curves have infinite order .
it seems however reasonable that the construction should generalize to a larger class of non - degenerate relative pre - modular categories .
this brings us to the goal of the present paper : we want to find conditions for a non - degenerate pre - modular @xmath1-category @xmath4 relative to @xmath5 under which its associated costantino - geer - patureau quantum invariant can be extended to a @xmath2-graded etqft .
moreover , since we are inspired by the construction provided in @xcite , we also want our result to agree with the @xmath10-graded tqfts defined for @xmath16 .
it turns out that the only additional requirement we have to make for @xmath4 in order to carry out the whole plan is actually fairly simple : we call it _ relative modularity _ , and it takes the form of a non - degeneracy condition , which is given here in definition [ relative_modular_graded_categories_def ] , that is just slightly stronger than the one we need in order to define costantino - geer - patureau invariants . since in particular this property is established for @xmath16 in lemma a.4 of @xcite , our construction extends the blanchet - costantino - geer - patureau @xmath10-graded tqfts down to 1-dimensional manifolds .
our etqfts will then be symmetric monoidal functors defined on symmetric monoidal 2-categories @xmath17 of decorated cobordisms satisfying some admissibility conditions , and taking values in symmetric monoidal 2-categories @xmath18 of complete @xmath2-graded linear categories .
we state here our main result .
[ main ] if @xmath4 is a modular @xmath1-category relative to @xmath5 then @xmath19 extends to a @xmath2-graded etqft @xmath20 the functor @xmath21 can actually be described in terms of the relative modular category @xmath4 which is used as a building block : connected objects of @xmath17 are mapped to complete @xmath2-graded linear categories which are @xmath2-graded extensions of homogeneous subcategories of projective objects of @xmath4 , and arbitrary objects are mapped to tensor products of these categories .
furthermore , generating 1-morphisms of @xmath17 can be translated into @xmath2-graded linear functors between these @xmath2-graded linear categories , and they can be interpreted as algebraic structures on them . this description presents us immediately with a new phenomenon : the functor @xmath21 is fully monoidal , but the images of certain objects of the domain 2-category are in general non - semisimple .
this result , which apparently contradicts theorem 3 of @xcite , can in fact be explained as follows : in the definition of the admissible cobordism 2-category @xmath17 we forbid some non - admissible decorated cobordisms by removing them from the categories of morphisms .
this makes some `` critical '' objects into non - dualizable ones , thus making room for non - semisimplicity .
we claimed that the symmetric monoidal functor of 2-categories @xmath21 extends @xmath19 , but what does this mean ?
the idea comes from the analogy with the following well - known property : every 3-dimensional tqft yields a quantum invariant of closed 3-dimensional manifolds . here is a recipe to read it off directly .
a closed 3-dimensional manifold can be interpreted in a unique way as a cobordism between two copies of the empty surface .
therefore a tqft associates with it a linear endomorphism of a certain vector space which , as a consequence of monoidality , is isomorphic to @xmath22 , the unit for the tensor product on vector spaces .
such a map is just a product with some fixed complex number .
since morphisms of a cobordism category are given by isomorphism classes of cobordisms , this complex number actually depends only on the diffeomorphism class of the closed 3-manifold , so we can intepret it as a quantum invariant . to compute it ,
it is sufficient to evaluate the linear endomorphism against a generator of the vector space associated with the empty surface .
the situation directly generalizes to the level of 2-categories , so that every 3-dimensional etqft yields a 3-dimensional tqft .
the procedure to recover it is completely analogous .
a closed surface can be interpreted in a unique way as a cobordism between two copies of the empty 1-dimensional manifold .
therefore a tqft associates with it a linear endofunctor of a certain linear category which , as a consequence of monoidality , is equivalent to @xmath23 , which is in turn equivalent to the unit for the complete tensor product of complete linear categories .
such a functor is , up to equivalence of categories , just a tensor product with some fixed vector space , which we can interpret as a tqft vector space . to compute it
, it is sufficient to choose a specified linear equivalence of categories , to translate the linear functor associated with the closed surface into a linear endofunctor of @xmath23 and to evaluate it against the 1-dimensional vector space @xmath22 , which is a generator under direct sum .
everything we said extends word by word to the @xmath2-graded case : every 3-dimensional @xmath2-graded tqft yields a quantum invariant of closed 3-dimensional manifolds and every 3-dimensional @xmath2-graded etqft yields a 3-dimensional @xmath2-graded tqft .
in particular , hidden behind the higher structure of the symmetric monoidal functor @xmath21 , we still have a recipe which associates a complex number with every admissible decorated closed 3-manifold and a @xmath2-graded vector space with every admissible decorated closed surface .
this recovers the quantum invariants of @xcite and generalizes the tqfts of @xcite to all relative modular categories .
the definition of the witten - reshetikhin - turaev quantum invariants of 3-manifolds contained in @xcite made use of surgery presentations which where colored with finite dimensional representations of the restricted version of quantum @xmath0 at a root of unity .
a famous result by turaev which first appeared in @xcite generalized this construction by characterizing its building blocks : non - degenerate pre - modular categories .
this allowed him to associate with every such category @xmath4 a quantum invariant . ]
@xmath24 of decorated closed 3-manifolds , where this time surgery presentations are colored with objects of the indexing category rather than representations of quantum @xmath0 .
it has been known for some time that these quantum invariants can be extended etqfts every time the category @xmath4 is modular .
one possible way to construct these semisimple etqfts is to plug @xmath24 into a machinery called the _ extended universal construction _
, which is derived from @xcite , to obtain a functor of 2-categories @xmath25 here @xmath26 is a symmetric monidal 2-category of @xmath4-decorated cobordisms of topmost dimension 3 , @xmath27 is the symmetric monoidal 2-category of complete linear categories and @xmath28 is a functor called the _ covariant extended quantization functor associated with @xmath24_. it can then be shown that @xmath28 is symmetric monoidal if , and in fact only if , @xmath4 is modular .
the idea of this paper is to try and adapt the same procedure to the non - semisimple setting of costantino , geer and patureau .
their construction , which again is based on surgery presentations , provides a quantum invariant . ]
@xmath19 of admissible decorated closed 3-manifolds for every non - degenerate pre - modular @xmath1-category @xmath4 relative to @xmath5 . by means of the extended universal construction
we can extend @xmath19 to a functor of 2-categories @xmath29 however , unlike in the semisimple case , @xmath28 is not an etqft on the nose , and its deviation from monoidality can be parametrized using the periodicity group @xmath2 .
nevertheless , we can define a @xmath2-graded extension @xmath30 of @xmath28 with target the symmetric monoidal 2-category of complete @xmath2-graded linear categories .
the idea is to replace the image of an object of @xmath17 under @xmath28 , which is a complete linear category , with a suitably defined complete @xmath2-graded linear category which integrates the obstruction to monoidality into its structure .
we can then show that the resulting functor @xmath21 is indeed symmetric monoidal , and thus provides a @xmath2-graded etqft .
the paper is organized as follows : we begin by introducing the main ingredients of our construction , with relative modular categories being defined in section [ relative_modular_graded_categories ] and admissible cobordisms being defined in section [ admissible_decorated_cobordisms ] . in sections from [ seventh_section_graded ] to [ graded_morita_reduction ] we state surgery axioms and we derive their consequences on the extended quantization functors obtained via the extended universal construction , much in the spirit of @xcite . in section [ gradation_section ]
we prove the failure of @xmath28 to be an etqft , we carefully analyse its deviation from monoidality and we define the @xmath2-graded extended quantization functors @xmath21 and @xmath31 . our main results are contained in sections [ monoidality_section ] and [ simmetry_section ] , where we show that @xmath21 is indeed a @xmath2-graded etqft .
sections [ image_of_objects_section ] and [ image_of_1-morphisms_section ] are devoted to a description of the image of @xmath21 in terms of the corresponding relative modular category @xmath4 .
appendix [ cobordisms ] and appendix [ ribbon_graphs ] fix the notation we use for cobordisms and ribbon graphs , appendix [ linear_categories ] and appendix [ graded_linear_categories ] contain definitions and properties of complete linear categories and of complete @xmath2-graded linear categories and appendix [ extended_universal_construction ] contains the details for the extended universal construction .
this section is devoted to the definition of the algebraic structures which are going to play the leading role in the construction .
they are a non - semisimple analogue of modular categories and , since their definition involves a lot of different ingredients , we first set the ground by recalling classical concepts like ribbon categories , group structures , colored ribbon graphs , traces on ideals and ambidextrous objects .
let us begin by recalling the very basic notions for the whole construction .
pivotal category _ is the given of : 1 . a strict monoidal category @xmath4 ; 2 .
a monoidal functor is a monoidal category then the standard monoidal structure on @xmath32 is determined by the opposite tensor product @xmath33 where @xmath34 denotes the standard braiding functor of categories @xmath4 and @xmath35 . ]
@xmath36 called the _
duality functor _ ; 3 .
a pair of dinatural transformations is a monoidal category then the constant functor @xmath37 maps every object to @xmath38 and every morphism to @xmath39 . ]
@xmath40 called the _ left evaluation _ and the _ left coevaluation _ respectively , satisfying @xmath41 for every object @xmath42 ; 4 . a monoidal natural isomorphism @xmath43 .
our notation for pivotal categories , and more in general for all categories equipped with some specified structures , will be slightly abusive , as we will always drop all references to extra data and refer only to the underlying categories .
if @xmath4 is a pivotal category then the image of an object @xmath44 under the duality functor @xmath45 will be denoted by @xmath46 and the image of a morphism @xmath47 will be denoted by @xmath48 .
if @xmath4 is a pivotal category we define the _ right evaluation _ and the _ right coevaluation _ to be the dinatural transformations @xmath49 given by @xmath50 if @xmath51 is a braiding on a pivotal category @xmath4 then we define the _ twist of @xmath4 _ to be the natural isomorphism @xmath52 associating with every object @xmath42 the morphism @xmath53 the twist satisfies @xmath54 if @xmath4 is a pivotal category then a braiding @xmath51 on @xmath4 is _ compatible with the pivotal structure _ if the twist @xmath55 satifies @xmath56 a _ ribbon category _ is then a pivotal category together with a compatible braiding .
we move on now to recall the definition of group structures on monoidal categories , and in order to do so we fix an abelian group @xmath1 . if @xmath4 is a strict monoidal category a _
@xmath1-structure on @xmath4 _ is the given of a family of full subcategories @xmath57 of @xmath4 indicized by elements of @xmath1 and satisfying @xmath58 for all @xmath59 and all @xmath60 .
a strict monoidal category @xmath4 equipped with a @xmath1-structure is called a _
@xmath1-category _ and @xmath1 is called the _ _ structure group__-structure is usually called a @xmath1-graded category , see @xcite , @xcite and @xcite .
we change terminology here because we will use the term graded category in the enriched sense , with grading appearing on morphisms rather than on objects . ] . for every @xmath61 the subcategory @xmath62
is called the _ homogeneous subcategory of index @xmath63_. if an object @xmath64 of @xmath4 belongs to the homogeneous subcategory of index @xmath63 then we will write @xmath65 and we will say @xmath64 is a
_ homogeneous object of index @xmath63_. we say a @xmath1-structure on a pivotal category @xmath4 is _ compatible with the pivotal structure _ if @xmath66 for all @xmath59 .
pivotal @xmath1-category _ is then a pivotal category equipped with a compatible @xmath1-structure , and a _
ribbon @xmath1-category _ is a pivotal @xmath1-category which is ribbon . in this subsection
we give a definition for the category of @xmath4-colored ribbon graphs . ] and for the associated reshetikhin - turaev functor which is adapted to the case of categories with group structure .
let us fix for this subsection a ribbon @xmath1-category @xmath4 . if @xmath67 is a 2-dimensional cobordism and if @xmath68 is a ribbon set then a _ @xmath1-homogeneous @xmath4-coloring of @xmath69 _ , denoted @xmath70 , is the given of a homogeneous object @xmath71 of @xmath4 called the _ color of @xmath72 _ for each vertex @xmath73 .
@xmath1-homogeneous @xmath4-colored ribbon set @xmath74 _ is the given of a ribbon set @xmath68 together with a @xmath1-homogeneous @xmath4-coloring @xmath70 .
if @xmath67 and @xmath75 are 2-dimensional cobordisms and if @xmath74 is a @xmath1-homogeneous @xmath4-colored ribbon set then every time we have a positive embedding of surfaces @xmath76 we obtain a @xmath4-colored ribbon set @xmath77 given by @xmath78 with color @xmath71 on @xmath79 for every vertex @xmath73 .
if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath74 and @xmath81 are @xmath1-homogeneous @xmath4-colored ribbon sets and if @xmath82 is a ribbon graph from @xmath69 to @xmath83 then a _ @xmath1-homogeneous @xmath4-coloring of @xmath84 extending @xmath64 and @xmath85 _ , denoted @xmath86 , is the given of : 1 . a homogeneous object @xmath87 of @xmath4 called the _ color of @xmath88 _ for each edge @xmath89 ; 2 . a morphism @xmath90 of @xmath4 called the _ color of @xmath91 _ for each coupon @xmath92 .
these data satisfy : 1 . the color @xmath87 of an edge @xmath89 intersecting @xmath93 along a vertex @xmath94 equals @xmath71 ; 2 .
the color @xmath87 of an edge @xmath89 intersecting @xmath95 along a vertex @xmath96 equals @xmath97 ; 3 .
the color @xmath90 of a coupon @xmath92 of type @xmath98 belongs to @xmath99 where @xmath100 is the edge of @xmath101 intersecting @xmath102 along its @xmath103-th input point , where @xmath104 is the edge of @xmath101 intersecting @xmath105 along its @xmath106-th output point , where @xmath107 is the sign given by the orientation of the @xmath103-th input point of @xmath108 , where @xmath109 is the sign given by the orientation of the @xmath106-th output point of @xmath108 and where @xmath110 and @xmath111 for all @xmath42 .
@xmath1-homogeneous @xmath4-colored ribbon graph @xmath112 from @xmath113 to @xmath114 _ is the given of a ribbon graph @xmath82 from @xmath69 to @xmath83 together with a @xmath1-homogeneous @xmath4-coloring @xmath86 extending @xmath64 and @xmath85 .
for @xmath1-homogeneous @xmath4-colorings of ribbon tangles we will sometimes use the same notation @xmath64 we use for @xmath1-homogeneous @xmath4-colorings of ribbon sets instead of @xmath115 , in order to stress the absence of coupons .
if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath116 , if @xmath117 is a 3-dimensional cobordism with corners from @xmath75 to @xmath118 and if @xmath112 is a @xmath1-homogeneous @xmath4-colored ribbon graph from @xmath74 to @xmath81 then every time we have positive embeddings of surfaces @xmath76 and @xmath119 and a positive embedding of manifolds with faces @xmath120 satisfying @xmath121 we obtain a @xmath4-colored ribbon graph @xmath122 from @xmath123 to @xmath124 given by @xmath125 with color @xmath87 on @xmath126 for every edge @xmath89 and with color @xmath90 on @xmath127 for every coupon @xmath92 .
every @xmath1-homogeneous @xmath4-coloring @xmath70 of a ribbon set @xmath68 determines a @xmath1-homogeneous @xmath4-colored ribbon tangle @xmath128 .
we can now begin to define the _ ribbon @xmath1-category @xmath129 of @xmath1-homogeneous @xmath4-colored ribbon graphs _ : an _ object @xmath130 of @xmath129 _ is a finite sequence @xmath131 where @xmath107 is an element of @xmath132 and @xmath133 is a homogeneous object of @xmath4 for all @xmath134 . if @xmath135 is an object of @xmath136 then we denote with @xmath137 the object of @xmath4 given by @xmath138 let @xmath139 denote the 2-dimensional cobordism from @xmath140 to @xmath141 given by . ]
@xmath142 where the collar @xmath143 is defined as @xmath144 for the embedding @xmath145 specified in remark [ trivial_cobordism ] . then with every object @xmath146 of
@xmath129 we can associate the standard @xmath1-homogeneous @xmath4-colored ribbon set @xmath147 inside @xmath139 whose oriented vertex set is given by @xmath148 with orientation specified by the sequence of signs @xmath149 , with framing tangent to @xmath150 and given by the positive unit vector at positive vertices and by the negative unit vector at negative vertices and with @xmath1-homogeneous @xmath4-coloring @xmath151 given by @xmath152 for all @xmath134 .
a _ morphism @xmath153 of @xmath129 _ is an equivalence class of @xmath1-homogeneous @xmath4-colored ribbon graphs inside @xmath154 from @xmath147 to @xmath155 .
two @xmath1-homogeneous @xmath4-colored ribbon graphs from @xmath147 to @xmath155 are equivalent if their oriented graphs are isotopic relative to @xmath156 and @xmath157 through an isotopy which preserves framings via pullback and colorings .
the notation for morphisms of @xmath129 will be a little abusive since we will not distinguish between an isotopy class of @xmath1-homogeneous @xmath4-colored ribbon graphs and any of its representatives .
morphisms of @xmath129 will simply be represented by oriented black diagrams in @xmath158 . in particular we will always implicitly assume the framing to be the blackboard one , and coupons will always be depicted with their horizontal faces oriented from left to right and with their vertical faces oriented from bottom to top .
when depicting ribbon graphs inside general cobordism with corners we will represent them in blue and we will be more careful , always specifying framing of ribbon sets and of edges and orientations of coupons .
we can always suppose that every morphism @xmath8 from @xmath130 to @xmath159 is represented , up to isotopy , by a vertical graph in some neighborhood of @xmath160 .
more precisely we can suppose that @xmath161 for some neighborhood @xmath162 of @xmath163 .
let us now describe the structure of the ribbon @xmath1-category @xmath129 : if we consider morphisms @xmath153 and @xmath164 in @xmath129 then the vertical composition of @xmath165 with itself determines a @xmath1-homogeneous @xmath4-colored ribbon graph inside @xmath166 which we denote @xmath167 .
composition @xmath168 of @xmath169 and @xmath8 _ is then defined as @xmath170 for the isomorphism @xmath171 & \mapsto & \begin{cases } \left ( ( x , y),\frac{t}{2 } \right ) & i = 0 \\ \left ( ( x , y),\frac{t+1}{2 } \right ) & i = 1 \end{cases } \end{array}\ ] ] the _ tensor product @xmath172 of objects @xmath130 and @xmath159 of @xmath129 _ is given by @xmath173 the _ tensor product @xmath174 of morphisms @xmath8 and @xmath169 of @xmath129 _ is given by @xmath175 for the embeddings @xmath176 the _ dual @xmath177 of an object @xmath178 _ is @xmath179 where @xmath180 and @xmath181 .
the _ braiding _ , the _ twist _ , the _ left evaluation _ and _ left coevaluation of @xmath129 _ are depicted in figure [ structure ] .
[ structure ] .,title="fig : " ] from now on the term @xmath1-homogeneous @xmath4-colored ribbon graph , if used without any reference to the ambient cobordism with corners , will denote morphisms of @xmath129 .
an _ elementary morphism @xmath182 of @xmath129 _ is a morphism of @xmath129 which features a single coupon of color @xmath183 and whose oriented graph is contaned in @xmath158 .
if @xmath4 is a ribbon @xmath1-category then there exists a unique braided monoidal functor @xmath184 mapping the object @xmath185 to @xmath64 , the object @xmath186 to @xmath46 , an elementary morphism to the color of its coupon and the braiding , the twist and the left evaluation and the left coevaluation of @xmath129 to those of @xmath4 .
the functor @xmath187 is the _ reshetikhin - turaev functor associated with @xmath4_. the reshetikhin - turaev functor @xmath187 yields in particular an invariant of framed oriented links colored with homogeneous objects of @xmath4 . in this subsection
we recall the notions of traces on ideals in ribbon linear categories and of ambidextrous objects which are central in the theory of renormalized invariant of links developed by geer , patureau and turaev in @xcite .
we say an object @xmath64 of a category @xmath4 is a _
retract _ of another object @xmath85 of @xmath4 if there exist morphisms @xmath188 and @xmath189 satisfying @xmath190 , and a subcategory @xmath35 of @xmath4 is said to be _ closed under retraction _ if every retract of every object of @xmath35 is an object of @xmath35 . moreover a subcategory @xmath35 of a monoidal category @xmath4 is said to be _ absorbent _ if for all objects @xmath64 of @xmath4 and @xmath85 of @xmath35 the tensor products @xmath191 and @xmath192 are objects of @xmath35 .
an _ ideal _
@xmath193 of a monoidal category @xmath4 is then a full subcategory of @xmath4 which is absorbent and closed under retraction .
the closure under retraction implies that every ideal @xmath193 in @xmath4 is replete , meaning that if some object @xmath64 of @xmath4 is isomorphic to some object of @xmath193 then @xmath64 is also an object of @xmath193 .
[ ideal_of_duals ] if @xmath4 is a pivotal category , if @xmath193 is an ideal of @xmath4 and if @xmath194 then @xmath195 .
the _ ideal @xmath196 generated by an object @xmath64 of a pivotal category @xmath4 _ is the smallest among the ideals @xmath193 of @xmath4 satisfying @xmath194 . if @xmath4 is a pivotal category then full subcategory @xmath197 of projective objects of @xmath4 is an ideal of @xmath4 , and it coincides with the full subcategory @xmath198 of injective objects
moreover if @xmath64 is a projective object of @xmath4 with epic evaluation then @xmath199 .
can be found in @xcite ] we will consider from now on pivotal and ribbon categories which are linear_. ] .
it is to be understood that a monoidal linear category is a linear category equipped with a monoidal structure whose unit is given by a simple object @xmath200 and whose tensor product is given by a bilinear functor @xmath201 .
analogously , a pivotal linear category will be a strict monoidal linear category equipped with a pivotal structure whose duality functor is linear .
a ribbon linear category will simply be a pivotal linear category which is ribbon .
the _ left partial trace of a morphism @xmath202 of a ribbon linear category @xmath4 _ is the morphism @xmath203 given by @xmath204 and the _ right partial trace of @xmath205 _ is the morphism @xmath206 in @xmath207 given by @xmath208 a _ trace @xmath209 on an ideal @xmath193 of a ribbon linear category @xmath4 _ is a family @xmath210 of linear maps satisfying : 1 .
@xmath211 for all objects @xmath64 of @xmath4 and @xmath85 of @xmath193 and for all morphisms @xmath202 ; 2 .
@xmath212 for all objects @xmath64 of @xmath193 and @xmath85 of @xmath4 and for all morphisms @xmath202 ; 3 .
@xmath213 for all objects @xmath64 and @xmath85 of @xmath193 and for all morphisms @xmath188 and @xmath189 .
a trace @xmath209 on an ideal @xmath193 of a ribbon linear category @xmath4 is _ non - degenerate _ if @xmath214 is a non - degenerate pairing for every @xmath194 and for every @xmath215 .
the _ dimension @xmath216 associated with a trace @xmath209 on an ideal @xmath193 of a ribbon linear category @xmath4 _ is the function @xmath217 the main propeller for the theory of traces on ideals in ribbon linear categories is the existence of a special kind of object : a simple object @xmath64 of a ribbon linear category @xmath4 is _ ambidextrous _ if for all @xmath218 are part of the structure of the monoidal linear functor @xmath36 . ]
@xmath219 for every @xmath220 .
if @xmath4 is a ribbon linear category and if @xmath64 is an ambidextrous object of @xmath4 then there exists a unique trace @xmath209 on @xmath196 which satisfies @xmath221 .
moreover if @xmath64 is projective with epic evaluation then the unique trace @xmath209 on @xmath197 satisfying @xmath221 satisfies @xmath222 for every simple projective object @xmath85 of @xmath4 with epic evaluation is provided by corollary 3.2.1 of @xcite . ] .
the last ingredient we will need for the definition of relative modular categories is the action of an abelian group @xmath2 .
if @xmath223 is a symmetric @xmath10-bilinear map then we denote with @xmath224 the symmetric monoidal linear category with set of objects @xmath225 , with vector spaces of morphisms satisfying @xmath226 with unit @xmath227 , with tensor product mapping each pair of objects @xmath228 of @xmath229 to @xmath230 and with braiding given by the morphisms @xmath231 for all @xmath232 .
realization of @xmath2 in a braided monoidal category @xmath4 _ is a braided monoidal functor @xmath233 . for every realization @xmath234 of @xmath2 in @xmath4 and for all @xmath232 we will denote with @xmath235 and with @xmath236 the morphisms of @xmath4 coming from the structure of @xmath234 , and we denote with @xmath237 the set of objects @xmath238 of @xmath4 .
if @xmath233 is a realization of @xmath2 in @xmath4 then all objects in @xmath239 are simple and @xmath240 is isomorphic to @xmath241 for all @xmath218 and for every @xmath242 .
in particular @xmath243 is simple if and only if @xmath64 is simple .
a _ free realization of @xmath2 in a braided monoidal category @xmath4 _ is a realization of @xmath2 in @xmath4 such that for every object @xmath64 of @xmath4 the tensor product @xmath243 is isomorphic to @xmath64 if and only if @xmath244 .
relative modular categories will be a non - semisimple version of modular categories with group structure .
however , for them to be manageable , we will need to be able to control their non - semisimplicity .
in particular , like in @xcite and in @xcite , we will want their generic homogeneous subcategory to be semisimple , with the non - semisimple part of the category localized in a `` small '' set of critical indices .
let us start by fixing the terminology for semisimple categories .
we say a dominating set . ]
@xmath245 for a linear category @xmath4 is _ reduced _ if @xmath246 for all @xmath247 .
a _ semisimple category _ is then a linear category @xmath4 which admits a reduced dominating set .
a subset @xmath3 of an abelian group @xmath1 is _ small symmetric _ if : 1 .
@xmath248 ; 2 .
@xmath249 for all @xmath250 and all @xmath251 .
if @xmath1 is an abelian group and @xmath6 is a small symmetric subset then all elements in @xmath252 are called _ generic _ and all elements in @xmath3 are called _
critical_. let us fix such an abelian group @xmath1 with a small symmetric subset @xmath6 .
we say a linear @xmath1-category @xmath4 is _ semisimple relative to @xmath3 _ if : 1 .
@xmath62 is semisimple for all @xmath253 ; 2 .
@xmath254 .
if @xmath4 is a semisimple @xmath1-category relative to @xmath3 then @xmath3 is called the _ set of critical indices _ or simply the _
critical set_. for a semisimple @xmath1-category relative to @xmath3 we will use the term _ relative semisimple category _ every time there will be no need to explicitly mention the structure group @xmath1 and the critical set @xmath6 .
a _ semisimple ribbon @xmath1-category relative to @xmath3 _ is just a ribbon linear @xmath1-category which is semisimple relative to @xmath3 .
[ relative_pre - modular_graded_categories_def ] if @xmath1 and @xmath2 are abelian groups and if @xmath6 is a small symmetric subset then a _ pre - modular @xmath1-category relative to @xmath5 _ is the given of : 1 . a semisimple ribbon @xmath1-category @xmath4 relative to @xmath6 ; 2 .
a set @xmath255 of simple projective objects of @xmath62 with epic evaluation for every @xmath253 and for some ordered finite index set @xmath256 ; 3 .
an ambidextrous projective object @xmath257 of @xmath4 with epic evaluation ; 4 .
a free realization @xmath258 .
these data satisfy the following conditions : 1 .
@xmath259 is a reduced dominating set for @xmath62 ; 2 .
@xmath260 ; 3 .
@xmath261 for all @xmath61 , @xmath59 and @xmath242 and for some @xmath10-bilinear homomorphism @xmath262 .
the group @xmath2 is called the _ periodicity group of @xmath4_. if @xmath4 is a pre - modular @xmath1-category relative to @xmath5 we will denote with @xmath263 the set of objects @xmath264 , which is in general not finite , and we will denote with @xmath209 the unique trace on @xmath197 satisfying @xmath265 . [ non - degeneracy_of_trace ] if @xmath4 is a pre - modular @xmath1-category relative to @xmath5 then the trace @xmath209 on @xmath197 is non degenerate .
we begin with a remark . if @xmath266 is a projective object of @xmath4 and if @xmath267 then the morphism @xmath268 is still epic ecause @xmath266 is dualizable .
therefore , since @xmath266 is projective , there exists a section @xmath269 of @xmath268 , which is a morphism in @xmath270 satisfying @xmath271 .
let us consider a homogeneous projective object @xmath266 , a homogeneous object @xmath272 of the same index and a non - zero morphism @xmath273 between them .
then if @xmath267 the morphism @xmath274 is a non - zero morphism of @xmath62 because @xmath275 for every section @xmath269 of @xmath268 .
then , since @xmath62 is semisimple and @xmath133 is simple , there exists a retraction @xmath276 of @xmath277 , which is a morphism in @xmath278 satisfying @xmath279 let @xmath280 denote the morphism @xmath281 and let @xmath282 denote the morphism @xmath283 then @xmath284 if @xmath4 is a pre - modular @xmath1-category relative to @xmath5 then the _ kirby color of index @xmath253 _ is the formal linear combination of objects @xmath285 if @xmath8 is a @xmath1-homogeneous @xmath4-colored ribbon graph , if @xmath286 is a framed knot disjoint from @xmath84 and if @xmath253 is a generic index then we denote with @xmath287 the formal linear combination of @xmath1-homogeneous @xmath4-colored ribbon graphs @xmath288 where @xmath289 .
although @xmath287 is not actually a morphism of @xmath129 , we can still naturally define its image @xmath290 under the reshetikhin - turaev functor as @xmath291 [ graded_delta_def ] for a pre - modular @xmath1-category @xmath4 relative to @xmath5 there exist constants @xmath292 satisfying @xmath293 for every @xmath59 with @xmath253 , where @xmath294 and @xmath295 are the images under the reshetikhin - turaev functor @xmath187 of the @xmath4-colored ribbon tangles depicted in figure [ graded_delta ] .
observe that @xmath296 and @xmath297 do not depend neither on @xmath59 nor on @xmath253 .
this is proved in lemma 5.10 of @xcite .
[ graded_delta ] -colored ribbon tangles representing @xmath294 and @xmath295.,title="fig : " ] [ relative_modular_graded_categories_def ] a _ modular @xmath1-category relative to @xmath5 _ is a pre - modular @xmath1-category relative to @xmath5 which admits a _ relative modularity parameter _
@xmath298 such that @xmath299 for all @xmath300 and for all @xmath253 , where @xmath301 is the image under the reshetikhin - turaev functor @xmath187 of the @xmath4-colored ribbon tangle depicted in figure [ non - semisimple_modularity ] .
[ non - semisimple_modularity ] if @xmath4 is a modular @xmath1-category relative to @xmath5 then the relative modularity parameter @xmath302 equals @xmath303 . the proof of this fact follows from the equality @xmath304 which holds for every @xmath59 , where @xmath305 is the image under the reshetikhin - turaev functor @xmath187 of the @xmath4-colored ribbon tangle depicted in figure [ non - degeneracy_of_modular_categories ] .
the above equality is in turn a direct consequence of the relative modularity condition for @xmath4 .
[ non - degeneracy_of_modular_categories ] -colored ribbon tangle representing @xmath306.,title="fig : " ] at first sight the relative modularity condition of definition [ relative_modular_graded_categories_def ] may seem strange , but it actually is a generalization of the usual modularity condition stated in terms of the invertibility of the positive hopf link matrix . let @xmath4 be a pre - modular @xmath1-category relative to @xmath5 . in the trivial case @xmath307 we have @xmath308 and we can choose @xmath38 as the specified projective ambidextrous object of @xmath4 .
then we have @xmath309 and , since @xmath310 , we have @xmath311 thanks to lemma 16 of @xcite . in particular
@xmath4 is semisimple with reduced dominating set @xmath312 for some finite ordered index set @xmath313 , so it is pre - modular also in the usual sense . in the following the term _ pre - modular category _
will denote a pre - modular @xmath314-category relative to @xmath315 .
if @xmath4 is a pre - modular category then the _ positive hopf link matrix of @xmath4 _ is the @xmath316 matrix whose @xmath317-th entry @xmath318 is defined as the evaluation of the reshetikhin - turaev functor @xmath187 against the @xmath4-colored framed link depicted in the left - hand part of figure [ s - matrix ] for all @xmath319 .
[ s - matrix ] -colored framed links representing the @xmath317-th entry of the positive and of the negative hopf link matrix of @xmath4 respectively.,title="fig : " ] a pre - modular category @xmath4 is modular if and only if it its positive hopf link matrix is invertible . if the positive hopf link matrix of @xmath4 is invertible than it is known that the ( relative ) modularity condition holds .
then let us suppose @xmath4 is modular and let us consider the negative hopf link matrix of @xmath4 whose @xmath317-th entry @xmath320 is defined as the evaluation of the reshetikhin - turaev functor @xmath187 against the @xmath4-colored framed link depicted in the right - hand part of figure [ s - matrix ] for all @xmath319 .
now we have @xmath321 for the @xmath4-colored ribbon tangles depicted in figure [ long_hopf_links ] for all @xmath322 . [ long_hopf_links ] -colored ribbon tangles representing @xmath323 and @xmath324.,title="fig : " ] therefore we have @xmath325 where @xmath326 is given by the evaluation of the reshetikhin - turaev functor @xmath187 against the @xmath4-colored framed link depicted in figure [ proof_equivalence_modularity ] for all @xmath327 . as a consequence we get @xmath328 and
we can conclude .
[ proof_equivalence_modularity ] -colored framed link representing @xmath326.,title="fig : " ]
this section is devoted to the definition of the domain 2-category of our extended tqfts , the symmetric monoidal 2-category of admissible 3-dimensional cobordisms .
we fix an abelian group @xmath1 and we introduce a new piece of decoration for cobordisms equipped with ribbon sets and ribbon graphs : @xmath1-colorings .
they consist of relative cohomology classes with coefficients in @xmath1 together with discrete sets of specified base points which are needed in order to induce @xmath1-colorings on horizontal and vertical gluings
. their role will be to determine indices for surgery presentations of 3-dimensional manifolds . if @xmath329 is a 1-dimensional smooth manifold without boundary then a _ @xmath1-coloring @xmath330 of @xmath329 _
is the given of : 1 .
a finite set @xmath331 called the _ base set _ whose elements are called _ base points _ ; 2 . a cohomology class @xmath332 .
these data satisfy the following condition : if @xmath333 is a non - empty connected component of @xmath329 then @xmath334 for exactly one point @xmath335 . if @xmath67 is a 2-dimensional cobordism from @xmath329 to @xmath336 , if @xmath69 is a ribbon set inside @xmath67 and if @xmath337 and @xmath338 are @xmath1-colorings of @xmath329 and @xmath336 respectively then a _ @xmath1-coloring @xmath339 of @xmath340 extending @xmath337 and @xmath338 _ is the given of : 1 . a finite set @xmath341 called the _ base set _ whose elements are called _ base points _ ; 2 . a cohomology class @xmath342 where @xmath343 these data satisfy the following conditions : 1 .
if @xmath344 is a non - empty connected component of @xmath345 then @xmath346 2 .
@xmath347 and @xmath348 for the embeddings @xmath349 induced by @xmath350 and @xmath351 respectively .
every @xmath1-coloring @xmath330 of @xmath329 naturally determines a @xmath1-coloring @xmath352 of @xmath353 extending @xmath330 and @xmath330 whose base set is given by @xmath354 and whose cohomology class is given by @xmath355 for the projection @xmath356 induced by the natural projection onto the second factor of @xmath357 . if @xmath67 and @xmath75 are 2-dimensional cobordisms from @xmath329 to @xmath336 , if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath82 is a ribbon graph from @xmath68 to @xmath358 , if @xmath337 and @xmath338 are @xmath1-colorings of @xmath329 and @xmath336 respectively , if @xmath359 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 respectively which both extend @xmath337 and @xmath338 then a _ @xmath1-coloring @xmath9 of @xmath362 extending @xmath359 and @xmath360 _ is a cohomology class @xmath363 where @xmath364 this data satisfies the following condition : @xmath365 , @xmath366 , @xmath367 and @xmath368 for the embeddings @xmath369 induced by @xmath370 , @xmath371 , @xmath372 , @xmath373 respectively and for the projections @xmath374 induced by the natural projections onto the first factors of @xmath375 and @xmath376 respectively . if @xmath67 and @xmath75 are 2-dimensional cobordisms from @xmath329 to @xmath336 , if @xmath80 and @xmath117 are 3-dimensional cobordisms with corners from @xmath67 to @xmath75 , if @xmath377 is a ribbon graph from @xmath68 to @xmath358 , if @xmath337 and @xmath338 are @xmath1-colorings of @xmath329 and @xmath336 respectively , if @xmath359 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 respectively which both extend @xmath337 and @xmath338 and if @xmath378 is a @xmath1-coloring of @xmath379 extending @xmath359 and @xmath360 then every isomorphism of cobordisms with corners @xmath380 induces a @xmath1-coloring @xmath381 of @xmath382 extending @xmath359 and @xmath360 given by the cohomology class @xmath383 where the induced isomorphism of pairs @xmath384 is still denoted @xmath205 by abuse of notation
if @xmath67 is a 2-dimensional cobordism from @xmath329 to @xmath336 , if @xmath330 and @xmath338 are @xmath1-colorings of @xmath329 and @xmath336 respectively and if @xmath68 is a ribbon set then very @xmath1-coloring @xmath339 of @xmath340 naturally determines a @xmath1-coloring @xmath385 of @xmath386 extending @xmath339 and @xmath339 whose base set is given by @xmath387 and whose cohomology class is given by @xmath388 for the map of pairs @xmath389 induced by the natural projection onto the first factor of @xmath390 .
let us fix a ribbon @xmath1-category @xmath4 . if @xmath67 is a 2-dimensional cobordism and
if @xmath68 is a ribbon set then we say a @xmath1-homogeneous @xmath4-coloring @xmath64 of @xmath69 and a @xmath1-coloring @xmath359 of @xmath340 are _ compatible _ if for every oriented vertex @xmath73 the homology class @xmath391 of a positive meridian of @xmath72 satisfies @xmath392 .
a _ @xmath393-coloring @xmath394 of @xmath340 _ is the given of a @xmath4-coloring @xmath64 of @xmath69 together with a compatible @xmath1-coloring @xmath359 of @xmath340 . if @xmath80 is a 3-dimensional cobordism with corners and
if @xmath82 is a ribbon graph then we say a @xmath1-homogeneous @xmath4-coloring @xmath115 of @xmath84 and a @xmath1-coloring @xmath9 of @xmath362 are _ compatible _ if for every oriented edge @xmath89 the homology class @xmath395 of a positive meridian of @xmath88 satisfies @xmath396 .
a _ @xmath393-coloring @xmath397 of @xmath362 _ is the given of a @xmath4-coloring @xmath115 of @xmath84 together with a compatible @xmath1-coloring @xmath9 of @xmath362 .
in this subsection we fix a ribbon @xmath1-category @xmath4 and we introduce the symmetric monoidal 2-category @xmath26 of 3-dimensional @xmath4-decorated cobordisms .
an _ object @xmath398 of @xmath26 _ is the given of a pair @xmath399 where : 1 .
@xmath329 is a smooth oriented closed 1-dimensional manifold ; 2 .
@xmath337 is a @xmath1-coloring of @xmath329 .
a _ 1-morphism @xmath400 of @xmath26 _ is the given of a 4-tuple @xmath401 where : 1 .
@xmath67 is a 2-dimensional cobordism from @xmath329 to @xmath336 ; 2 .
@xmath402 is a @xmath1-homogeneous @xmath4-colored ribbon set ; 3 .
@xmath359 is a @xmath1-coloring of @xmath340 extending @xmath337 and @xmath338 which is compatible with the @xmath4-coloring @xmath70 ; 4 .
@xmath403 is a lagrangian subspace of a finite dimensional real vector space @xmath404 equipped with an antisymmetric bilinear form is lagrangian if it satisfies @xmath405 . ] with respect to the intersection pairing @xmath406 .
the _ identity 1-morphism @xmath407 associated with an object @xmath408 of @xmath26 _ is the 4-tuple @xmath409 a _
2-morphism @xmath410 of @xmath26 _ between 1-morphisms @xmath411 is the given of an equivalence class of 4-tuples @xmath412 where : 1 .
@xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 ; 2 .
@xmath413 is a @xmath1-homogeneous @xmath4-colored ribbon graph from @xmath113 to @xmath114 ; 3 .
@xmath9 is a @xmath1-coloring of @xmath362 extending @xmath414 and @xmath415 which is compatible with the @xmath4-coloring @xmath86 ; 4 .
@xmath416 is called the _
signature defect_. two such 4-tuples @xmath417 and @xmath418 are equivalent if @xmath419 and if there exists an isomorphism of cobordisms with corners @xmath380 satisfying @xmath420 and @xmath421 .
the _ identity 2-morphism @xmath422 associated with a 1-morphism @xmath423 of @xmath26 _ is the equivalence class of the 4-tuple @xmath424 in order to define horizontal and vertical compositions of 1-morphisms and 2-morphisms we need to be able to induce uniquely determined @xmath1-colorings on horizontal and vertical gluings of @xmath4-decorated cobordisms .
the following observation will be repeatedly used for this purpose .
[ gluing_cohomology ] let @xmath425 , @xmath426 , @xmath427 be topological spaces , let @xmath428 , @xmath429 be subspaces and let @xmath430 , @xmath431 be homeomorphisms .
if @xmath432 , @xmath433 , @xmath434 are subsets satisfying @xmath435 and @xmath436 let us consider relative cohomology classes @xmath437 satisfying @xmath438 for the embeddings of pairs @xmath439 induced by @xmath440 and @xmath441 respectively .
then there exists a relative cohomology class @xmath442 satisfying @xmath443 and @xmath444 for the natural embeddings of pairs @xmath445 as follows from the mayer - vietoris sequence @xmath446 { } } \cdots \end{aligned}\ ] ] such a cohomology class is furthermore unique if the map @xmath447 is zero .
this happens , for instance , when @xmath448 .
if @xmath400 and @xmath449 are 1-morphisms of @xmath26 then the horizontal composition of @xmath75 with @xmath67 determines a unique @xmath1-coloring @xmath450 of @xmath451 extending @xmath330 and @xmath452 defined as follows : its base set is given by @xmath453 and its cohomology class is given by @xmath454 where @xmath455 is the unique cohomology class in @xmath456 given by remark [ gluing_cohomology ] and where @xmath457 is the inclusion of the pair @xmath458 into the pair @xmath459 with @xmath460 the _ horizontal composition @xmath461 of 1-morphisms @xmath449 and @xmath400 of @xmath26 _ is given by the 4-tuple @xmath462 if @xmath463 and @xmath464 are 1-morphisms and if @xmath465 and @xmath466 are 2-morphisms of @xmath26 then the horizontal composition of @xmath117 with @xmath80 determines a unique @xmath1-coloring @xmath467 of @xmath468 extending @xmath469 and @xmath470 given by @xmath454 where @xmath455 is the unique cohomology class in @xmath456 given by remark [ gluing_cohomology ] and where @xmath457 is the inclusion of the pair @xmath458 into the pair @xmath459 with @xmath460 the _ horizontal composition @xmath471 of 2-morphisms @xmath472 and @xmath473 of @xmath26 _ for 1-morphisms @xmath463 and @xmath464 is given by the equivalence class of the 4-tuple @xmath474 if @xmath475 are 1-morphisms and if @xmath476 and @xmath477 are 2-morphisms of @xmath26 then the vertical composition of @xmath117 with @xmath80 determines a @xmath1-coloring @xmath478 of @xmath479 extending @xmath339 and @xmath480 given by @xmath481 where @xmath482 is the unique class in @xmath483 given by remark [ gluing_cohomology ] and where @xmath457 is the inclusion of the pair @xmath484 into the pair @xmath485 with @xmath486 the _ vertical composition @xmath487 of 2-morphisms @xmath488 and @xmath410 of @xmath26 _ for 1-morphisms @xmath475 is given by the equivalence class of the 4-tuple @xmath489 the _ unit of @xmath26 _ is the unique object whose manifold is empty and it will be denoted @xmath140 .
let @xmath490 be topological spaces and let @xmath491 be subsets .
if @xmath492 and @xmath493 are cohomology classes then we denote with @xmath494 the cohomology class in @xmath495 defined as @xmath496 for the natural inclusions @xmath497 the _ tensor product @xmath498 of objects @xmath398 and @xmath499 of @xmath26 _ is given by the pair @xmath500 .
the _ tensor product @xmath501 of 1-morphisms @xmath502 and @xmath503 of @xmath26 _ is given by the 4-tuple @xmath504 the _ tensor product @xmath505 of 2-morphisms @xmath473 and @xmath472 of @xmath26 _ is given by the equivalence class of the 4-tuple @xmath506 [ braiding_1-morphism ] the _ braiding 1-morphism @xmath507 associated with objects @xmath398 and @xmath499 of @xmath26 _ is given by the 4-tuple for the flip cobordism @xmath508 . ]
@xmath509 [ braiding_2-morphism ]
the _ braiding 2-morphism @xmath510 associated with 1-morphisms @xmath502 and @xmath503 of @xmath26 _ is given by the 4-tuple for the flip cobordism with corners @xmath511 . ]
@xmath512 in this subsection we fix a pre - modular @xmath1-category @xmath4 relative to @xmath5 and we define the symmetric monoidal 2-category of 3-dimensional admissible @xmath4-decorated cobordisms associated with any pre - modular @xmath1-category @xmath4 relative to @xmath5
. it will be on this category that our quantum invariants , tqfts and etqfts will be defined .
if @xmath67 is a 2-dimensional cobordism and if @xmath68 is a ribbon set then a vertex @xmath73 is said to be _ projective with respect to a @xmath4-coloring @xmath64 of @xmath69 _ if @xmath71 is a projective object of @xmath4 .
a @xmath4-coloring @xmath70 is _ admissible _ if there exists a projective vertex @xmath73 with respect to @xmath64 .
if @xmath80 is a 3-dimensional cobordism with corners and if @xmath82 is a ribbon graph then an edge @xmath89 is said to be _ projective with respect to a @xmath4-coloring @xmath115 of @xmath84 _ if @xmath87 is a projective object of @xmath4 .
a @xmath4-coloring @xmath86 is _ admissible _ if there exists a projective edge @xmath89 with respect to @xmath115 .
if @xmath67 is a 2-dimensional cobordism and if @xmath68 is a ribbon set then a simple closed oriented curve @xmath513 is _ generic with respect to a @xmath1-coloring @xmath339 of @xmath340 _ if @xmath514 as a homology class .
a @xmath1-coloring @xmath339 of @xmath340 is _ admissible _ if there exists a generic simple closed oriented curve @xmath513 with respect to @xmath339 .
if @xmath80 is a 3-dimensional cobordism with corners and if @xmath82 is a ribbon graph then an oriented knot @xmath515 is _ generic with respect to a @xmath1-coloring @xmath9 of @xmath362 _ if @xmath516 as a homology class .
a @xmath1-coloring @xmath9 of @xmath362 is _ admissible _ if there exists a generic oriented knot @xmath515 with respect to @xmath9 .
if @xmath517 is a 2-dimensional cobordism with connected components @xmath518 for @xmath519 and if @xmath68 is a ribbon set which decomposes as @xmath520 with @xmath521 then every @xmath393-coloring @xmath522 of @xmath340 naturally restricts to a @xmath393-coloring of @xmath523 for all @xmath519 which will be denoted @xmath524 .
analogously if @xmath525 is a 3-dimensional cobordism with corners with connected components @xmath526 for @xmath519 and if @xmath82 is a ribbon graph which decomposes as @xmath527 with @xmath528 then every @xmath393-coloring @xmath397 of @xmath362 naturally restricts to a @xmath393-coloring of @xmath529 for all @xmath519 which will be denoted @xmath530 .
if @xmath517 is a 2-dimensional cobordism with connected components @xmath518 for @xmath519 and if @xmath68 is a ribbon set which decomposes as @xmath520 with @xmath521 then a @xmath393-coloring @xmath522 of @xmath340 is _ admissible _ if either @xmath133 or @xmath531 is admissible for every connected component @xmath518 of @xmath67 disjoint from the incoming boundary @xmath532 , and it is _ strongly admissible _ if either @xmath133 or @xmath531 is admissible for every connected component @xmath518 of @xmath67 .
if @xmath525 is a 3-dimensional cobordism with corners with connected components @xmath526 for @xmath519 and if @xmath82 is a ribbon graph which decomposes as @xmath527 with @xmath528 then a @xmath393-coloring @xmath397 of @xmath362 is _ admissible _ if if either @xmath533 or @xmath534 is admissible for every connected component @xmath526 of @xmath80 disjoint from the incoming horizontal boundary @xmath93 , and it is _ strongly admissible _ if either @xmath533 or @xmath534 is admissible for every connected component @xmath526 of @xmath80 .
a 1-morphism @xmath423 of @xmath26 is _ admissible _ if @xmath522 is an admissible @xmath393-coloring for @xmath340 , and it is _ strongly admissible _ if @xmath522 is a strongly admissible @xmath393-coloring for @xmath340 .
the previous definition has the following direct consequences : 1 .
if @xmath398 is an object of @xmath26 then @xmath535 is an admissible 1-morphism .
if @xmath449 and @xmath400 are admissible 1-morphisms of @xmath26 then @xmath536 is an admissible 1-morphism .
if @xmath537 and @xmath538 are admissible 1-morphisms of @xmath26 then @xmath539 is an admissible 1-morphism .
a 2-morphism @xmath540 of @xmath26 is _ admissible _ if @xmath397 is an admissible @xmath393-coloring for @xmath362 .
the previous definition has the following direct consequences : 1 .
if @xmath537 is an admissible 1-morphism of @xmath26 then @xmath541 is an admissible 2-morphism .
if @xmath466 and @xmath465 are admissible 2-morphisms of @xmath26 for 1-morphisms @xmath463 and @xmath464 then @xmath542 is an admissible 2-morphism .
if @xmath477 and @xmath476 are admissible 2-morphisms of @xmath26 for 1-morphisms @xmath475 then @xmath543 is an admissible 2-morphism .
if @xmath544 and @xmath545 are admissible 2-morphisms of @xmath26 then @xmath546 is an admissible 2-morphism .
the symmetric monoidal 2-category @xmath17 of admissible 3-dimensional cobordisms is the symmetric monoidal 2-category whose objects are objects of @xmath26 , whose morphisms are admissible morphisms of @xmath26 and whose symmetric monoidal structure is inherited by that of @xmath26 . in this subsection
we consider a pre - modular @xmath1-category @xmath4 relative to @xmath5 and we fix the terminology for quantum invariants , tqfts and etqfts .
we also recall the main ideas of the extended universal construction , the fundamental machinery our investigation is based on , postponing a detailed account to appendix [ extended_universal_construction ] .
morphisms of @xmath17 from @xmath140 to itself are naturally arranged in a symmetric monoidal category with respect to the tensor product of 1-morphisms and 2-morphisms and with unit given by the empty 1-morphism @xmath547 .
this category will be denoted @xmath548 for brevity .
analogously , the set of 2-morphisms of @xmath17 from @xmath547 to itself forms a commutative monoid with respect to the tensor product of 2-morphisms and with unit given by the empty 2-morphism @xmath549 .
this monoid will be denoted @xmath550 for brevity .
an _ invariant on @xmath550 _ is a function from @xmath550 to @xmath22 .
a _ covariant quantization functor on @xmath548 _ is a functor from @xmath548 to @xmath551 .
a _ contravariant quantization functor on @xmath548 _ is a functor from @xmath552 to @xmath551 .
a _ covariant extended quantization functor on @xmath17 _ is a functor from @xmath17 to @xmath553 .
a _ contravariant extended quantization functor on @xmath17 _ is a functor from @xmath554 to @xmath553 .
a _ quantum invariant on @xmath550 _ is an invariant on @xmath550 which is a commutative monoid homomorphism .
tqft on @xmath548 _ is a covariant quantization functor on @xmath548 which is symmetric monoidal .
etqft on @xmath17 _ is a symmetric monoidal functor from @xmath17 to @xmath27 .
@xmath2-graded tqft on @xmath548 _ is a symmetric monoidal functor from @xmath548 to @xmath555 .
@xmath2-graded etqft on @xmath17 _ is a symmetric monoidal functor from @xmath17 to @xmath18 .
the extended universal construction is a machinery which associates with every invariant @xmath556 on @xmath550 a covariant and a contravariant extended quantization functors on @xmath17 denoted @xmath557 and @xmath558 respectively .
the image of an object @xmath398 of @xmath17 under @xmath557 is a linear category @xmath559 called the _ covariant universal algebroid of @xmath398 _ whose objects are 1-morphisms of @xmath17 of the form @xmath560 and whose morphism vector spaces @xmath561 are given by certain quotients , which are defined using @xmath556 , of the free complex vector spaces generated by 2-morphisms of @xmath17 of the form @xmath562 .
the image of a 1-morphism @xmath400 of @xmath17 under @xmath557 is a covariant linear functor @xmath563 called the _ covariant universal linear functor of @xmath537 _ which maps every object @xmath564 of @xmath559 to the object @xmath565 of @xmath566 and every morphism @xmath567 $ ] of @xmath559 to the morphism @xmath568 $ ] of @xmath566 .
the image of a 2-morphism @xmath410 of @xmath17 under @xmath557 is a natural transformation @xmath569 called the _ covariant universal natural transformation of @xmath544 _ which associates with every object @xmath564 of @xmath559 the morphism @xmath570 $ ] of @xmath566 .
the contravariant extended quantization functor @xmath558 is defined similarly .
we move on to the study of the extended quantization functors given by the extended universal construction .
we begin by introducing a set of axioms for quantum invariants on @xmath550 which , when met by a particular quantum invariant , enables us to better handle universal algebroids , linear functors and natural transformations .
this is much in the spirit of @xcite and @xcite .
[ pre - modular_from_now_on ] from now on @xmath4 will denote a fixed pre - modular @xmath1-category relative to @xmath5 and @xmath571 will denote a fixed index in @xmath256 for every generic @xmath253 .
this choice specifies a fixed projective simple object @xmath572 of @xmath62 .
let @xmath423 , @xmath573 be 1-morphisms of @xmath17 from @xmath398 to @xmath499 , let @xmath80 be a cobordism with corners from @xmath67 to @xmath75 , let @xmath112 be a @xmath1-homogeneous @xmath4-colored ribbon graph from @xmath113 to @xmath114 and let @xmath574 be an integer .
if @xmath575 is a framed link disjoint from @xmath84 and if @xmath9 is a @xmath1-coloring of @xmath576 extending @xmath339 and @xmath360 which is compatible with @xmath86 and which satisfies @xmath577 for a positive meridian @xmath578 of @xmath579 for all @xmath580 then we denote with @xmath581 the formal linear combination of 2-morphisms of @xmath17 @xmath582 although @xmath583 is not actually a 2-morphism of @xmath17 , we will treat it as such by abuse of notation . for every generic @xmath253 the _ @xmath63-colored index 0 surgery 1-morphism @xmath584 of @xmath17 _
is given by @xmath585 [ graded_surgery_index_1_def ] for every generic @xmath253 the _
@xmath63-colored index 1 1-morphism @xmath586 of @xmath17 _ is given by @xmath587 where the @xmath4-colored ribbon set @xmath588 is given by @xmath589 for the poles @xmath590 and @xmath591 , with orientation given by the sign @xmath592 on @xmath593 and @xmath594 and by the sign @xmath595 on @xmath596 and @xmath597 , with framing tangent to @xmath598 and with @xmath4-coloring given by the object @xmath599 for every point , where the base set @xmath600 is given by @xmath601 for some equatorial point @xmath602 and where @xmath603 is the only cohomology class yielding a @xmath1-coloring of @xmath604 which is compatible with the @xmath4-coloring of @xmath605 .
see figure [ graded_surgery_index_1 ] for a graphical representation .
[ graded_surgery_index_1 ] .
the ribbon set @xmath605 is given by the four framed blue vertices .
the base set @xmath606 is given by the two starred red equatorial points .
the oriented red equators , which form a basis for the first homology group , are evaluated to @xmath63 by @xmath603.,title="fig : " ] for every generic @xmath253 the _
@xmath63-colored index 2 1-morphism @xmath607 of @xmath17 _ is given by @xmath608 where the base set @xmath609 is given by @xmath610 , where @xmath611 is the cohomology class determined by @xmath612 and @xmath613 for the positive meridian @xmath614 and for the positive longitude @xmath615 and where the lagrangian subspace @xmath616 is generated by @xmath617 .
see figure [ graded_surgery_index_2 ] for a graphical representation .
[ graded_surgery_index_2 ] .
the base set @xmath618 is given by the starred red intersection point between the meridian and the longitude .
these two oriented red curves , which form a basis for the first homology group , are evaluated to @xmath63 and to 0 respectively by @xmath611.,title="fig : " ] for every generic @xmath253 the _
@xmath63-colored index 0 attaching 2-morphism @xmath619 of @xmath17 _ is given by @xmath620 [ graded_surgery_index_1_attaching_def ] for every generic @xmath253 the _
@xmath63-colored index 1 attaching 2-morphism @xmath621 of @xmath17 _ is given by @xmath622 where the @xmath4-colored ribbon tangle @xmath623 is given by @xmath624 with orientation determined by that of @xmath625 , with framing tangent to @xmath626 and with @xmath4-coloring uniquely determined by the @xmath4-coloring already present on @xmath627 and where @xmath628 is the only @xmath1-coloring of @xmath629 extending @xmath603 .
see figure [ graded_surgery_index_1_attaching ] for a graphical representation .
[ graded_surgery_index_1_attaching ] . the ribbon tangle @xmath630 is represented by the two vertical blue edges with blackboard framing .
the cohomology class @xmath628 is completely determined by @xmath603.,title="fig : " ] for every generic @xmath253 the _
@xmath63-colored index 2 attaching 2-morphism @xmath631 of @xmath17 _ is given by @xmath632 where the framed knot @xmath633 is given by @xmath634 with orientation induced by @xmath635 and with framing tangent to @xmath636 and where @xmath637 is the only @xmath1-coloring of @xmath638 extending @xmath611 .
see figure [ graded_surgery_index_2_attaching ] for a graphical representation . [ graded_surgery_index_2_attaching ] . the framed knot @xmath633 is given by the oriented blue core of the solid torus with blackboard framing .
the cohomology class @xmath637 is completely determined by @xmath611.,title="fig : " ] for every generic @xmath253 the _
@xmath63-colored index 0 belt 2-morphism @xmath639 of @xmath17 _ is given by @xmath640 where the @xmath4-colored framed knot @xmath641 is given by the unknot @xmath642 with any orientation , with framing tangent to @xmath643 and with @xmath4-coloring given by the object @xmath599 and where @xmath644 is the only @xmath1-coloring of @xmath645 which is compatible with the @xmath4-coloring of @xmath646 .
[ graded_surgery_index_1_belt_def ] for every generic @xmath253 the _
@xmath63-colored index 1 belt 2-morphism @xmath647 of @xmath17 _ is given by @xmath648 where the @xmath4-colored ribbon tangle @xmath649 is given by @xmath650 with orientation determined by that of @xmath625 , with framing tangent to @xmath651 and with @xmath4-coloring uniquely determined by the @xmath4-coloring already present on @xmath627 and where @xmath652 for the natural projection @xmath653 see figure [ graded_surgery_index_1_belt ] for a graphical representation .
[ graded_surgery_index_1_belt ] . the ribbon tangle @xmath654 is represented by the two vertical blue edges with blackboard framing .
the red line joining the two starred red equatorial points , which , together with any oriented red equator , forms a basis for the first homology group , is evaluated to 0 by @xmath655.,title="fig : " ] for every generic @xmath253 the _
@xmath63-colored index 2 belt 2-morphism @xmath656 of @xmath17 _ is given by @xmath657 where @xmath658 is the only @xmath1-coloring of @xmath659 extending @xmath611 .
see figure [ graded_surgery_index_2_belt ] for a graphical representation .
[ graded_surgery_index_2_belt ] .
the cohomology class @xmath658 is completely determined by @xmath611.,title="fig : " ] a quantum invariant @xmath556 on @xmath550 is said to _ satisfy the surgery axioms _ if there exist @xmath660 such that for every 2-morphism @xmath661 of @xmath17 we have @xmath662 for @xmath663 and for every @xmath253 .
[ first_surgery_remark ] let @xmath556 be a non - trivial quantum invariant on @xmath550 satisfying the surgery axioms .
then : 1 . if @xmath664 is a closed 2-morphism of @xmath17 which satisfies @xmath665 then @xmath666 implies @xmath667 for all @xmath253 ; 2 .
let @xmath668 denote the 2-morphism of @xmath17 given by @xmath669 where , using the notation of definition [ graded_surgery_index_1_attaching_def ] , @xmath670 is obtained from @xmath630 by reversing the orientation of the tangle @xmath671
. then the chain of equalities @xmath672 implies @xmath673 for all @xmath253 ; 3 . for every @xmath253 and every @xmath674 let @xmath675 denote the 2-morphism of @xmath17 given by @xmath676 where , using the notation of definition [ graded_surgery_index_1_def ] , the @xmath1-coloring @xmath677 satisfies @xmath678 for the incoming horizontal boundary identification @xmath679 and @xmath680 let also @xmath681 denote the 2-morphism of @xmath17 given by @xmath682 where , using the notation of definition [ graded_surgery_index_1_belt_def ] , @xmath683 is obtained from @xmath654 by reversing the orientation of the tangle @xmath684
. then we have @xmath685 for every @xmath253 and every @xmath674 .
let @xmath686 be the @xmath4-colored framed link depicted in figure [ index_2_surgery_link_1 ] .
if @xmath556 is a non - trivial quantum invariant on @xmath550 which satisfies the surgery axioms then we have @xmath687 for every @xmath253 , where @xmath688 is the only @xmath1-coloring of @xmath689 wich is compatible with the @xmath4-coloring of @xmath690 .
indeed , for all @xmath691 , let @xmath692 denote the 2-morphism of @xmath17 given by @xmath693 where the @xmath4-colored ribbon graph @xmath694 is given by @xmath695 with opposite orientations on the two components , with framing tangent to @xmath696 and with constant @xmath4-coloring given by the object @xmath697 and where @xmath698 is the only @xmath1-coloring of @xmath699 extending @xmath700 wich is compatible with the @xmath4-coloring of @xmath701
. then we have @xmath702 for all @xmath691 .
[ index_2_surgery_link_1 ] -colored framed link @xmath686 with blackboard framing.,title="fig : " ] let @xmath556 be a non - trivial quantum invariant on @xmath550 satisfying the surgery axioms . then @xmath703 with @xmath704 where the @xmath4-colored framed knot @xmath705 is given by @xmath706 with any orientation , with framing parallel to @xmath707 and with @xmath4-coloring given by the object @xmath599 and where @xmath708 is the unique @xmath1-coloring of @xmath709 which is compatible with the @xmath4-coloring of @xmath710 .
[ graded_surgery_equivalence ] let @xmath556 be a quantum invariant on @xmath550 , let @xmath398 be an object of @xmath17 , let @xmath711 be objects of @xmath559 and let @xmath712 be a 2-morphism of @xmath17 with @xmath713 and @xmath253 . if @xmath556 satisfies the surgery axioms then @xmath714 = \lambda_{k , g } \cdot \left [ { \mathbf{m}}_{{\mathbf{\gamma } } } \ast \left ( { \mathbf{a}}_{k , g } \otimes { \mathrm{id}}_{{\mathbf{\sigma}}_{{\mathbf{\gamma } } } } \right ) \right]\ ] ] as vectors of @xmath715 .
we have to show that @xmath716 where for every 2-morphism @xmath717 of @xmath17 @xmath718 is the natural transformation coming from the extended universal construction ( see appendix [ extended_universal_construction ] for the definition ) .
but then the equality we have to prove is equivalent to the equality @xmath719 for every 2-morphism @xmath661 of @xmath17 , which is precisely the @xmath63-colored index @xmath720 surgery axiom .
in this section we derive the main consequences of surgery axioms on universal algebroids : all morphisms between a fixed pair of strongly admissible objects of a universal algebroid associated with a quantum invariant which satisfies the surgery axioms are generated by all possible decorations on a fixed connected cobordism with corners .
[ surgery_notation ] we fix here our notation for surgery .
if @xmath80 and @xmath117 are 3-dimensional cobordisms with corners from @xmath67 to @xmath75 then a surgery framed link @xmath721 for @xmath117 inside @xmath80 determines , up to isomorphism , a cobordism with corners @xmath722 from @xmath67 to @xmath75 , where @xmath723 is the exterior of @xmath721 in @xmath80 , together with an isomorphism @xmath724 of cobordisms with corners restricting to an indentification @xmath725 of the exteriors of @xmath721 which determines an isomorphism between @xmath117 and @xmath726 if @xmath721 is a surgery framed link for @xmath117 inside @xmath80 then @xmath578 denotes the @xmath103-th meridian @xmath727 corresponding to the @xmath103-th copy of @xmath728 in the incoming horizontal boundary @xmath729 of @xmath723 .
let @xmath411 be 1-morphisms of @xmath17 and let @xmath80 be a cobordism with corners from @xmath67 to @xmath75 . if @xmath730 is a framed link then for every @xmath4-colored ribbon graph @xmath112 from @xmath113 to @xmath114 the inclusion @xmath731 induces an isomorphism between the first homology groups of the two pairs . in particular
if @xmath732 is a 2-morphism of @xmath17 of the form @xmath733 and if @xmath721 gives a surgery presentation for @xmath117 inside @xmath80 then we denote with @xmath734 the unique @xmath1-coloring of @xmath735 satisfying @xmath736 where @xmath457 denotes the inclusion of the pair @xmath737 into the pair @xmath738 where @xmath739 denotes the inclusion of the pair @xmath740 into the pair @xmath741 and where @xmath742 also denotes , by abuse of notation , the isomorphism of pairs induced by @xmath742 between @xmath737 and @xmath743 let @xmath411 be 1-morphisms of @xmath17 , let @xmath80 be a non - empty connected cobordism with corners from @xmath67 to @xmath75 , let @xmath732 be a 2-morphism of @xmath17 of the form @xmath733 for some non - empty connected cobordism with corners @xmath117 from @xmath67 to @xmath75 , let @xmath744 be a framed link giving a surgery presentation for @xmath117 and let @xmath724 be an isomorphism of cobordisms with corners .
we say that _
@xmath721 gives a computable surgery presentation for @xmath545 _ , or that _ @xmath721 is a computable surgery framed link for @xmath545 _ , if @xmath745 for all @xmath134 , and in this case we denote with @xmath746 the formal linear combination of @xmath4-colored ribbon graphs inside @xmath80 given by @xmath747 . [ computable ] if @xmath411 are 1-morphisms of @xmath17 , if @xmath80 is a non - empty connected cobordism with corners from @xmath67 to @xmath75 and if @xmath732 is a non - empty connected 2-morphism of @xmath17 then there exists a computable surgery framed link for @xmath545 inside @xmath80 .
a proof of this lemma in the case where @xmath748 can be found in @xcite .
the proof of the general case is completely analogous .
[ graded_surgery_remark ] if @xmath556 is a quantum invariant on @xmath550 which satisfies the surgery axioms , if @xmath398 is an object of @xmath17 , if @xmath749 are objects of @xmath559 , if @xmath750 is a non - empty connected cobordism with corners from @xmath751 to @xmath752 and if @xmath717 is a non - empty connected 2-morphism of @xmath17 of the form @xmath753 then lemmas [ computable ] and [ graded_surgery_equivalence ] yield the equality @xmath754 = \lambda_{2,g_1 } \cdots \lambda_{2,g_k } \cdot [ { \mathrm{m}}_{\vargamma},{\mathrm{l}}^{\omega '' } \cup f_{{\mathrm{l}}}^{-1}({\mathrm{t}}''^{\varphi''}),f_{{\mathrm{l}}}^{\ # } \omega'',n]\ ] ] between morphisms of @xmath715 , where @xmath755 is some computable surgery framed link for @xmath756 inside @xmath750 , where @xmath757 is contained inside @xmath758 up to isotopy and where @xmath416 is determined via the maslov index formula .
[ g - circle ] the _
@xmath63-colored 1-sphere @xmath759 _ for an index @xmath61 is the object of @xmath26 given by @xmath760 where @xmath761 is the base set given by @xmath762 and where @xmath763 is the unique cohomology class satisfying @xmath764 .
if @xmath63 is generic then @xmath765 is said to be a _ generic circle _ , while if @xmath63 is critical then @xmath765 is said to be a _
critical circle_. [ ( epsilon , v)-disc ] the
_ @xmath130-colored 2-disc @xmath766 _ for an object @xmath130 of @xmath129 is the 1-morphism of @xmath26 given by @xmath767 where @xmath768 is the base set given by @xmath769 and where @xmath770 is the unique cohomology class compatible with the @xmath4-coloring of @xmath771 which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath772 .
[ t^phi - cylinder ] the _
@xmath8-colored 3-cylinder @xmath773 _ for a morphism @xmath153 of @xmath129 with @xmath774 is the 2-morphism of @xmath26 given by @xmath775 where @xmath776 is the unique @xmath1-coloring extending @xmath777 and @xmath778 which is compatible with the @xmath4-coloring of @xmath84 and which evaluates to 0 the relative homology class of the oriented arc @xmath779 .
a quantum invariant @xmath556 on @xmath550 satisfies the _ skein axiom _ if for all objects @xmath130 and @xmath159 of @xmath129 satisfying @xmath780 and for all morphisms @xmath781 of @xmath129 satisfying @xmath782 we have @xmath783 for every choice of a 1-morphism @xmath784 of @xmath17 and of 2-morphisms @xmath785 and @xmath786 of @xmath17 .
[ evocation_equivalence ] let @xmath556 be a quantum invariant on @xmath550 , let @xmath130 and @xmath159 be objects of @xmath129 satisfying @xmath780 , let @xmath787 be morphisms of @xmath129 satisfying @xmath782 , let @xmath398 be an object of @xmath17 , let @xmath564 , @xmath788 be objects of @xmath789 , let @xmath790 be a 1-morphism of @xmath17 and let @xmath791 and @xmath792 be 2-morphisms of @xmath17 .
if @xmath556 satisfies the skein axiom then @xmath793 = \left [ { \mathbf{m}}''_{{\mathbf{\gamma } } } \ast \left ( { \mathrm{id}}_{{\mathbf{\sigma } } ' } \circ ( { \mathbf{d}}^2 \times { \mathbf{i}})_{{\mathrm{t}}'^{\varphi ' } } \right ) \ast { \mathbf{m}}_{{\mathbf{\gamma } } } \right]\ ] ] as vectors of @xmath715 .
the proof of this lemma is analogous to the proof of lemma [ graded_surgery_equivalence ] .
let @xmath794 be the 1-morphism of @xmath17 given by @xmath795 where @xmath609 is given by @xmath610 , where @xmath796 is the cohomology class determined by @xmath797 and @xmath798 for the oriented longitude @xmath615 and meridian @xmath799 and where the lagrangian @xmath616 is generated by the homology class of @xmath617 .
the only difference between @xmath800 and the @xmath63-colored index 2 surgery 1-morphism @xmath801 is that the @xmath1-coloring @xmath802 of @xmath801 is given by a cohomology class satisfying @xmath613 and @xmath612 .
let @xmath803 be the 2-morphism of @xmath17 given by @xmath804 where @xmath805 is the only @xmath1-coloring on @xmath806 extending @xmath807 let @xmath808 be the 2-morphism of @xmath17 given by @xmath809 where @xmath810 is the framed link depicted in figure [ projective_trick ] and where @xmath811 is the only @xmath1-coloring on @xmath812 extending @xmath807 which is compatible with the @xmath4-coloring on @xmath721 .
[ projective_trick ] .
the framed link @xmath721 is represented by the two blue knots @xmath813 and @xmath814 with blackboard framing .
the cohomology class @xmath811 evaluates the oriented red longitude to @xmath63 and the oriented red meridian to 0.,title="fig : " ] [ projective_trick_lemma ] if @xmath556 is a quantum invariant on @xmath550 which satisfies the surgery axioms then @xmath815 = \lambda_{2,g } \lambda_{2,-g } \cdot \left [ ( { \mathbf{s}}^1 \times \overline{{\mathbf{d}}^2})_{{\mathrm{l}},g } \right]\ ] ] as vectors in @xmath816 .
surgery along @xmath721 yields a cobordism with corners @xmath817 which is isomorphic with @xmath818 because surgery along @xmath814 cancels the effects of surgery along @xmath813 .
[ same_color_everywhere ] let @xmath556 be a quantum invariant on @xmath550 wich satisfies the surgery axioms and the skein axiom .
let @xmath398 be an object of @xmath17 , let @xmath711 be objects of @xmath559 , let @xmath562 be a 2-morphism of @xmath17 of the form @xmath819 and let @xmath253 be a generic index .
then the vector @xmath820 $ ] of @xmath715 can be represented by a linear combination @xmath821\ ] ] where every @xmath4-colored ribbon graph @xmath822 contains an edge colored with @xmath599 in every connected component of @xmath823 .
since the following proof can be carried out in each connected component of @xmath823 separately , let us assume that @xmath823 is connected .
let us start by supposing that @xmath8 contains a projective edge @xmath88 colored with a projective object @xmath266 of @xmath4 .
then , thanks to the skein axiom , we can replace a small portion of @xmath88 with an embedded copy of the @xmath4-colored ribbon graph @xmath824 represented in figure [ projective_trick_2 ] for any section . ]
@xmath825 of @xmath826 .
[ projective_trick_2 ] -colored ribbon graph @xmath824.,title="fig : " ] let us now suppose that @xmath8 does not contain a projective edge . then , thanks to the admissibility condition , there exists some generic oriented knot @xmath827 satisfying @xmath516 .
now lemma [ projective_trick_lemma ] enables us to produce a @xmath4-colored framed link inside a tubular neighborhood of @xmath828 whose every component is projective , and we reduce the previous case .
if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath74 and @xmath81 are @xmath1-homogeneous @xmath4-colored ribbon sets and if @xmath339 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 which are compatible with @xmath64 and @xmath85 respectively then a _ @xmath4-skein inside @xmath80 relative to @xmath829 and @xmath830 _ is an equivalence class of pairs @xmath831 where @xmath112 is a @xmath1-homogeneous @xmath4-colored ribbon graph from @xmath113 to @xmath114 and where @xmath9 is a @xmath1-coloring of @xmath362 extending @xmath339 and @xmath360 which is compatible with @xmath115 .
two pairs @xmath831 and @xmath832 are equivalent if there exists an isomorphism of cobordisms with corners @xmath833 isotopic to the identity relative to @xmath834 satisfying @xmath420 and @xmath421 .
we denote with @xmath835{cals.pdf}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))$ ] the free complex vector space generated by @xmath4-skeins inside @xmath80 relative to @xmath829 and @xmath830 .
our notation for @xmath4-skeins will be a little abusive since we will not distinguish between an equivalence class of @xmath4-colored ribbon graphs and any of its representatives . if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath74 and @xmath81 are @xmath1-homogeneous @xmath4-colored ribbon sets and if @xmath339 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 which are compatible with @xmath64 and @xmath85 respectively then a @xmath4-skein @xmath831 inside @xmath80 relative to @xmath829 and @xmath830 is _ admissible _ if @xmath397 is an admissible @xmath393-coloring of @xmath362 .
we denote with @xmath836{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))$ ] the free complex vector space generated by admissible @xmath4-skeins inside @xmath80 relative to @xmath829 and @xmath830 .
[ graded_connection ] let @xmath398 be an object of @xmath17 , let @xmath749 be objects of @xmath559 and let @xmath750 be a non - empty connected 3-dimensional cobordism with corners from @xmath751 to @xmath752 . if @xmath556 is a quantum invariant on @xmath550 wich satisfies the surgery axioms and the skein axiom then the natural linear map @xmath837{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}}_{\vargamma};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}''^{v''},\vartheta''_{b '' } ) ) & \rightarrow & { \mathrm{hom}}_{\lambda_{z}({\mathbf{\gamma } } ) } ( { \mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma } } } ) \\ & ( { \mathrm{t}}^{\varphi},\omega ) & \mapsto & [ { \mathrm{m}}_{\vargamma},{\mathrm{t}}^{\varphi},\omega,0 ] \end{array}\ ] ] is surjective .
we have to show that for every morphism @xmath838 \in { \mathrm{hom}}_{\lambda_{z}({\mathbf{\gamma}})}({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}})\ ] ] there exist some ribbon graphs @xmath839 together with some admissible @xmath393-colorings @xmath530 of @xmath840 for @xmath841 such that @xmath842 = [ { \mathrm{m}}''_{\vargamma},{\mathrm{t}}''^{\varphi''},\omega'',n''].\ ] ] first of all we can suppose that @xmath843 is connected : indeed if it is not then , thanks to lemma [ same_color_everywhere ] , we can choose some @xmath253 and we can suppose that every connected component of @xmath843 contains an edge of @xmath844 which is colored with @xmath599 .
then , thanks to lemma [ graded_surgery_equivalence ] , a finite sequence of @xmath63-colored index 1 surgeries on @xmath845 connecting its components will determine a vector of @xmath715 which is a non - zero scalar multiple of @xmath846 $ ] .
now thanks to the surgery lemma [ computable ] and to remark [ graded_surgery_remark ] we know there exists a computable framed link @xmath847 together with some integer @xmath416 such that @xmath848 & = \lambda_{2,g_1 } \cdots \lambda_{2,g_k } \cdot [ { \mathrm{m}}_{\vargamma},{\mathrm{l}}^{\omega '' } \cup f_{{\mathrm{l}}}^{-1}({\mathrm{t}}''^{\varphi''}),f_{{\mathrm{l}}}^{\ # } \omega'',n ] \\ & = \lambda_{2,g_1 } \cdots \lambda_{2,g_k } \kappa^n \cdot [ { \mathrm{m}}_{\vargamma},{\mathrm{l}}^{\omega '' } \cup f_{{\mathrm{l}}}^{-1}({\mathrm{t}}''^{\varphi''}),f_{{\mathrm{l}}}^{\ # } \omega'',0 ] .
\qedhere \end{aligned}\ ] ] at this point in the construction the contravariant universal algebroids divert fom the covariant ones .
indeed the admissibility condition is equivalent to the strong admissibility condition for all objects and morphisms in covariant universal algebroids , while in the contravariant case the two properties differ .
this discrepancy is not visible in lemmas [ graded_surgery_equivalence ] , [ evocation_equivalence ] and [ projective_trick_lemma ] , which can be directly translated into analogous results for the corresponding contravariant universal algebroids . on the other hand , lemmas [ same_color_everywhere ] an [ graded_connection ]
require stronger hypotheses .
[ same_color_everywhere_prime ] let @xmath556 be a quantum invariant on @xmath550 satisfying the surgery axioms and the skein axiom .
let @xmath398 be an object of @xmath17 , let @xmath849 be objects of @xmath850 , let @xmath851 be a strongly admissible 2-morphism of @xmath17 of the form @xmath852 and let @xmath253 be a generic index .
then the vector @xmath853 $ ] of @xmath854 can be represented by a linear combination @xmath855\ ] ] where every @xmath4-colored ribbon graph @xmath856 contains an edge colored with @xmath599 in every connected component of @xmath857 .
[ graded_connection_prime ] let @xmath398 be an object of @xmath17 , let @xmath858 be strongly admissible objects of @xmath850 and let @xmath859 be a non - empty connected 3-dimensional cobordism with corners from @xmath860 to @xmath861 . if @xmath556 satisfies the surgery axioms and the skein axiom then the natural linear map @xmath862{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}}'_{\vargamma};({\mathrm{p}}'^{v'},\vartheta'_{b'}),({\mathrm{p}}'''^{v'''},\vartheta'''_{b ' '' } ) ) & \rightarrow & { \mathrm{hom}}_{\lambda'_{z}({\mathbf{\gamma } } ) } ( { \mathbf{\sigma}}'_{{\mathbf{\gamma}}},{\mathbf{\sigma}}'''_{{\mathbf{\gamma } } } ) \\ & ( { \mathrm{t}}'^{\varphi'},\omega ' ) & \mapsto & [ { \mathrm{m}}'_{\vargamma},{\mathrm{t}}'^{\varphi'},\omega',0 ] \end{array}\ ] ] is surjective .
the stronger hypotheses introduced in lemmas [ same_color_everywhere_prime ] and [ graded_connection_prime ] allow for the exact same proofs that were given in the covariant context .
in this section we recall the definitions for the geer - patureau - turaev renormalized invariants of admissible closed @xmath4-colored ribbon graphs , which were first introduced in @xcite , and for the costantino - geer - patureau quantum invariants of admissible closed decorated 3-dimensional manifolds , which were first constructed in @xcite via surgery presentations , and we show that the latter satisfy the surgery axioms introduced earlier .
a closed morphism @xmath863 of @xmath129 is _ admissible _ if the closed 2-morphism @xmath864 is admissible for the unique @xmath1-coloring @xmath776 of @xmath865 which is compatible with the @xmath4-coloring @xmath115 of @xmath84 .
we denote with @xmath866 the set of admissible closed morphisms of @xmath129 .
recall we confuse closed ribbon graphs contained in @xmath165 with ribbon graphs contained in @xmath867 .
if @xmath863 is an admissible closed morphism of @xmath129 and if @xmath89 is a projective edge with respect to @xmath115 then a morphism @xmath868 of @xmath129 is _ obtained from @xmath8 by cutting it open at @xmath88 _ if @xmath869 the _ geer - patureau - turaev renormalized invariant @xmath870 of admissible closed @xmath4-colored ribbon graphs _ is the function @xmath871 where @xmath89 is any projective edge with respect to @xmath115 , where @xmath872 is obtained from @xmath8 by cutting it open at @xmath88 and where @xmath209 is the unique trace on @xmath197 satisfying @xmath873 for the specified ambidextrous projective object @xmath257 of @xmath4 .
the geer - patureau - turaev renormalized invariant @xmath870 of admissible closed @xmath4-colored ribbon graphs is well - defined .
a proof of this result , which consists in showing that the definition of @xmath870 does not depend on the choice of the projective edge @xmath88 where to cut open @xmath84 , can be found in @xcite .
if @xmath4 satisfyies the non - degeneracy condition @xmath874 for the parameters @xmath296 and @xmath297 given in remark [ graded_delta_def ] then the _ costantino - geer - patureau quantum invariant associated with @xmath4 _ is the unique quantum invariant on @xmath550 mapping each closed connected 2-morphism @xmath540 of @xmath17 to the complex number @xmath875 where : 1 .
@xmath876 is a computable surgery framed link for @xmath544 in @xmath867 ; 2 .
@xmath877 is the identification of the exteriors of @xmath721 ; 3 .
@xmath878 is the signature of @xmath721 ; 4 .
@xmath879 is a choice of a square root of @xmath303 ; 5 .
@xmath880 . a proof of the fact that @xmath19 is a well - defined invariant on @xmath550 is contained in @xcite .
the costantino - geer - patureau quantum invariant @xmath19 satisfies the skein axiom .
the result is a direct consequence of the definition of the geer - patureau - turaev renormalized invariant @xmath870 of admissible closed @xmath4-colored ribbon graphs in terms of the reshetikhin - turaev functor @xmath187 .
if @xmath4 is modular relative to @xmath5 then the costantino - geer - patureau quantum invariant @xmath19 satisfies the surgery axioms with @xmath881 and with @xmath882 for every @xmath253 .
moreover @xmath883 .
the 0-surgery axiom holds for every generic @xmath253 with @xmath884 thanks to the trivial computation @xmath885 the 1-surgery axiom holds for every generic @xmath253 with @xmath886 thanks to two easy computations .
first of all , using the notation of remark [ first_surgery_remark ] , we have @xmath887 as follows from the evaluation of the geer - patureau - turaev renormalized invariant @xmath870 against the admissible framed @xmath4-colored link @xmath888 depicted in figure [ index_2_surgery_link_2 ] , which gives @xmath889 thanks to the relative modularity condition of definition [ relative_modular_graded_categories_def ] .
now the claim follows from the fact that if @xmath890 are morphisms of @xmath129 such that both @xmath891 and @xmath892 are admissible closed ribbon graphs then we have the equality @xmath893 [ index_2_surgery_link_2 ] .
the two components colored with @xmath894 form a computable surgery framed link for @xmath895 inside @xmath867 whose signature is equal to 1 , while the two components colored with @xmath599 are decorations.,title="fig : " ] finally the 2-surgery axiom holds for every generic @xmath253 with @xmath882 .
indeed if @xmath876 is a computable surgery framed link for @xmath896 and if we set @xmath897 then @xmath898 is a computable surgery framed link for @xmath899 and we obtain equalities @xmath900 as a corollary lemma [ graded_connection ] applies to @xmath19 whenever @xmath4 is relative modular . the only quantum invariant we will consider from now on is @xmath19 . we will therefore adopt a lighter notation for the extended universal construction : instead of using the full subscript @xmath19 for universal algebroids , linear functors and natural transformations we will simply refer to the relative modular category @xmath4 .
in this section we define admissible skein modules which project onto vector spaces of morphisms of universal algebroids associated with @xmath19 , much in the spirit of @xcite and @xcite .
from now we will suppose that the pre - modular @xmath1-category @xmath4 relative to @xmath5 we fixed in remark [ pre - modular_from_now_on ] is actually modular relative to @xmath5 .
in particular @xmath19 satisfies the surgery axioms .
if @xmath411 are 1-morphisms of @xmath26 then we denote with @xmath901 the complex vector space generated by the set of 2-morphisms @xmath410 of @xmath26 . then we denote with @xmath902 the subspace of @xmath901 generated by vectors of the form @xmath903 for some 1-morphism @xmath904 of @xmath26 , for some 2-morphisms @xmath905 and @xmath906 of @xmath26 , for some objects @xmath907 of @xmath129 satisfying the equality @xmath774 and for some morphisms @xmath908 of @xmath136 satisfying the equality @xmath909 as vectors of @xmath910 .
let @xmath411 be 1-morphisms of @xmath26 of the form @xmath911 and let @xmath80 be a 3-dimensional cobordism with corners from @xmath67 to @xmath75 .
let @xmath912 denote the subspace of @xmath835{cals.pdf}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))$ ] given by the preimage of the subspace @xmath902 of @xmath901 under the natural linear map @xmath913{cals.pdf}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b ' } ) ) & \rightarrow & { \mathcal{v}}({\mathbf{\sigma}},{\mathbf{\sigma } } ' ) \\ & ( { \mathrm{t}}^{\varphi},\omega ) & \mapsto & ( { \mathrm{m}},{\mathrm{t}}^{\varphi},\omega,0 ) .
\end{array}\ ] ] a vector of @xmath914 will be called a _
@xmath4-skein relation_. we also denote with @xmath915 the subspace of @xmath835{cals.pdf}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))$ ] generated by vectors of the form @xmath916 for some @xmath242 and some framed knot @xmath917 disjoint from @xmath84 and colored with @xmath918 , where @xmath919 is a parallel copy of @xmath920 determined by the framing and where @xmath457 is the inclusion of pairs @xmath921 a vector of @xmath922 will be called a _
@xmath234-skein relation_. if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath74 and @xmath81 are @xmath1-homogeneous @xmath4-colored ribbon sets and if @xmath339 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 which are compatible with @xmath64 and @xmath85 respectively then the _ skein module of @xmath80 relative to @xmath829 and @xmath830 _ is the complex vector space @xmath923 defined as the quotient of @xmath924{cals.pdf}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))\ ] ] with respect to @xmath925 two @xmath4-skeins in @xmath835{cals.pdf}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))$ ] are said to be _ skein equivalent _ if their difference is a sum of a @xmath4-skein relation with a @xmath234-skein relation .
we also consider the admissible versions of the vector spaces of @xmath4-skein relations and @xmath234-skein relations .
namely , we define @xmath926 to be the intersection @xmath927{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))\ ] ] and we define @xmath928 to be the intersection @xmath929{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'})).\ ] ] if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 , if @xmath74 and @xmath81 are @xmath1-homogeneous @xmath4-colored ribbon sets and if @xmath339 and @xmath360 are @xmath1-colorings of @xmath340 and @xmath361 which are compatible with @xmath64 and @xmath85 respectively then the _ admissible skein module of @xmath80 relative to @xmath829 and @xmath830 _ is the complex vector space @xmath930 defined as the quotient of @xmath931{cals.pdf}}}_{{\mathcal c}}({\mathrm{m}};({\mathrm{p}}^v,\vartheta_b),({\mathrm{p}}'^{v'},\vartheta'_{b'}))\ ] ] with respect to @xmath932 [ admissible_skein ] let @xmath398 be an object of @xmath17 , let @xmath749 be objects of @xmath933 and let @xmath750 be a cobordism with corners from @xmath751 to @xmath752 .
the linear map @xmath934 \end{array}\ ] ] is well - defined .
first of all , if @xmath935 is a @xmath4-skein relation in @xmath836{cals.pdf}}}({\mathrm{m}}_{\vargamma};{\mathrm{p}}^v,{\mathrm{p}}''^{v''})$ ] then we have to show that @xmath936 for the natural transformations @xmath937 coming from the extended universal construction ( see appendix [ extended_universal_construction ] for a definition ) .
in particular , it suffices to show that @xmath19 vanishes on @xmath938 .
this is true thanks to the very definition of @xmath19 in terms of the geer - patureau - turaev renormalized invariant @xmath870 of admissible closed @xmath4-colored ribbon graphs , which is itself defined in terms of the reshetikhin - turaev functor @xmath187
. now if @xmath831 is an admissible @xmath4-skein in @xmath750 and if @xmath939 is a framed knot disjoint from @xmath84 then we have to show that @xmath940 this follows from the definition of @xmath870 in terms of @xmath187 and from the second and third condition in definition [ relative_pre - modular_graded_categories_def ] .
the same result holds also for the contravariant universal algebroids associated with @xmath19 .
in other words , vectors in morphism spaces of contravariant universal algebroids are invariant under skein equivalence just like vectors in morphism spaces of covariant universal algebroids .
we use here the results of the previous sections in order to give a better description for covariant universal algebroids associated with the costantino - geer - patureau invariant , to which we will confine from now on .
in particular , we give for every covariant universal algebroid an explicit description of a dominating set .
this will allow for a careful analysis in the following sections of the obstruction for the monoidality of extended quantization functors . from now on we will fix a choice , for every critical @xmath941 , of a generic index @xmath942 satisfying @xmath943 .
such a choice exists because @xmath3 is small symmetric in @xmath1 .
we also introduce the short notation @xmath944 for @xmath945 , where @xmath946 is the fixed projective simple object of index @xmath947 specified in remark [ pre - modular_from_now_on ] .
[ critical_fusion ] if @xmath941 then @xmath948 is dominated by the set @xmath949 in other words if @xmath64 is a projective object of @xmath950 then we can always find objects @xmath951 , elements @xmath952 and morphisms @xmath953 satisfying @xmath954 . indeed , since @xmath64 is projective , there exists . ] a section @xmath955 of @xmath956 . then , since @xmath957 has index @xmath958 and since @xmath959 is semisimple and dominated by @xmath960 , we can conclude .
let us fix an object @xmath408 of @xmath17 .
every cobordism @xmath961 from @xmath140 to @xmath329 whose @xmath103-th connected component @xmath962 is given by a cobordism from @xmath140 to @xmath333 induces a decomposition @xmath963 and the _ index @xmath964 of @xmath962 with respect to @xmath398 _ is defined as the element of @xmath1 given by @xmath965 \rangle$ ] for the fundamental class @xmath966 $ ] of @xmath333 and for the inclusion @xmath967 if @xmath968 then @xmath962 is said to be a _ generic component of @xmath751 with respect to @xmath398 _ while if @xmath969 then @xmath962 is said to be a _ critical component of @xmath751 with respect to @xmath398_. let @xmath751 be a cobordism from @xmath140 to @xmath329 .
a _ fundamental ribbon set @xmath69 for @xmath751 with respect to @xmath398 _ is a ribbon set inside @xmath751 decomposed as a disjoint union @xmath970 where :
@xmath971 is given by a single negative ribbon vertex in every critical component of @xmath751 with respect to @xmath398 ; 2 .
@xmath972 is given by a single positive ribbon vertex in every component of @xmath751 ; 3 .
@xmath973 is given by a single positive ribbon vertex in every component of @xmath751 .
let @xmath69 be a fundamental ribbon set for @xmath751 with respect to @xmath398 .
if @xmath974 is contained in the connected component @xmath962 of @xmath751 then the _ index @xmath975 of @xmath976 with respect to @xmath398 _ is defined as the element of @xmath3 given by @xmath969 .
a _ fundamental @xmath393-coloring @xmath522 of @xmath977 extending @xmath330 _ is a @xmath393-coloring satisfying : 1 .
@xmath978 for every @xmath979 ; 2 .
@xmath980 for every @xmath981 ; 3 .
@xmath982 for every @xmath983 . the compatibility between @xmath64 and @xmath339 together with the assumption that @xmath943 for every @xmath941 guarantees the genericity of the index of @xmath984 for every @xmath983 .
we denote with @xmath985 the set of fundamental @xmath393-colorings of @xmath977 extending @xmath330 .
if @xmath69 is a fundamental ribbon set for @xmath751 and if we have a fixed choice for a base set @xmath986 and for a lagrangian @xmath987 then we denote with @xmath988 the object of @xmath933 given by @xmath989 for every fundamental @xmath393-coloring @xmath394 of @xmath977 extending @xmath330 .
[ graded_generators ] let @xmath408 be an object of @xmath17 , let @xmath751 be a non - empty cobordism from @xmath140 to @xmath329 , let @xmath69 be a fundamental ribbon set for @xmath751 with respect to @xmath398 , let @xmath990 be a base set for @xmath751 and let @xmath987 be a lagrangian
. then @xmath991 dominates @xmath933 .
let @xmath992 and @xmath993{fourth_derivative.pdf}}}_{{\mathbf{\gamma } } } = ( \sigma^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{\vargamma},{{\mathrm{p}}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}},\vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } , { \mathcal l}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}})$ ] be objects of @xmath933 .
if @xmath750 is a connected cobordism with corners from @xmath752 to @xmath751 and if @xmath843 is a connected cobordism with corners from @xmath751 to @xmath994{fourth_derivative.pdf}}}_{\vargamma}$ ] then , thanks to the connection lemma [ graded_connection ] , every morphism inside @xmath995{fourth_derivative.pdf}}}_{{\mathbf{\gamma}}})$ ] can be described by some linear combination of admissible @xmath4-skeins inside @xmath996 relative to @xmath997{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}},\vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}}).\ ] ] but every such @xmath4-skein can be written , up to isotopy and skein equivalence , as a linear combination of admissible @xmath4-skeins whose @xmath4-colored ribbon graphs meet @xmath998 transversely in @xmath69 inducing a fundamental @xmath393-coloring of @xmath977 .
this follows from the semisimplicity of @xmath999 for generic components @xmath962 of @xmath751 with respect to @xmath398 , and it follows from remark [ critical_fusion ] for critical components @xmath1000 of @xmath751 with respect to @xmath398 . in other words
every morphism @xmath1001 \in { \mathrm{hom}}_{\lambda_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}''_{{\mathbf{\gamma}}},{\mathbf{\sigma}}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{{\mathbf{\gamma}}})\ ] ] can be decomposed as @xmath1002\ ] ] for some fundamental @xmath393-colorings @xmath1003 of @xmath977 , for some admissible @xmath4-skein @xmath1004 inside @xmath750 relative to @xmath1005 and for some admissible @xmath4-skein @xmath1006 inside @xmath843 relative to @xmath1007{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}},\vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}})\ ] ] for every @xmath841 .
if @xmath69 is a ribbon graph inside a 2-dimensional cobordism @xmath67 from @xmath329 to @xmath336 then we will denote with @xmath1008 the ribbon graph inside the cobordism @xmath1009 from @xmath336 to @xmath329 whose oriented vertex set @xmath1010 is obtained from @xmath1011 by reversing the orientation of every vertex and whose framing at @xmath1012 is given by the framing of @xmath69 at @xmath72 for every @xmath1013 .
if @xmath394 is a @xmath393-coloring of @xmath340 then @xmath1014 will denote the @xmath393-coloring of @xmath1015 defined by @xmath1016 for every @xmath1013 and by @xmath1017 .
let @xmath408 be an object of @xmath17 and let @xmath860 be a cobordism from @xmath329 to @xmath140 .
a connected component @xmath1018 of @xmath860 is _ generic with respect to @xmath398 _ if @xmath1019 is generic with respect to @xmath398 and it is _ critical with respect to @xmath398 _ if @xmath1019 is critical with respect to @xmath398 .
a ribbon set @xmath1020 is a _ fundamental ribbon set for @xmath860 with respect to @xmath398 _ if @xmath1021 is a fundamental ribbon set for @xmath1022 .
let @xmath83 be a fundamental ribbon set for @xmath860 with respect to @xmath398 .
a _ fundamental @xmath393-coloring @xmath1023 of @xmath1024 extending @xmath330 _ is a @xmath393-coloring such that @xmath1025 is a fundamental @xmath393-coloring of @xmath1026 .
we denote with @xmath1027 the set of fundamental @xmath393-colorings of @xmath1024 extending @xmath330 .
if @xmath83 is a fundamental ribbon set for @xmath860 and if we have a fixed choice for a base set @xmath1028 and for a lagrangian @xmath1029 then we denote with @xmath1030 the object of @xmath1031 the 1-morphism of @xmath17 given by @xmath1032 for every fundamental @xmath393-coloring @xmath1023 of @xmath1024 extending @xmath330 .
[ graded_triviality ] let @xmath398 be an object of @xmath17 , let @xmath1033 be objects of @xmath933 and let @xmath1034 be 2-morphisms of @xmath17 .
let @xmath860 be a non - empty cobordism from @xmath329 to @xmath140 , let @xmath83 be a fundamental ribbon set for @xmath860 with respect to @xmath398 , let @xmath1035 be a base set for @xmath860 , let @xmath1029 be a lagrangian , let @xmath80 be a connected cobordism with corners from @xmath140 to @xmath1036 and let @xmath117 be a connected cobordism with corners from @xmath1037 to @xmath140 .
then a linear combination @xmath1038 $ ] determines a trivial morphism in @xmath1039 if and only if the expression @xmath1040 vanishes for all fundamental @xmath393-colorings @xmath1023 of @xmath1024 extending @xmath330 , for all admissible @xmath4-skeins @xmath831 inside @xmath80 relative to @xmath1041 and for all @xmath4-skeins @xmath832 inside @xmath117 relative to @xmath1042 the vector @xmath1043 \in { \mathrm{hom}}_{\lambda_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}})\ ] ] is trivial if and only if @xmath1044 this happens if and only if we have the equality @xmath1045 for every object @xmath1046 of @xmath1031 and for all 2-morphisms @xmath1047 and @xmath1048 of @xmath17 of the form @xmath1049 thanks to lemma [ same_color_everywhere ] we can suppose that @xmath844 restricts to an admissible @xmath4-colored ribbon graph in every connected component of @xmath1050 . therefore , up to sliding projective arcs of @xmath844 through all connected components of @xmath1051
, we can suppose that @xmath1052 is strongly admissible .
then , just like in the proof of lemma [ graded_generators ] , we can use lemma [ graded_connection_prime ] , isotopy and skein equivalence to obtain a decomposition @xmath1053 = \sum_{j=1}^{\ell } \mu_j \cdot [ { \mathrm{m}}'''_{\vargamma},{\mathrm{t}}'''^{\varphi'''_j}_j,\omega'''_j,0 ] \ast [ { \mathrm{m}}'_{\vargamma},{\mathrm{t}}'^{\varphi'_j}_j,\omega'_j,0]\ ] ] for some connected cobordisms with corners @xmath859 from @xmath1054 to @xmath860 and @xmath1055 from @xmath860 to @xmath861 , and for some admissible @xmath4-skeins @xmath1056 inside @xmath859 relative to @xmath1057 and @xmath1058 inside @xmath1055 relative to @xmath1059 for all @xmath1060 .
now we can apply lemma [ graded_connection ] to @xmath1061 \ast [ { \mathbf{m}}'']\ ] ] and to @xmath1062 \ast [ ( { \mathrm{m}}'''_{\vargamma},{\mathrm{t}}'''^{\varphi'''_j}_j,\omega'''_j,0 ) \circ { \mathrm{id}}_{{\mathbf{\sigma}}''_{{\mathbf{\gamma}}}}]\ ] ] in order to conclude .
in this section we analyse the failure of @xmath28 at being an etqft .
we describe the obstruction to monoidality as @xmath2-suspension systems on universal algebroids .
this allows for the construction of @xmath2-graded extensions of these linear categories , which leads to the definition of the @xmath2-graded extended quantization functors @xmath1063 and @xmath1064 .
we begin by studying the universal algebroid @xmath1065 , which measures the obstruction to monoidality for @xmath28 .
[ zero_critico ] we suppose from now on that the element @xmath1066 is critical .
indeed , should 0 be generic , we could choose @xmath38 as the specified projective ambidextrous object of @xmath4 .
then we would have @xmath309 and , since @xmath310 , the geer - patureau - turaev renormalized invariant @xmath870 would be equivalent to @xmath187 thanks to corollary 17 of @xcite .
then @xmath19 would boil down to witten - reshetikhin - turaev with @xmath1-structure .
in particular the construction could be carried on also without this assumption , but since it would force us to use two different notations for the rest of the exposition while producing just a very mild generalization of what was already well - known , we prefer to directly concentrate on the new situation .
let @xmath1067 denote the fundamental ribbon set for @xmath1068 given by @xmath1069 , by @xmath1070 and by @xmath1071 with framing given by @xmath1072 for the curve @xmath1073 consider the base set @xmath1074 and let @xmath1075 be the cohomology class in @xmath1076 given by @xmath1077 for positive meridians @xmath1078 of @xmath1079 for @xmath1080 .
for every @xmath242 and for every @xmath1081 we denote with @xmath1082 the fundamental @xmath393-coloring of @xmath1083 determined by @xmath1084 and by @xmath1085 , and we will simply denote with @xmath1086 the object of @xmath1065 given by @xmath1087 such an object will be called a
_ @xmath1088-colored 2-sphere_. [ spheres_figure ] -colored 2-sphere @xmath1086.,title="fig : " ] [ spheres_lemma ] for all @xmath1089 and for all @xmath1090 we have @xmath1091 thanks to lemma [ graded_connection ] we know that every morphism in @xmath1092 can be represented by some admissible @xmath4-skein inside the trivial cobordism with corners @xmath1093 relative to @xmath1094 up to isotopy and skein equivalence we can moreover restrict to admissible @xmath4-skeins of the form @xmath1095 where @xmath1096 consists of an oriented graph @xmath1097 contained in @xmath1098 featuring a single coupon @xmath91 with color @xmath1099 like the one represented in figure [ first_image_spheres_lemma_graded ] .
[ first_image_spheres_lemma_graded ] -colored ribbon graph @xmath1096 inside @xmath1093.,title="fig : " ] in other words we have a surjective linear map from @xmath1100 to @xmath1092 and we can study its kernel . in order to do
so we can specify @xmath1101 as a non - empty 2-dimensional cobordism from @xmath140 to @xmath140 , we can specify @xmath1102 as a 3-dimensional cobordism with corners from @xmath140 to @xmath1103 and we can specify @xmath1093 as a 3-dimensional cobordism with corners from @xmath1103 to @xmath140 .
then , thanks to lemma [ graded_triviality ] , the triviality of the vector @xmath1104 $ ] in @xmath1105 can be tested by choosing @xmath1106 as a fundamental ribbon set , @xmath1107 as a base set , @xmath1108 as a cohomology class , by considering all fundamental @xmath393-colorings @xmath1109 of @xmath1110 , by considering all admissible @xmath4-skeins @xmath832 inside @xmath1102 relative to @xmath1111 by considering all admissible @xmath4-skeins @xmath1112 inside @xmath1093 relative to @xmath1113 and by computing the costantino - geer - patureau invariant @xmath19 on the resulting closed 2-morphism of @xmath17 . up to isotopy , skein equivalence and multiplication by some non - zero scalar this amounts . ] to evaluating the geer - patureau - turaev renormalized invariant @xmath870 against the admissible closed @xmath4-colored ribbon graph @xmath1114 for all @xmath253 and for all @xmath1115 where @xmath1114 is given by the admissible closed ribbon graph of figure [ killing_fundamental_spheres ] .
[ killing_fundamental_spheres ] .,title="fig : " ] the middle @xmath918-colored strand can be pulled out on top of the closed component encircling it at the cost of a multiplicative factor @xmath1116 .
then , thanks to the relative modularity condition for @xmath4 , we get @xmath1117 for the admissible closed ribbon graph @xmath1118 depicted in figure [ reduction_fundamental_spheres ] .
[ reduction_fundamental_spheres ] .,title="fig : " ] the region enclosed by the dotted rectangle represents a morphism in @xmath1119 since this vector space is naturally isomorphic with @xmath1120 then the semisimplicity of @xmath1121 implies @xmath1122 the dimension of @xmath1123 is furthermore exactly equal to 1 because the non - triviality of @xmath1124 $ ] follows from the evaluation of @xmath870 against @xmath1118 for @xmath1125 , for @xmath1126 and for @xmath1127 [ precisazione ] in order to give a complete proof , we would also have to consider the case of a critical index @xmath941 for the closed component of the ribbon graph in figure [ killing_fundamental_spheres ] , which of course would determine a kirby color which is not defined .
the problem is that the chosen surgery presentation would no longer be computable , and this is a recurring phenomenon with relative pre - modular categories which makes arguments a little cumbersome , like in @xcite .
we will explain here how to adapt the proof given above to the critical case by passing to a computable surgery presentation , and this can be done every time an analgous situation shows up , so we will not give details again . by using the skein axiom we can produce a strand with color @xmath946 in a small region of the graph along the bottom left edge colored with @xmath1128 .
then we can change the index of the critical component by sliding the new arc over it .
if we take care of doing this by subtraction we obtain a computable surgery presentation because the index of the surgery knot is changed from @xmath1129 to @xmath958 , which is generic .
we can then pull the middle @xmath918-colored strand out on top of the diagram and the cost of doing this , in terms of multiplicative factors , is equal to @xmath1130 .
we can then use the relative modularity condition just like in the generic case , and remark that now the @xmath946-colored strand can be returned to its original position , so that the modification we performed on @xmath1131 can in fact be undone .
the rest of the proof is unchanged .
since we established that for every @xmath1132 and for every @xmath242 the @xmath1088-colored 2-sphere @xmath1086 is a zero object of @xmath1065 , we will switch from now on to the simpler notations @xmath1133 and we will refer to @xmath1134 as the _
@xmath1135-colored 2-sphere_. [ pi_non - triviale ] lemmas [ graded_generators ] and [ spheres_lemma ] immediately imply that @xmath1065 is morita equivalent to its full linear subcategory with set of objects @xmath1136 .
this is of course not morita equivalent to the unit algebroid @xmath22 unless @xmath1137 , in which case we can prove that @xmath1138 is monoidal .
therefore , as in remark [ zero_critico ] , we will suppose from now on that @xmath2 is not trivial , and thus that @xmath28 is not an etqft .
the next subsubsections are dedicated to the study of vector spaces of morphisms between tensor products of fundamental spheres in @xmath1065 .
the _ 0-colored 3-disc _
@xmath1139 is the 2-morphism of @xmath17 given by @xmath1140 where @xmath1141 is the @xmath4-colored ribbon graph contained in @xmath1142 depicted in figure [ discs_figure ] whose only coupon is colored with the invertible morphism @xmath235 given by the braided monoidal functor @xmath1143 and where @xmath1144 is the only @xmath1-coloring of @xmath1145 which is compatible with the @xmath4-coloring on @xmath1146 .
the _ inverse 0-colored 3-disc _ @xmath1147 is the 2-morphism of @xmath17 given by @xmath1148 where @xmath1149 is the @xmath4-colored ribbon graph contained in @xmath1150 and represented in figure [ discs_figure ] whose only coupon is colored with the invertible morphism @xmath1151 given by the braided monoidal functor @xmath1143 and where @xmath1152 is the only @xmath1-coloring of @xmath1153 which is compatible with the @xmath4-coloring on @xmath1154 .
[ discs_figure ] and the inverse 0-colored 3-disc @xmath1155 .
the ribbon graphs @xmath1156 and @xmath1157 are represented in blue with blackboard framing .
the arrows on the faces of the coupons specify the orientations of their horizontal boundaries and of their vertical boundaries .
the cohomology classes @xmath1144 and @xmath1152 are completely determined by the @xmath4-colorings on @xmath1156 and on @xmath1157 respectively.,title="fig : " ] [ index_1_consequence ] we have the equality @xmath1158 = \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^0 } \right]\ ] ] between vectors of @xmath1159 .
this result is an immediate consequence of the @xmath1160-colored index 1 surgery axiom .
since the equality @xmath1161 = \left [ { \mathrm{id}}_{{\mathrm{id}}_{\varnothing } } \right]\ ] ] follows immediately from the @xmath1160-colored index 0 surgery axiom , we have @xmath1162 = \left [ ( { \mathbf{d}}^3)^0 \right]^{-1 } \in { \mathrm{hom}}_{\lambda_{{\mathcal c}}(\varnothing)}(({\mathbf{s}}^2)^0,{\mathrm{id}}_{\varnothing})\ ] ] and the objects @xmath547 and @xmath1163 of @xmath1065 are isomorphic .
we define the _ 3-pant cobordism . ]
@xmath1164 _ as the 3-dimensional cobordism from @xmath1165 to @xmath1068 whose support @xmath1166 is given by @xmath1167 where @xmath1168 is the open ball of center @xmath1169 and radius @xmath1170 in @xmath1171 , whose incoming horizontal boundary identification is given by @xmath1172 and whose outgoing horizontal boundary identification is given by @xmath1173 .
we define the _
inverse 3-pant cobordism @xmath1174 _ as the 3-dimensional cobordism from @xmath1068 to @xmath1165 whose support is given by @xmath1175 , whose incoming horizontal boundary identification is given by @xmath1173 and whose outgoing horizontal boundary identification is given by @xmath1176 .
let us consider @xmath232 .
the _ @xmath228-colored 3-pant _
@xmath1177 is the 2-morphism of @xmath17 given by @xmath1178 where @xmath1179 is the @xmath4-colored ribbon graph contained in @xmath1180 and represented in figure [ pants_figure ] , whose only coupon is colored with the invertible morphism @xmath1181 given by the braided monoidal functor @xmath1143 , and where @xmath1182 is the unique @xmath1-coloring of @xmath1183 which is compatible with the @xmath4-coloring of @xmath1184 and which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1185 .
[ pants_figure ] -colored 3-pant @xmath1186 .
the ribbon graph @xmath1187 is given by the three blue edges together with the middle blue trivalent graph with blackboard framing .
the arrows on the faces of the coupon specify the orientation of its horizontal boundary and of its vertical boundary .
the cohomology class @xmath1182 is completely determined by the @xmath4-coloring on @xmath1187 and by the vanishing condition for the evaluations against relative homology classes contained in the lower hemisphere.,title="fig : " ] the _ inverse @xmath228-colored 3-pant _
@xmath1188 is the 2-morphism of @xmath17 given by @xmath1189 where @xmath1190 is the @xmath4-colored ribbon graph contained in @xmath1191 and represented in figure [ inverse_pants_figure ] , whose only coupon is colored with the invertible morphism @xmath1192 given by the braided monoidal functor @xmath1143 , and where @xmath1193 is the only @xmath1-coloring of @xmath1194 which is compatible with the @xmath4-coloring on @xmath1195 and which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1196 .
[ inverse_pants_figure ] -colored 3-pant @xmath1197 .
the ribbon graph @xmath1198 is given by the three blue edges together with the middle blue trivalent graph with blackboard framing .
the arrows on the faces of the coupon specify the orientation of its horizontal boundary and of its vertical boundary .
the cohomology class @xmath1193 is completely determined by the @xmath4-coloring on @xmath1198 and by the vanishing condition for the evaluations against relative homology classes contained in the lower hemisphere.,title="fig : " ] [ pants_lemma ] for all @xmath1199 we have @xmath1200 thanks to lemma [ graded_connection ] every morphism in @xmath1201 can be represented by some admissible @xmath4-skein inside the 3-pant cobordism @xmath1164 relative to @xmath1202 and @xmath1203 .
up to isotopy and skein equivalence , just like for lemma [ spheres_lemma ] , we have a surjective linear map from @xmath1204 to @xmath1205 and we have to study its kernel . in order to do so we can specify @xmath1101 as a non - empty 2-dimensional cobordism from @xmath140 to @xmath140 , we can specify @xmath1174 as a 3-dimensional cobordism from @xmath140 to @xmath1206 and we can specify @xmath1093 as a 3-dimensional cobordism from @xmath1103 to @xmath140 . then the triviality of the image of a morphism @xmath1207 in @xmath1208 can be tested by evaluating the geer - patureau - turaev renormalized invariant @xmath870 against the admissible closed @xmath4-colored ribbon graph @xmath1118 depicted in figure [ reduction_pants ] for every @xmath1209 , for every @xmath1210 and for all @xmath1211 [ reduction_pants ] .,title="fig : " ] as before , the region enclosed by the dotted rectangle represents a morphism in @xmath1212 . since this vector space is naturally isomorphic with @xmath1213 the semisimplicity of @xmath1121 implies @xmath1214 the dimension of @xmath1215 is furthermore exactly 1 because the non - triviality of @xmath1216 $ ] follows from the evaluation of @xmath870 against @xmath1118 for @xmath1217 , for @xmath1218 and for @xmath1219 given by the images under the reshetikhin - turaev functor @xmath187 of the @xmath4-colored ribbon graphs depicted in figure [ non - zero_pants ] .
[ non - zero_pants ] -colored ribbon graphs representing @xmath205 , @xmath1220 and @xmath1221.,title="fig : " ] the proof of the second statement is completely analogous to the proof of the first one .
the morphism space @xmath1222 is generated by @xmath1216 $ ] for all @xmath232 and @xmath1223 is generated by @xmath1224 $ ] for all @xmath1225 .
[ pants_composition_1 ] for all @xmath232 we have the equality @xmath1226 = { \mathcal{d}}{\mathrm{d}}_0^{-1 } \cdot \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^u } \otimes { \mathrm{id}}_{({\mathbf{s}}^2)^{u ' } } \right]\ ] ] between vectors of @xmath1227 .
figure [ pants_proof ] depicts two skein equivalent 2-morphisms of @xmath17 representing the vector @xmath1226 \in { \mathrm{hom}}_{\lambda_{{\mathcal c}}(\varnothing)}(({\mathbf{s}}^2)^u \otimes ( { \mathbf{s}}^2)^{u'},({\mathbf{s}}^2)^u \otimes ( { \mathbf{s}}^2)^{u'}).\ ] ] then proof follows now from lemma [ graded_surgery_equivalence ] .
[ pants_proof ] of @xmath17 is realized as an admissible cobordism with corners whose support is a submanifold of @xmath1228 .
the incoming horizontal boundary is given by the two innermost spheres while the ougoing horizontal boundary is given by the two outermost ones .
the leftmost and the rightmost blue edges of the @xmath4-colored ribbon graph meet at the infinity . in the bottom part
, the same 2-morphism is rearranged , up to skein equivalence , as a different vertical composition .
the unlimited region between the two exterior spheres can be interpreted as the @xmath1160-colored index 1 surgery belt 2-morphism @xmath1229.,title="fig : " ] [ pants_composition_2 ] for all @xmath232 we have the equality @xmath1230 = { \mathcal{d}}{\mathrm{d}}_0^{-1 } \cdot \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{u + u ' } } \right]\ ] ] between vectors of @xmath1231 .
thanks to remark [ pi_non - triviale ] we have @xmath1232 in particular @xmath1233 $ ] must be of the form @xmath1234 $ ] for some @xmath1235 . thanks to lemma [ pants_composition_1 ] we have the chain of equalities @xmath1236 & = { \mathcal{d}}^{-2 } { \mathrm{d}}_0 ^ 2 \cdot \left [ ( \overline{{\mathbf{p}}^3})^{u , u ' } \ast ( { \mathbf{p}}^3)^{u , u ' } \ast ( \overline{{\mathbf{p}}^3})^{u , u ' } \ast ( { \mathbf{p}}^3)^{u , u ' } \right ] \\ & = { \mathcal{d}}^{-2 } { \mathrm{d}}_0 ^ 2 \lambda \cdot \left [ ( \overline{{\mathbf{p}}^3})^{u , u ' } \ast { \mathrm{id}}_{({\mathbf{s}}^2)^{u+u ' } } \ast ( { \mathbf{p}}^3)^{u , u ' } \right ] \\ & = { \mathcal{d}}^{-2 } { \mathrm{d}}_0 ^ 2 \lambda \cdot \left [ ( \overline{{\mathbf{p}}^3})^{u , u ' } \ast ( { \mathbf{p}}^3)^{u , u ' } \right ] \\ & = { \mathcal{d}}^{-1 } { \mathrm{d}}_0 \lambda \cdot \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^u } \otimes { \mathrm{id}}_{({\mathbf{s}}^2)^{u ' } } \right ] \end{aligned}\ ] ] which implies @xmath1237 . lemmas
[ pants_composition_1 ] and [ pants_composition_2 ] combine to give @xmath1238 = \left [ ( { \mathbf{p}}^3)^{u , u ' } \right]^{-1 } \in { \mathrm{hom}}_{\lambda_{{\mathcal c}}(\varnothing)}(({\mathbf{s}}^2)^{u + u'},({\mathbf{s}}^2)^{u + u'}).\ ] ] [ unitarity_of_discs ] for every @xmath242 we have the equalities @xmath1239 = \left [ ( { \mathbf{p}}^3)^{u,0 } \ast \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \otimes ( { \mathbf{d}}^3)^{0 } \right ) \right ] = \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \right]\ ] ] between vectors of @xmath1240 and @xmath1241 = \left [ \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \otimes
( \overline{{\mathbf{d}}^3})^0 \right ) \ast ( \overline{{\mathbf{p}}^3})^{u,0 } \right ] = \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \right]\ ] ] between vectors of @xmath1240 .
the proof of this lemma follows from an obvious isomorphism of cobordisms with corners together with the equality @xmath1242 which is part of the definition of braided monoidal functor between braided monoidal categories . [ associativity_of_pants ] for all @xmath1243 we have the equalities @xmath1244 \\ = \ ; & \left [ ( { \mathbf{p}}^3)^{u , u ' + u '' } \ast \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \otimes ( { \mathbf{p}}^3)^{u',u '' } \right ) \right ] \end{aligned}\ ] ] between vectors of @xmath1245 and @xmath1246 \\ = \ ; & \left [ \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{u } } \otimes ( \overline{{\mathbf{p}}^3})^{u',u '' } \right ) \ast ( \overline{{\mathbf{p}}^3})^{u , u ' + u '' } \right ] \end{aligned}\ ] ] between vectors of @xmath1247 .
the proof of this lemma follows from an obvious isomorphism of cobordisms with corners together with the equality @xmath1248 which is part of the definition of braided monoidal functor between braided monoidal categories . for all @xmath232
the _ twisted @xmath228-colored 3-pant _ @xmath1249 is the 2-morphism of @xmath17 given by @xmath1250 [ twisted_pants_lemma ] for all @xmath232 we have the equality @xmath1251 = \nu(u , u ' ) \cdot \left [ ( { \mathbf{p}}^3)^{u , u ' } \right]\ ] ] between vectors of @xmath1252 .
for the proof of this lemma we adopt the notation used in the proof of lemma [ pants_lemma ] . we know that @xmath1253 $ ] generates the vector space @xmath1254 so there have to exist coefficients @xmath1255 for all @xmath232 satisfying @xmath1251 = \lambda_{u , u ' } \cdot \left [ ( { \mathbf{p}}^3)^{u , u ' } \right].\ ] ] in order to compute @xmath1256 we choose @xmath1174 as a cobordism from @xmath140 to @xmath1206 and @xmath1093 as a cobordism from @xmath1103 to @xmath140 .
now it is sufficient to compare the evaluations of the geer - patureau - turaev renormalized invariant @xmath870 against the admissible closed @xmath4-colored ribbon graphs @xmath1118 and @xmath1257 depicted in figure [ reduction_pants ] for @xmath1217 , for @xmath1218 and for @xmath1258 given by the images under the reshetikhin - turaev functor @xmath187 of the ribbon graphs depicted in figure [ non - zero_pants ] and [ non - zero_pants_tildef ] .
[ non - zero_pants_tildef ] -colored ribbon graph representing @xmath1259.,title="fig : " ] by applying twice the skein condition for relative modular categories we can pull the @xmath918-colored strand in figure [ non - zero_pants_tildef ] on top of all the @xmath1128-colored strands .
then the equality @xmath1260 coming from the definition of braided monoidal functor leads to the equality @xmath1261 which proves the assertion .
we sidetrack for a moment in order to define an operation we will apply to our extended quantization functors @xmath1138 and @xmath1262 .
this will allow for the definition of a symmetric monoidal functor at the cost of changing the target symmetric monoidal 2-category . if @xmath1263 is a linear category then the category @xmath1264 of linear functors from @xmath1263 to @xmath1263 together with their natural transformations is monoidal with unit given by @xmath1265 and with tensor product given by composition .
[ def_suspension_system ] a _ @xmath2-suspension system on a linear category @xmath1263 _ is a monoidal functor @xmath1266 .
if @xmath1267 is a @xmath2-suspension system on a linear category @xmath1263 then we denote with @xmath1268 the natural transformations coming from the structure of @xmath1267 and we denote with @xmath1269 the functor @xmath1270 , which is called the _
@xmath1135-suspension functor_. [ def_gradation ] if @xmath1263 is a linear category and if @xmath1267 is a @xmath2-suspension system on @xmath1263 then the _ @xmath2-graded extension of @xmath1263 along @xmath1267 _ is the @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] with set of objects @xmath1272{$\mathbb v$ } } } } ) : = { \mathrm{ob}}(\lambda),\ ] ] with vector spaces of degree @xmath1135 morphisms @xmath1273{\scriptsize $ \mathbb v$}}}}}^u(v , v ' ) : = { \mathrm{hom}}_{\lambda}(v , s^{-u}(v'))\ ] ] for all objects @xmath64 and @xmath85 of @xmath1263 and for every degree @xmath242 , with identities @xmath1274 for every object @xmath64 of @xmath1263 and with compositions @xmath1275 for all @xmath1276 and @xmath1277 . the _ standard @xmath2-suspension system on @xmath1278
_ is the @xmath2-suspension system @xmath1267 on @xmath1278 defined by @xmath1279 for every @xmath2-graded vector space @xmath1280 and every @xmath242 , where @xmath1281 is the functor defined by @xmath1282 .
we denote with @xmath1283 the @xmath2-graded extension of @xmath1278 along the standard @xmath2-suspension system @xmath1267 .
the results of the previous subsections can now be summarized as follows . for every object @xmath398 of @xmath17 and for every @xmath242
we have a functor @xmath1284 mapping each object @xmath564 of @xmath933 to @xmath1285 and each morphism @xmath820 $ ] of @xmath933 to @xmath1286 $ ] .
we have a natural isomorphism @xmath1287 associating with each object @xmath564 of @xmath933 the morphism @xmath1288 $ ] .
we also have for each @xmath232 natural isomorphisms @xmath1289 associating with each object @xmath564 of @xmath933 the morphism @xmath1290 $ ] .
analogously , for every object @xmath398 of @xmath17 and for every @xmath242 we have a functor @xmath1291 mapping each object @xmath1292 of @xmath1031 to @xmath1293 and each morphism @xmath853 $ ] of @xmath1031 to @xmath1294 $ ] .
we have a natural isomorphism @xmath1295 associating with each object @xmath1292 of @xmath1031 the morphism @xmath1296 $ ] .
we also have for each @xmath232 natural isomorphisms @xmath1289 associating with each object @xmath1292 of @xmath1031 the morphism @xmath1297 $ ] .
[ suspension_spheres ] for every object @xmath398 of @xmath17 the functors @xmath1269 together with the natural transformations @xmath1298 and @xmath1299 for @xmath232 form a @xmath2-suspension system on @xmath933 and on @xmath1031 .
we are now ready to improve the extended universal construction by defining a @xmath2-graded version of the extended quantization functors @xmath1138 and @xmath1262 .
the _ covariant universal @xmath2-graded algebroid _
@xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] associated with an object @xmath398 of @xmath17 is the @xmath2-graded extension of @xmath933 along the @xmath2-suspension system @xmath1267 given by proposition [ suspension_spheres ] .
the _ contravariant universal @xmath2-graded algebroid _
@xmath1271{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})$ ] associated with an object @xmath398 of @xmath17 is the @xmath2-graded extension of @xmath1031 along the @xmath2-suspension system @xmath1267 given by proposition [ suspension_spheres ] .
the _ covariant universal @xmath2-graded linear functor _
@xmath1300{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ) \rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')\ ] ] associated with a 1-morphism @xmath400 of @xmath17 is the @xmath2-graded linear functor mapping every object @xmath564 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] to the object @xmath565 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] and mapping every degree @xmath1135 morphism @xmath1301 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}^u({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}})$ ] to the degree @xmath1135 morphism @xmath1302 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}^u ( { \mathbf{\sigma}}\circ { \mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}\circ { \mathbf{\sigma}}''_{{\mathbf{\gamma}}}).\ ] ] the _ contravariant universal @xmath2-graded linear functor _
@xmath1303{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma } } ' ) \rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})\ ] ] associated with a 1-morphism @xmath400 of @xmath17 is the @xmath2-graded linear functor mapping every object @xmath1304 of @xmath1271{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}}')$ ] to the object @xmath1305 of @xmath1271{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})$ ] and mapping every degree @xmath1135 morphism @xmath1306 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}}')}^u({\mathbf{\sigma}}'_{{\mathbf{\gamma}}'},{\mathbf{\sigma}}'''_{{\mathbf{\gamma}}'})$ ] to the degree @xmath1135 morphism @xmath1307 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})}^u ( { \mathbf{\sigma}}'_{{\mathbf{\gamma } } ' } \circ { \mathbf{\sigma}},{\mathbf{\sigma}}'''_{{\mathbf{\gamma } } ' } \circ { \mathbf{\sigma}}).\ ] ] the _ covariant universal @xmath2-graded natural transformation _ @xmath1308{bbeta.pdf}}}_{{\mathcal c}}({\mathbf{m } } ) : { \mathbb f}_{{\mathcal c}}({\mathbf{\sigma } } ) \rightarrow { \mathbb f}_{{\mathcal c}}({\mathbf{\sigma}}')\ ] ] associated with a 2-morphism @xmath410 of @xmath17 is the @xmath2-graded natural transformation associating with every object @xmath564 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] the degree 0 morphism @xmath1309 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}^0({\mathbf{\sigma}}\circ { \mathbf{\sigma}}_{{\mathbf{\gamma } } } , { \mathbf{\sigma } } ' \circ { \mathbf{\sigma}}_{{\mathbf{\gamma}}}).\ ] ] the _ contravariant universal @xmath2-graded natural transformation _ @xmath1308{bbeta.pdf}}}'_{{\mathcal c}}({\mathbf{m } } ) : { \mathbb f}'_{{\mathcal c}}({\mathbf{\sigma } } ) \rightarrow { \mathbb f}'_{{\mathcal c}}({\mathbf{\sigma}}')\ ] ] associated with a 2-morphism @xmath410 of @xmath17 is the @xmath2-graded natural transformation associating with every object @xmath1292 of @xmath1271{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}}')$ ] the degree 0 morphism @xmath1310 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})}^0({\mathbf{\sigma}}'_{{\mathbf{\gamma } } ' } \circ { \mathbf{\sigma } } , { \mathbf{\sigma}}'_{{\mathbf{\gamma } } ' } \circ { \mathbf{\sigma}}').\ ] ] the _ covariant @xmath2-graded extended quantization functor _ @xmath1311 associated with @xmath19 is the functor mapping each object @xmath398 of @xmath17 to @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] , each 1-morphism @xmath537 of @xmath17 to @xmath1312 and each 2-morphism @xmath544 of @xmath17 to @xmath1313{bbeta.pdf}}}_{{\mathcal c}}({\mathbf{m}})$ ] .
the _ contravariant @xmath2-graded extended quantization functor _ @xmath1314 associated with @xmath19 is the functor mapping each object @xmath398 of @xmath17 to @xmath1271{$\mathbb v$}}}}'_{{\mathcal c}}({\mathbf{\gamma}})$ ] , each 1-morphism @xmath537 of @xmath17 to @xmath1315 and each 2-morphism @xmath544 of @xmath17 to @xmath1313{bbeta.pdf}}}'_{{\mathcal c}}({\mathbf{m}})$ ] .
[ graded_generators_upgraded ] let @xmath408 be an object of @xmath17 , let @xmath751 be a cobordism from @xmath140 to @xmath329 , let @xmath69 be a fundamental ribbon set for @xmath751 with respect to @xmath398 , let @xmath990 be a base set for @xmath751 and let @xmath987 be a lagrangian .
then the set @xmath1316 dominates @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] .
let @xmath992 and @xmath993{fourth_derivative.pdf}}}_{{\mathbf{\gamma } } } = ( \sigma^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{\vargamma},{{\mathrm{p}}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}},\vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } , { \mathcal l}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}})$ ] be objects of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] .
if @xmath750 is a connected cobordism with corners from @xmath752 to @xmath1317 and if @xmath843 is a connected cobordism with corners from @xmath751 to @xmath1318{fourth_derivative.pdf}}}_{\vargamma}$ ] then , thanks to the connection lemma [ graded_connection ] , every morphism inside @xmath1319{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}''_{{\mathbf{\gamma}}},{\mathbf{\sigma}}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{{\mathbf{\gamma } } } ) = { \mathrm{hom}}_{\lambda_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}''_{{\mathbf{\gamma}}},({\mathbf{s}}^2)^{-u } \otimes { \mathbf{\sigma}}^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{{\mathbf{\gamma}}})\ ] ] can be described by some linear combination of admissible @xmath4-skeins inside @xmath1320{fourth_derivative.pdf}}}_{\vargamma } : = { \mathrm{m}}_{\vargamma } \cup_{{\mathrm{s}}^2 \sqcup \sigma_{\vargamma } } \left ( ( { \mathrm{s}}^2 \times { \mathrm{i } } ) \sqcup { \mathrm{m}}''_{\vargamma } \right ) \cup_{{\mathrm{s}}^2 \sqcup { \mathrm{s}}^2 \sqcup \sigma^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}_{\vargamma } } \left ( { \mathrm{p}}^3 \sqcup ( \sigma^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{\vargamma } \times { \mathrm{i } } ) \right)\ ] ] relative to @xmath1321{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } , ( \vartheta_{{\mathrm{s}}^2})_{b_{s^2 } } \sqcup \vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } \right).\ ] ] but , just like in the proof of lemma [ graded_generators ] , and thanks to lemmas [ spheres_lemma ] and [ pants_lemma ] , every such admissible @xmath4-skein @xmath831 can be written , up to isotopy and skein equivalence , as a linear combination of admissible @xmath4-skeins inducing a decomposition of @xmath1322{fourth_derivative.pdf}}}_{\vargamma},{\mathrm{t}}^{\varphi},\omega,0 \right]$ ] into a linear combination of morphisms of the form @xmath1323 \otimes { \mathrm{id}}_{{\mathbf{\sigma}}^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf}}}_{{\mathbf{\gamma } } } } \right ) \ast \left ( \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{-v_i } } \right ] \otimes \left[{\mathrm{m}}''_{\vargamma},{\mathrm{t}}''^{\varphi''_i}_i,\omega''_i,0 \right ] \right ) \ast \left[{\mathrm{m}}_{\vargamma},{\mathrm{t}}^{\varphi_i}_i,\omega_i,0 \right]\ ] ] for some fundamental @xmath393-colorings @xmath1003 of @xmath977 and for some fundamental @xmath393-colorings @xmath1324 of @xmath1083 with @xmath841 , for some @xmath4-skein @xmath1004 inside @xmath750 relative to @xmath1325 and for some @xmath4-skein @xmath1006 inside @xmath843 realtive to @xmath1326{fourth_derivative.pdf}}}}^{v^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } , ( \vartheta_{{\mathrm{s}}^2})_{b_{s^2 } } \sqcup \vartheta^{{\includegraphics[height=\heightof{\scriptsize $ \prime$}]{fourth_derivative.pdf}}}_{b^{{\includegraphics[height=\heightof{\tiny $ \prime$}]{fourth_derivative.pdf } } } } \right)\ ] ] for every @xmath841 .
this section contains the first part of the proof of our main result , which is the monoidality of the functor @xmath21 .
[ notation_constant_functor ] if @xmath64 is an object of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] then we will denote with @xmath1327{$\mathbb v$}}}}$ ] the unique @xmath2-graded linear functor mapping the unique object of the unit @xmath2-graded algebroid @xmath22 to @xmath64 .
[ monoidality_unit ] the @xmath2-graded linear functor @xmath1328{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] is a @xmath2-morita equivalence .
@xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] is dominated by the set @xmath1329 , as follows immediately from proposition [ graded_generators_upgraded ] and from theorem [ graded_cauchy_completion_equivalence ] .
we will constantly confuse @xmath1330 with @xmath499 and @xmath1331 with @xmath398 for all objects @xmath398 and @xmath499 of @xmath17 by a very slight abuse of notation .
this means that the braiding 1-morphisms @xmath1332 and @xmath1333 of @xmath17 given by definition [ braiding_1-morphism ] will be confused with @xmath1334 and with @xmath535 respectively .
then for all 1-morphisms @xmath1335 of @xmath17 we will use the notation @xmath1336 and @xmath1337 for the braiding 2-morphisms @xmath1338 and @xmath1339 and we will consider them as a 2-morphisms from @xmath539 to @xmath1340 and from @xmath1341 to @xmath1342 respectively .
we introduce a short notation which will be useful in the following : consider objects @xmath1343 , @xmath1344 , 1-morphisms @xmath1345 , @xmath1346 and 2-morphisms @xmath1347 of @xmath17 .
then we denote with @xmath1348 the 2-morphism of @xmath17 given by @xmath1349 we denote with @xmath1350 the 2-morphism of @xmath17 given by @xmath1351 let us consider objects @xmath398 and @xmath499 of @xmath17 and let us define the transformation @xmath1352 as follows : with every object @xmath1353 of @xmath1354 the transformation @xmath1355 associates the @xmath2-graded linear functor @xmath1356{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ) { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ' ) \rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')\ ] ] mapping every object @xmath1357 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ) { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] to the object @xmath1358 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')$ ] and mapping every degree @xmath1359 morphism @xmath1360 \otimes [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \ ] ] of @xmath1361{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma } } } ) \otimes { \mathrm{hom}}^{u'}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}({\mathbf{\sigma}}_{{\mathbf{\gamma}}'},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}'})$ ] to the degree @xmath1359 morphism @xmath1362\ ] ] of @xmath1363{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')}({\mathbf{\sigma}}_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}_{{\mathbf{\gamma } } ' } , { \mathbf{\sigma}}''_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}''_{{\mathbf{\gamma}}'})$ ] . with every 1-morphism @xmath1364 of @xmath1354 the transformation @xmath1355 associates the @xmath2-graded natural transformation @xmath1365 associating with every object @xmath1357 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ) { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] the degree 0 morphism with @xmath1366 by a very slight abuse of notation . ]
@xmath1367 of @xmath1368{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } '' \otimes { \mathbf{\gamma}}''')}\left ( ( { \mathbf{\sigma}}\circ { \mathbf{\sigma}}_{{\mathbf{\gamma } } } ) \otimes ( { \mathbf{\sigma } } ' \circ { \mathbf{\sigma}}_{{\mathbf{\gamma}}'}),({\mathbf{\sigma}}\otimes { \mathbf{\sigma } } ' ) \circ ( { \mathbf{\sigma}}_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}_{{\mathbf{\gamma } } ' } ) \right)$ ] . to actually check that @xmath1369 is a transformation we need to go through some very lengthy but also very standard applications of structural properties of the symmetric monoidal 2-category @xmath17 . we will not report them here as they are not very enlightening .
we just stress one fact : the proof that @xmath1370 is a functor uses the naturality of the braiding of @xmath17 together with the identity @xmath1371 \\ & = \nu(v',v '' ) \cdot \left [ ( { \mathbf{p}}^3)^{v + v'',v ' + v ' '' } \ast \left ( ( { \mathbf{p}}^3)^{v , v '' } \otimes ( { \mathbf{p}}^3)^{v',v'''}\right ) \right ] \end{aligned}\ ] ] which holds for all @xmath1372 and which itself follows from lemmas [ associativity_of_pants ] and [ twisted_pants_lemma ] .
this is used to prove the equality . ]
\otimes [ { \mathbf{m}}''^{u'''}_{{\mathbf{\gamma } } ' } ] \right ) \circ { \boldsymbol{\eta}}^{u + u'}_{{\mathbf{\gamma}},{\mathbf{\gamma } } ' } \left ( [ { \mathbf{m}}^{u}_{{\mathbf{\gamma } } } ] \otimes [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \right ) \\ & = \nu(u',u '' ) \cdot { \boldsymbol{\eta}}^{u + u ' + u '' + u'''}_{{\mathbf{\gamma}},{\mathbf{\gamma } } ' } \left ( \left ( [ { \mathbf{m}}''^{u''}_{{\mathbf{\gamma } } } ] \star [ { \mathbf{m}}^{u}_{{\mathbf{\gamma } } } ] \right ) \otimes \left ( [ { \mathbf{m}}''^{u'''}_{{\mathbf{\gamma } } ' } ] \star [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \right ) \right ) . \end{aligned}\ ] ] [ monoidality_tensor_product ] for all objects @xmath398 and @xmath499 of @xmath17 the @xmath2-graded linear functor @xmath1356{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ) { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ' ) \rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')\ ] ] is a @xmath2-morita equivalence .
if @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] is dominated by @xmath1374 and if @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] is dominated by @xmath1375 then @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')$ ] is dominated by @xmath1376 therefore @xmath1370 clearly defines a bijection between generators . to see that it is faithful let us consider morphisms @xmath1360 \in { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}})\ ] ] and @xmath1377 \in { \mathrm{hom}}^{u'}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}({\mathbf{\sigma}}_{{\mathbf{\gamma}}'},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}'})\ ] ] such that @xmath1378 \otimes [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \right ) = 0 $ ] .
if we suppose that @xmath1301 \neq 0 $ ] then there have to exist some objects @xmath1292 , @xmath1379 of @xmath1031 and some 2-morphisms @xmath1380 of @xmath17 such that @xmath1381 now for all objects @xmath1304 , @xmath1382 of @xmath1031 and for all 2-morphisms @xmath1383 of @xmath17 we have @xmath1384 thanks to lemmas [ pants_composition_1 ] and [ pants_composition_2 ] .
but the first member vanishes thanks to the assumption @xmath1378 \otimes [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \right ) = 0 $ ] , so that @xmath1385 = 0 $ ] . to see that it is also full we need to show that every morphism in @xmath1273{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')}^u \left ( { \mathbf{\sigma}}_{{\mathbf{\gamma}}}^{(v,\vartheta ) } \otimes { \mathbf{\sigma}}^{(v',\vartheta')}_{{\mathbf{\gamma } } ' } , { \mathbf{\sigma}}''^{(v'',\vartheta'')}_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}''^{(v''',\vartheta''')}_{{\mathbf{\gamma } } ' } \right)\ ] ] can be written as @xmath1386 \otimes [ { \mathbf{m}}^{u_i}_{i,{\mathbf{\gamma } } ' } ] \right)\ ] ] for some @xmath1387 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}^{u - u_i } \left ( { \mathbf{\sigma}}_{{\mathbf{\gamma}}}^{(v,\vartheta)},{\mathbf{\sigma}}''^{(v'',\vartheta'')}_{{\mathbf{\gamma } } } \right)\ ] ] and for some @xmath1388 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}^{u_i } \left ( { \mathbf{\sigma}}^{(v',\vartheta')}_{{\mathbf{\gamma}}'},{\mathbf{\sigma}}''^{(v''',\vartheta''')}_{{\mathbf{\gamma } } ' } \right).\ ] ] to do so let us consider connected cobordisms with corners @xmath750 from @xmath751 to @xmath1389 and @xmath1390 from @xmath1391 to @xmath1392 .
let @xmath1393 denote the cobordism with corners obtained by gluing @xmath1394 to @xmath1395 along @xmath1396 then every morphism in @xmath1273{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}\otimes { \mathbf{\gamma}}')}^u \left ( { \mathbf{\sigma}}_{{\mathbf{\gamma}}}^{(v,\vartheta ) } \otimes { \mathbf{\sigma}}^{(v',\vartheta')}_{{\mathbf{\gamma } } ' } , { \mathbf{\sigma}}''^{(v'',\vartheta'')}_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}''^{(v''',\vartheta''')}_{{\mathbf{\gamma } } ' } \right)\ ] ] can be presented by some admissible @xmath4-skein inside @xmath1393 .
then up to isotopy and skein equivalence lemma [ pants_lemma ] allows us to conclude .
in this section we complete the proof of theorem [ main ] by establishing the symmetry of the monoidal functor @xmath21 .
the _ flip morphism @xmath1397 \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ' \otimes { \mathbf{\gamma}})}^0({\mathbf{\sigma}}_{{\mathbf{\gamma } } ' } \otimes { \mathbf{\sigma}}_{{\mathbf{\gamma } } } , \beta_{{\mathbf{\gamma}},{\mathbf{\gamma } } ' } \circ ( { \mathbf{\sigma}}_{{\mathbf{\gamma } } } \otimes { \mathbf{\sigma}}_{{\mathbf{\gamma}}'}))\ ] ] associated with objects @xmath564 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] and @xmath1398 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] _ is the degree 0 morphism of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ' \otimes { \mathbf{\gamma}})$ ] given by @xmath1399.\ ] ] for all objects @xmath398 , @xmath499 of @xmath17 , @xmath564 , @xmath788 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})$ ] and @xmath1398 , @xmath1400 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')$ ] and for every morphism @xmath1401 \otimes [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } ] \in { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}})}^u({\mathbf{\sigma}}_{{\mathbf{\gamma}}},{\mathbf{\sigma}}''_{{\mathbf{\gamma } } } ) \otimes { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma}}')}^{u'}({\mathbf{\sigma}}_{{\mathbf{\gamma}}'},{\mathbf{\sigma}}''_{{\mathbf{\gamma}}'})\ ] ] the diagram ( p0 ) at ( 0:3.7 ) @xmath1402 ; ( p1 ) at ( 90:3.7 ) @xmath1403 ; ( p2 ) at ( 2 * 90:3.7 ) @xmath1404 ; ( p3 ) at ( 3 * 90:3.7 ) @xmath1405 ; ( p1 ) edge[- > ] node[descr ] @xmath1406 \circ { \boldsymbol{\eta}}^{u+u'}_{{\mathbf{\gamma}},{\mathbf{\gamma } } ' } \left ( \left [ { \mathbf{m}}^u_{{\mathbf{\gamma } } } \right ] \otimes \left [ { \mathbf{m}}^{u'}_{{\mathbf{\gamma } } ' } \right ] \right)$ ] ( p0 ) ( p2 ) edge[- > ] node[descr ] @xmath1407 \quad$ ] ( p1 ) ( p2 ) edge[- > ] node[descr ] @xmath1408 \otimes \left [ { \mathbf{m}}^u_{{\mathbf{\gamma } } } \right ] \right ) \quad$ ] ( p3 ) ( p3 ) edge[- > ] node[descr ] @xmath1409 $ ] ( p0 ) ; of morphisms of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{\gamma } } ' \otimes { \mathbf{\gamma}})$ ] is commutative . the proof of this theorem is a rather lengthy application of the naturality property of the braiding @xmath1410 of @xmath17 , so we will not write it in detail .
the only additional ingredient which is needed is the following equality @xmath1411 \\ & = \nu(u , u ' ) \cdot \left [ ( { \mathbf{p}}^3)^{0,-u'-u } \ast \left ( ( { \mathbf{d}}^3)^0 \otimes ( { \mathbf{p}}^3)^{-u',-u } \right ) \right ] \\ & = \left [ ( { \mathbf{p}}^3)^{0,-u'-u } \ast \left ( ( { \mathbf{d}}^3)^0 \otimes ( \tilde{{\mathbf{p}}}^3)^{-u',-u } \right ) \right ] \end{aligned}\ ] ] between morphisms of @xmath1412 , which is a consequence of lemmas [ unitarity_of_discs ] , [ associativity_of_pants ] and [ twisted_pants_lemma ] and of the equality @xmath1413 which obviously holds for all @xmath232 .
in this section we describe the @xmath2-graded tqft produced by @xmath21 . indeed , as it was explained in the introduction , every @xmath2-graded etqft on @xmath17 yields a @xmath2-graded tqft on @xmath548 , and @xmath2-graded linear functors associated with closed 1-morphisms of @xmath17 determine @xmath2-graded vector spaces which we explicitly characterize . when the relative modular category @xmath4 is taken to be the category @xmath16 of finite - dimensional complex - weight representations of the unrolled version of quantum @xmath0 at @xmath1414 then our description recovers the @xmath10-graded tqft defined in @xcite
the _ covariant universal @xmath2-graded vector space _ @xmath1415 associated with a 1-morphism @xmath1416 of @xmath17 is the @xmath2-graded vector space whose space of degree @xmath1135 vectors @xmath1417 is given by . ]
@xmath1418 for every @xmath242 .
the _ contravariant universal @xmath2-graded vector space _ @xmath1419 associated with a 1-morphism @xmath1416 of @xmath17 is the @xmath2-graded vector space whose space of degree @xmath1135 vectors @xmath1420 is given by @xmath1421 for every @xmath242 .
the _ covariant universal @xmath2-graded linear map _
@xmath1422{bbf.pdf}}_{{\mathcal c}}({\mathbf{m } } ) : { \mathbb v}_{{\mathcal c}}({\mathbf{\sigma } } ) \rightarrow { \mathbb v}_{{\mathcal c}}({\mathbf{\sigma}}')\ ] ] associated with a 2-morphism @xmath476 between 1-morphisms @xmath1423 of @xmath17 is the @xmath2-graded linear map whose degree @xmath1135 component @xmath1424{bbf.pdf}}^u_{{\mathcal c}}({\mathbf{m}})$ ] is given by . ]
@xmath1425 for every @xmath242 .
the _ contravariant universal @xmath2-graded linear map _ @xmath1422{bbf.pdf}}'_{{\mathcal c}}({\mathbf{m } } ) : { \mathbb v}'_{{\mathcal c}}({\mathbf{\sigma } } ' ) \rightarrow { \mathbb v}'_{{\mathcal c}}({\mathbf{\sigma}})\ ] ] associated with a 2-morphism @xmath476 between 1-morphisms @xmath1423 of @xmath17 is the @xmath2-graded linear map whose degree @xmath1135 component @xmath1424{bbf.pdf}}'^u_{{\mathcal c}}({\mathbf{m}})$ ] is given by @xmath1426 for every @xmath242 .
the _ covariant @xmath2-graded quantization functor _ @xmath1427 associated with @xmath19 is the functor mapping each object @xmath537 of @xmath17 to @xmath1415 and each morphism @xmath544 of @xmath17 to @xmath1424{bbf.pdf}}_{{\mathcal c}}({\mathbf{m}})$ ] . the _ contravariant @xmath2-graded quantization functor _ @xmath1428 associated with @xmath19 is the functor mapping each object @xmath537 of @xmath17 to @xmath1419 and each morphism @xmath544 of @xmath17 to @xmath1424{bbf.pdf}}'_{{\mathcal c}}({\mathbf{m}})$ ] .
[ non - degenerate_pairing_on_graded_tqft_vector_spaces ] for every object @xmath537 of @xmath548 the @xmath2-graded pairing @xmath1429 whose degree 0 component @xmath1430 maps every vector @xmath1431 \otimes [ { \mathbf{m}}^{u}_{{\mathbf{\sigma } } } ] \in { \mathbb v}'^{-u}_{{\mathcal c}}({\mathbf{\sigma } } ) \otimes { \mathbb v}^u_{{\mathcal c}}({\mathbf{\sigma}})\ ] ] to @xmath1432 is non - degenerate thanks to remark [ non - degenerate_pairing_on_tqft_vector_spaces ] .
let us consider the functor @xmath1273{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , \cdot ) : { \check{\mathrm{c}}\mathrm{ob}}_3^{{\mathcal c } } \rightarrow { \mathrm{vect}}_{{\mathbb c}}^{\varpi}\ ] ] mapping every object @xmath537 of @xmath548 to the @xmath2-graded vector space @xmath1433{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}})$ ] and mapping every morphism @xmath476 of @xmath548 to the @xmath2-graded linear map @xmath1433{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing},{\mathbf{m}})$ ] whose degree @xmath1135 component is given by @xmath1434{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing},{\mathbf{m } } ) : & { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma } } ) & \rightarrow & { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma } } ' ) \\ & \left [ { \mathbf{m}}^u_{\varnothing } \right ] & \mapsto & \left [ \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{-u } } \otimes { \mathbf{m}}\right ) \ast { \mathbf{m}}^u_{\varnothing } \right ] , \end{array}\ ] ] and let us consider the natural transformation @xmath1435{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , \cdot ) \rightarrow { \mathbb{u}}_{{\mathcal c}}\ ] ] associating with every object @xmath537 of @xmath548 the @xmath2-graded linear map @xmath1436 whose degree @xmath1135 component is given by @xmath1437{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma } } ) & \rightarrow & { \mathbb v}^u_{{\mathcal c}}({\mathbf{\sigma } } ) \\ & \left [ { \mathbf{m}}^u_{\varnothing } \right ] & \mapsto & \left [ { \mathbf{m}}^u_{\varnothing } \right ] . \end{array}\ ] ] [ tqft_isomorphism ] the natural transformation @xmath1438 is an isomorphism . since @xmath1439 is clearly surjective for every degree @xmath1135 in @xmath2 and for every object @xmath537 of @xmath548 , we just need to show that it is also injective .
let us consider a vector @xmath1440 $ ] of @xmath1361{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}})$ ] such that @xmath1440 = 0 $ ] in @xmath1417 .
this means @xmath1441 for every 2-morphism @xmath1442 of @xmath17 .
now @xmath1440 $ ] is trivial in @xmath1361{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}})$ ] too if furthermore @xmath1443 for all 1-morphisms @xmath1444 of @xmath17 and for all 2-morphisms @xmath1445 of @xmath17 .
but since @xmath1446 we can conclude .
the covariant @xmath2-graded quantization functor @xmath1447 is a @xmath2-graded tqft on @xmath548 .
in order to establish the relationship between @xmath1447 and @xmath1063 we need to find a @xmath2-morita equivalence between @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] and @xmath1283 .
it turns out we have a very natural one which can be defined in terms of covariant universal @xmath2-graded vector spaces .
let us begin by laying the ground with some preliminary remarks .
the proofs of lemmas [ graded_connection ] , [ graded_connection_prime ] and [ graded_triviality ] can be directily adapted to the simpler context of the universal construction in order to obtain the following analogue results , which we state for the convenience of the reader .
[ graded_connection_tqft ] let @xmath1416 be a 1-morphism of @xmath17 of the form @xmath1448 and let @xmath1449 be a non - empty connected 3-dimensional cobordism from @xmath140 to @xmath67 . if @xmath556 is a quantum invariant on @xmath550 wich satisfies the surgery axioms and the skein axiom then the natural linear maps @xmath1450{cals.pdf}}({\mathrm{m}}_{\sigma};(\varnothing^{\varnothing},\varnothing_{\varnothing}),({\mathrm{p}}^{v},\vartheta_{b } ) ) & \rightarrow & { \mathrm{v}}_{z}({\mathbf{\sigma } } ) \\ & ( { \mathrm{t}}^{\varphi},\omega ) & \mapsto & [ { \mathrm{m}}_{\sigma},{\mathrm{t}}^{\varphi},\omega,0 ] \end{array}\ ] ] and @xmath1451{cals.pdf}}({\mathrm{m}}'_{\sigma};({\mathrm{p}}'^{v'},\vartheta'_{b'}),(\varnothing^{\varnothing},\varnothing_{\varnothing } ) ) & \rightarrow & { \mathrm{v}}'_{z}({\mathbf{\sigma } } ) \\ & ( { \mathrm{t}}'^{\varphi'},\omega ' ) & \mapsto & [ { \mathrm{m}}'_{\sigma},{\mathrm{t}}'^{\varphi'},\omega',0 ] \end{array}\ ] ] are surjective . [ graded_triviality_tqft ] let @xmath1416 be a 1-morphism of @xmath17 of the form @xmath1452 let @xmath1453 be 2-morphisms of @xmath17 for @xmath134 and let @xmath1454 be a connected cobordism with corners from @xmath67 to @xmath140 . if @xmath556 is a quantum invariant on @xmath550 wich satisfies the surgery axioms and the skein axiom then a linear combination @xmath1455 $ ] determines a trivial vector of @xmath1456 if and only if the expression @xmath1457 vanishes for all @xmath4-skeins @xmath832 inside @xmath1454 relative to @xmath1458 and @xmath1459 .
let us consider for every object @xmath1460 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] the @xmath2-graded pairing @xmath1461{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing } ) \otimes { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}}_{\varnothing } ) \rightarrow { \mathbb c}\ ] ] whose degree 0 component @xmath1462{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing } ) \otimes { \mathrm{hom}}^{u}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}}_{\varnothing } ) \rightarrow { \mathbb c}\ ] ] maps every vector @xmath1463 \otimes \left [ { \mathbf{m}}^{u}_{\varnothing } \right ] \in { \mathrm{hom}}^{-u}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing } ) \otimes { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}}_{\varnothing } ) \ ] ] to @xmath1464 [ non - degenerate_pairing_on_graded_etqft_hom_spaces ] for every object @xmath1465 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] the @xmath2-graded pairing @xmath1466 is non - degenerate .
let us start by supposing that @xmath1467 \in { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing } , { \mathbf{\sigma}}_{\varnothing } ) $ ] is non - zero .
let us fix a connected cobordism @xmath1468 from @xmath1469 to @xmath1068 and let @xmath1470 denote the connected cobordism @xmath1471 from @xmath1472 to @xmath140 .
then thanks to proposition [ tqft_isomorphism ] and to lemma [ graded_triviality_tqft ] there has to exist some @xmath4-skein @xmath1473 inside @xmath1470 relative to @xmath1474 satisfying @xmath1475 but @xmath1476 = \left [ ( \overline{{\mathbf{d}}^3})^0 \ast ( { \mathbf{p}}^3)^{-u , u } \ast \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{-u } } \otimes { \mathbf{m}}'^{-u}_{\varnothing } \right ) \right]\ ] ] for some @xmath1477{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing})$ ] thanks to lemmas [ spheres_lemma ] , [ index_1_consequence ] and [ pants_lemma ] .
let us suppose now that @xmath1478 \in { \mathrm{hom}}^{-u}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing})$ ] is non - zero .
let us fix a connected cobordism @xmath1479 from @xmath140 to @xmath1472 and let @xmath1480 denote the connected cobordism @xmath1481 from @xmath140 to @xmath1482 . then
thanks to lemma [ graded_triviality ] there have to exist @xmath4-skeins @xmath1483 inside @xmath1480 relative to @xmath1484 and @xmath1485 inside @xmath1486 relative to @xmath1487 satisfying @xmath1488 but then @xmath1489 thanks to lemma [ pants_lemma ] , there exists some non - zero @xmath1235 satisfying @xmath1490 = \lambda \cdot \left [ ( \overline{{\mathbf{d}}^3})^0 \ast ( { \mathbf{p}}^3)^{-u , u } \ast \beta_{({\mathbf{s}}^2)^{u},({\mathbf{s}}^2)^{-u } } \right]\ ] ] thanks to lemmas [ spheres_lemma ] , [ index_1_consequence ] and [ pants_lemma ] and we must have @xmath1491 = \left [ \beta_{({\mathbf{s}}^2)^{-u},{\mathbf{\sigma}}_{\varnothing } } \ast { \mathbf{m}}^{u}_{{\mathbf{\sigma}}_{\varnothing } } \right]\ ] ] for some @xmath1492{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing})$ ] thanks to lemma [ spheres_lemma ] . let us consider the functor @xmath1493{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing ) \rightarrow { \mathbb{v}\mathrm{ect}}_{{\mathbb c}}^{\varpi,{\mathrm{fg}}}\ ] ] mapping every object @xmath1460 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] to the @xmath2-graded vector space @xmath1494 and mapping every degree @xmath1135 morphism @xmath1440 $ ] of @xmath1361{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathbf{\sigma}}''_{\varnothing})$ ] to the @xmath2-graded linear map @xmath1495 ) : { \mathbb v}_{{\mathcal c}}({\mathbf{\sigma}}_{\varnothing } ) \rightarrow s^{-u}({\mathbb v}_{{\mathcal c}}({\mathbf{\sigma}}''_{\varnothing}))\ ] ] whose degree @xmath1496 component is given by @xmath1497 ) ^v : & { \mathbb v}^v_{{\mathcal c}}({\mathbf{\sigma}}_{\varnothing } ) & \rightarrow & { \mathbb v}^{u + v}_{{\mathcal c}}({\mathbf{\sigma}}''_{\varnothing } ) \\ & \left [ { \mathbf{m}}^v_{{\mathbf{\sigma}}_{\varnothing } } \right ] & \mapsto & \left [ \left ( ( { \mathbf{p}}^3)^{-v ,- u } \otimes { \mathrm{id}}_{{\mathbf{\sigma}}''_{\varnothing } }
\right ) \ast \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{-v } } \otimes { \mathbf{m}}^u_{\varnothing }
\right ) \ast { \mathbf{m}}^v_{{\mathbf{\sigma}}_{\varnothing } } \right ] \end{array}\ ] ] [ graded_universal_algebroid_empty_equivalence ] the @xmath2-graded linear functor @xmath1493{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing ) \rightarrow { \mathbb{v}\mathrm{ect}}_{{\mathbb c}}^{\varpi,{\mathrm{fg}}}\ ] ] is a @xmath2-morita equivalence .
since @xmath1283 is dominated by the unit @xmath2-graded algebroid @xmath22 and since @xmath1498 is ismorphic to it as an object of @xmath1283 then , thanks to theorem [ graded_cauchy_completion_equivalence ] , we just need to show that @xmath1499 is faithful and full . to see this
let us fix objects @xmath1465 and @xmath1500 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] .
let @xmath1501 \in { \mathrm{hom}}^u_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathbf{\sigma}}''_{\varnothing})\ ] ] be a morphism satisfying @xmath1502 \right ) = 0.\ ] ] this means that @xmath1503 for all 2-morphisms @xmath1504 and @xmath1505 of @xmath17 .
then we can first choose connected cobordisms @xmath1479 from @xmath140 to @xmath1472 and @xmath1506 from @xmath1507 to @xmath140 , and then specify @xmath1508 as a connected cobordism from @xmath140 to @xmath1482 and @xmath1509 as a connected cobordism from @xmath1510 to @xmath140 .
this means that , thanks to lemma [ graded_triviality_tqft ] , we have @xmath1511 = 0\ ] ] if and only if @xmath1512 for every @xmath1513 , for all admissible @xmath4-skeins @xmath1483 inside @xmath1480 relative to @xmath1484 and for all @xmath4-skeins @xmath1485 inside @xmath1514 relative to @xmath1515 therefore faithfulness follows from lemmas [ spheres_lemma ] and [ pants_lemma ] . finally in order to show fullness let us fix some @xmath1516 and let us choose a basis @xmath1517,\ldots , [ { \mathbf{m}}^v_{k,{\mathbf{\sigma}}_{\varnothing } } ] \}\ ] ] of @xmath1518 , its dual basis @xmath1519,\ldots , [ { \mathbf{m}}'^{-v}_{k,\varnothing } ] \}\ ] ] of @xmath1520{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathbf{\sigma}}_{\varnothing},{\mathrm{id}}_{\varnothing})$ ] with respect to the non - degenerate pairing induced by propositions [ tqft_isomorphism ] and [ non - degenerate_pairing_on_graded_etqft_hom_spaces ] and a basis @xmath1521,\ldots , [ { \mathbf{m}}''^{u+v}_{k'',\varnothing } ] \}\ ] ] of @xmath1522{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}(\varnothing)}({\mathrm{id}}_{\varnothing},{\mathbf{\sigma}}''_{\varnothing})$ ] .
then a basis for @xmath1523 is given by the linear maps @xmath1524
\star \left [ { \mathbf{m}}'^{-v}_{i,\varnothing } \right ] \right ) \right)^v\ ] ] for all @xmath841 and @xmath1525 .
we complete the picture with a standard linear algebra result which we state without proof .
if @xmath1526 is an object of @xmath1283 let @xmath1527 denote the @xmath2-graded linear functor mapping every object @xmath1280 of @xmath1283 to @xmath1528 and every morphism @xmath1424{bbf.pdf}}$ ] of @xmath1283 to @xmath1424{bbf.pdf}}\otimes { \mathrm{id}}_{{\mathbb w}}$ ]
. then if @xmath1529 is a @xmath2-graded linear functor let @xmath1313{bbeta.pdf}}}_{{\mathbb f } } : { \mathbb f}\rightarrow ( \cdot \otimes { \mathbb f}({\mathbb c}))$ ] denote the @xmath2-graded natural transformation associating with every @xmath2-graded vector space @xmath1280 the @xmath2-graded linear map whose degree @xmath1135 component @xmath1530{bbeta.pdf}}}_{{\mathbb f } } \right)^u_{{\mathbb v } } : \left ( { \mathbb f}({\mathbb v } ) \right)^u \rightarrow \bigoplus_{u ' \in \varpi } { \mathbb v}^{u - u ' } \otimes \left ( { \mathbb f}({\mathbb{c } } ) \right)^{u'}\ ] ] maps every degree @xmath1135 vector @xmath1531 of @xmath1532 to the degree @xmath1135 vector @xmath1533 of @xmath1534 , where @xmath1535 is a basis of @xmath1536 and where @xmath1537 is its dual basis of @xmath1538 for every @xmath1513 . [ graded_linear_functor ] the @xmath2-graded natural transformation @xmath1313{bbeta.pdf}}}_{{\mathbb f } } : { \mathbb f}\rightarrow ( \cdot \otimes { \mathbb f}({\mathbb c}))$ ] is a well - defined isomorphism . [ computation_goal ] our goal will be to produce computations of universal @xmath2-graded vector spaces associated with certain closed 1-morphisms of @xmath17 in terms of the relative modular category @xmath4 .
we will think about this question as follows : if @xmath1416 is a 1-morphism of @xmath17 then we look for a @xmath2-graded linear functor @xmath1539 together with a @xmath2-graded natural isomorphism - graded linear category @xmath22 features a single object , every such @xmath2-graded linear functor and every such @xmath2-graded natural isomorphism are constant , so we are actually looking for a single finitely generated @xmath2-graded vector space together with a @xmath2-graded linear isomorphism from @xmath1415 to it . ]
( p0 ) at ( 45:2 cm ) @xmath1283 ; ( p1 ) at ( 45 + 90:2 cm ) @xmath22 ; ( p2 ) at ( 45 + 2 * 90:2 cm ) @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] ; ( p3 ) at ( 45 + 3 * 90:2 cm ) @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}(\varnothing)$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1313{bbeta.pdf}}}_{{\mathbf{\sigma}}}$ ] ; ( p1 ) edge[- > ] node[above ] @xmath1541 ( p0 ) ( p1 ) edge[- > ] node[left ] @xmath547 ( p2 ) ( p2 ) edge[- > ] node[below ] @xmath1312 ( p3 ) ( p3 ) edge[- > ] node[right ] @xmath1499 ( p0 ) ; the idea is then to try and decompose the 1-morphism @xmath1416 as a compostion of tensor products of elementary 1-morphisms which are easier to describe and compute .
we describe here the covariant universal @xmath2-graded algebroids in terms of @xmath2-graded extensions of ideals of projective homogeneous objects of @xmath4 . for every index @xmath63 in @xmath1
there exists a standard @xmath2-suspension system on @xmath1542 given by the linear functors @xmath1543 mapping every object @xmath64 to @xmath243 and every morphism @xmath205 to @xmath1544 .
we denote with @xmath1545 the @xmath2-graded extension of @xmath1542 along this standard @xmath2-suspension system .
we begin with some preliminary results .
the following is a key technical result is needed in order to relate the @xmath2-suspension systems we are considering on universal @xmath2-graded algebroids to those we just introduced on ideals of projective homogeneous objects of @xmath4 .
[ fundamental_equivalence_lemma ] for all @xmath232 we have the equality @xmath1546 where @xmath1547 and @xmath1548 are the images under the reshetikhin - turaev functor @xmath187 of the @xmath4-colored ribbon graphs represented in figure [ fundamental_equivalence_figure ] .
[ fundamental_equivalence_figure ] -colored ribbon graphs representing @xmath1549 and @xmath1548.,title="fig : " ] the equality follows immediately from the properties of the @xmath10-bilinear homomorphism @xmath262 of definition [ relative_pre - modular_graded_categories_def ] and from the relative modularity condition of definition [ relative_modular_graded_categories_def ] .
[ shift_isomorphism_remark ] let us consider for all indices @xmath1550 and for every object @xmath64 of @xmath1551 the object of @xmath129 and for every morphism @xmath8 of @xmath129 we introduced the @xmath63-colored 1-sphere @xmath759 of definition [ g - circle ] and the @xmath130-colored 2-disc @xmath1552 of definition [ ( epsilon , v)-disc ] . ] @xmath1553 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] determined by the unique cohomology class @xmath1554 which is compatible with the @xmath4-coloring of @xmath1555 and which evaluates to @xmath1556 every relative homology class that can be represented by some oriented arc from @xmath768 to @xmath1557 contained in @xmath772 .
let @xmath1558 denote the 2-morphism @xmath1559 of @xmath17 determined by the unique @xmath1-coloring @xmath1560 extending @xmath1561 and @xmath1562 .
then @xmath1563 $ ] is a degree 0 isomorphism from @xmath1564 to @xmath1565 .
see figure [ shift_isomorphism_figure ] for a graphical representation of @xmath1566 .
[ shift_isomorphism_figure ] of @xmath17.,title="fig : " ] we denote with @xmath1567 the 3-dimensional cobordism with corners from @xmath139 to @xmath1568 whose support @xmath427 is given by is the open ball of center @xmath1169 and radius @xmath1170 in @xmath1171 . ]
@xmath1569 whose incoming horizontal boundary @xmath1570 is given by @xmath1571 with identification given by @xmath1572 , whose outgoing horizontal boundary @xmath1573 is given by @xmath1574 with identification induced by the diffeomorphisms @xmath1575 and @xmath1576 and whose outgoing vertical boundary identification is given by @xmath1577 .
[ notation_for_composition_with_section ] from now on we adopt the following notation : for every generic index @xmath63 in @xmath252 and for every projective homogeneous object @xmath64 of @xmath4 we choose a fixed section @xmath1578 of the epimorphism @xmath1579 .
then for every degree @xmath1135 in @xmath2 , for every index @xmath63 in @xmath1 and for every object @xmath64 of @xmath1542 we denote with @xmath1580 the retraction @xmath1581 of the map @xmath1582 and we denote with denote with @xmath1583 the idempotent @xmath1584 finally , for every degree @xmath1135 morphism @xmath1585 of @xmath1586 we denote with @xmath1587 the morphism @xmath1588 so that we have the equality @xmath1589 .
let @xmath1590{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)\ ] ] be the @xmath2-graded linear functor mapping every object @xmath64 of @xmath1545 to the object @xmath1591 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] and mapping every degree @xmath1135 morphism @xmath1585 of @xmath1592 to the degree @xmath1135 morphism @xmath1593 : = \left [ { \mathrm{y}},{\mathrm{t}}_{{\mathrm{y}}}^{f^u_{g_0}},\omega_{f^u_{g_0}},0 \right]\ ] ] of @xmath1594{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}({\mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v'')})$ ] , where @xmath1595 is the @xmath4-colored ribbon graph contained in @xmath1596 represented in figure [ punctured_cylinder ] and where @xmath1597 is the unique cohomology class which is compatible with the @xmath4-coloring of @xmath1598 and which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1599 .
[ punctured_cylinder ] morphism @xmath1600 $ ] .
the ribbon graph @xmath1598 is represented in blue with blackboard framing .
the cohomology class @xmath1597 is completely determined by the @xmath4-coloring on @xmath1598 and by the specified evaluations against the relative homology classes represented by the oriented red curves.,title="fig : " ] [ description_universal_algebroids ] the @xmath2-graded linear functor @xmath1601{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] is a well - defined @xmath2-morita equivalence .
the first property we have to establish is the functoriality of @xmath1602 .
the identity @xmath1603 = \left [ { \mathbf{y}}^0_{({\mathrm{id}}_{v})_{g_0 } } \right]\ ] ] follows from lemma [ index_1_consequence ] .
furthermore , the identity @xmath1604 = \left [ { \mathbf{y}}^{u'}_{f'^{u'}_{g_0 } } \right ] \star \left [ { \mathbf{y}}^{u}_{f^{u}_{g_0 } } \right]\ ] ] follows from lemma [ fundamental_equivalence_lemma ] : thanks to lemma [ graded_triviality ] we reduce to study two closed admissible @xmath4-colored ribbon graphs @xmath1605 and @xmath1606 whose diagrams differ only in some small region where they look like the @xmath4-colored ribbon graphs represented in the left - hand part and in the right - hand part of figure [ fundamental_equivalence_figure ] respectively . therefore in order to conclude
we just need to notice that @xmath1605 features an additional component of the surgery link which leaves the signature unchanged , so that . ]
@xmath1607 now , for what concerns the morita equivalence , the family of objects @xmath1608 dominates @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] thanks to remark [ shift_isomorphism_remark ] .
if @xmath64 and @xmath1609 are objects of @xmath1545 then let us show that the degree @xmath1135 component @xmath1610{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}^u \left ( { \mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v '' ) } \right)\ ] ] is surjective for every @xmath242 .
thanks to lemma [ graded_connection ] we know that every morphism in @xmath1433{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}^u ( { \mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v '' ) } ) $ ] can be represented by some admissible @xmath4-skein inside @xmath1567 relative to @xmath1611 up to isotopy and skein equivalence we can moreover restrict to admissible @xmath4-skeins of the form @xmath1612 where @xmath205 is a morphism in @xmath1613 . remark that we can suppose that the @xmath1-coloring @xmath1614 evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1615
because if @xmath1616 is the unique @xmath1-coloring of @xmath1617 extending @xmath1618 and @xmath1619 which is compatible with the @xmath4-coloring of @xmath1598 and which evaluates to @xmath1556 the relative homology class of the oriented arc between the two different components of @xmath1573 which is represented in red in figure [ punctured_cylinder ] then we have the equality . ]
@xmath1620 = \psi(g',u ) \cdot \left [ { \mathrm{y}},{\mathrm{t}}_{{\mathrm{y}}}^{f},\omega_{f},0 \right]\ ] ] between vectors of @xmath1433{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}^u({\mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v'')})$ ]
. therefore every degree @xmath1135 morphism of @xmath1433{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}^u({\mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v'')})$ ] is of the form @xmath1621 $ ] for some morphism @xmath205 of @xmath1613 .
now it is enough to show that @xmath1622 = \left [ { \mathbf{y}}^u_{p_{v'',-u } \circ f } \right ] = { \mathbb f}^u_g \left ( r_{v'',-u } \circ f \right)\ ] ] for every @xmath1623 . in order to do
so we can prove the first equality by applying lemma [ graded_triviality ] , and this is done by distinguishing the generic case @xmath253 from the critical case @xmath1624 . in the generic case we have to check that the evaluations of the geer - patureau - turaev renormalized invariant @xmath870 against the admissible closed @xmath4-colored ribbon graphs @xmath1625 and @xmath1626 coincide for every @xmath1627 , for every @xmath1513 and for all @xmath1628 where @xmath1625 is given by the admissible closed ribbon graph represented in the left - hand part of figure [ reduction_cylinder ] . in the critical case we have to check that the evaluations of the geer - patureau - turaev renormalized invariant @xmath870 against the admissible closed @xmath4-colored ribbon graphs @xmath1629 and @xmath1630 coincide for every @xmath1631 , for every @xmath1513 and for all @xmath1632 where @xmath1629 is given by the admissible closed ribbon graph represented in the right - hand part of figure [ reduction_cylinder ] .
these equalities show the surjectivity of @xmath1633 and they also show that for every @xmath1634 the definition of @xmath1635 is actually independent of the choice of the section @xmath1636 of @xmath1637 which was made in remark [ notation_for_composition_with_section ] .
[ reduction_cylinder ] and @xmath1629.,title="fig : " ] finally , in order to see that @xmath1610{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)}^u \left ( { \mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v '' ) } \right)\ ] ] is also injective it is sufficient to use the non - degeneracy of the trace @xmath209 on @xmath1542 : if @xmath1638 is a non - trivial degree @xmath1135 morphism then there exists some morphism @xmath1639
satisfying @xmath1640 . we can therefore choose 2-morphisms @xmath1641 satisfying @xmath1642
we give some results about covariant universal @xmath2-graded linear functors induced by a set of generating 1-morphisms of @xmath17 composed of 2-discs , 2-pants and 2-cylinders .
this allows for the computation of the covariant universal @xmath2-graded vector spaces associated with some elementary closed 1-morphisms of @xmath17 .
we have a commutative diagram of @xmath2-graded linear functors of the form - graded linear functor @xmath1602 is given in theorem [ description_universal_algebroids ] , the @xmath185-disc is introduced in definition [ ( epsilon , v)-disc ] and we use the notation of remark [ notation_constant_functor ] for constant @xmath2-graded linear functors . ] ( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] ; ( p1 ) at ( 90:2 ) @xmath1644 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1645 ; ( p3 ) at ( atan(0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}(\varnothing)$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1646 ( p0 ) ( p2 ) edge[- > ] node[above ] @xmath1647 ( p1 ) ( p2 ) edge[- > ] node[left ] @xmath1648 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1649 ( p0 ) ; for every index @xmath63 of @xmath1 and for every object @xmath64 of @xmath1545 .
let @xmath63 be an index in @xmath1 and let @xmath64 be an object of @xmath62 .
the _ dual @xmath1650-colored 2-disc _
@xmath1651 is the 1-morphism of @xmath17 given by @xmath1652 we denote with @xmath1653 the 3-dimensional cobordism from @xmath140 to @xmath1654 whose support @xmath556 is given by @xmath1655 whose outgoing horizontal boundary identification is induced by the diffeomorphisms @xmath1656 and @xmath1657 & \mapsto & \begin{cases } \left ( -\sqrt{1-x^2-y^2},x , y \right ) & i = 0 \\ \left ( \sqrt{1-x^2-y^2},x , y \right ) & i = 1 \end{cases } \end{array}\ ] ] for every index @xmath63 in @xmath1 and for every object @xmath64 in @xmath62 let @xmath1658 be the @xmath2-graded linear functor mapping every object @xmath85 of @xmath1545 to the @xmath2-graded vector space @xmath1659 whose space of degree @xmath1135 vectors is given by @xmath1660 and mapping every degree @xmath1661 morphism @xmath1662 of @xmath1663 to the @xmath2-graded linear map @xmath1664 whose degree @xmath1135 component @xmath1665 maps every vector @xmath1585 of @xmath1666 to the vector @xmath1667 of @xmath1668 .
then for every object @xmath85 of @xmath1545 we define the @xmath2-graded linear map @xmath1530{bbeta.pdf}}}_{(-,v ) } \right)_{v ' } : { \mathbb{h}\mathrm{om}}_{{\mathcal c}}(v , v ' ) \rightarrow { \mathbb v}_{{\mathcal c}}(\overline{{\mathbf{d}}^2}_{(-,v ) } \circ { \mathbf{d}}^2_{(+,v')})\ ] ] whose degree @xmath1135 component @xmath1669{bbeta.pdf}}}_{(-,v ) } ) ^u_{v'}$ ] maps every vector @xmath1585 of @xmath1666 to the vector . ]
@xmath1670 : = \left [ { \mathrm{z}},{\mathrm{t}}_{{\mathrm{z}}}^{f^u_{g_0}},\omega_{f^u_{g_0}},0 \right]\ ] ] of @xmath1671 , where @xmath1672 is the @xmath4-colored ribbon graph contained in @xmath1673 represented in figure [ sphere_vector_space ] and where @xmath1597 is the unique cohomology class which is compatible with the @xmath4-coloring of @xmath1674 and which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1675 .
[ sphere_vector_space ] vector @xmath1676 $ ] .
the ribbon graph @xmath1674 is represented in blue with blackboard framing .
the cohomology class @xmath1597 is completely determined by the @xmath4-coloring on @xmath1674 and by the specified evaluation against the relative homology class represented by the oriented red curve.,title="fig : " ] [ counit ] the @xmath2-graded linear maps @xmath1669{bbeta.pdf}}}_{(-,v ) } ) _ { v'}$ ] define a @xmath2-graded natural isomorphism - graded linear functor @xmath1499 is given in proposition [ graded_universal_algebroid_empty_equivalence ] . ]
( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}(\varnothing)$ ] ; ( p1 ) at ( atan(0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1677 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1644 ; ( p3 ) at ( 3 * 90:2 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{(-,v)}$ ] ; ( p0 ) edge[- > ] node[right ] @xmath1679 ( p1 ) ( p2 ) edge[- > ] node[above ] @xmath1680 ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath1646 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1681 ( p0 ) ; for every index @xmath63 of @xmath1 and for every object @xmath64 of @xmath1545 .
the result is established by showing that the linear maps @xmath1530{bbeta.pdf}}}_{(-,v ) } \right)^u_{v ' } : { \mathbb{h}\mathrm{om}}_{{\mathcal c}}^u(v , v ' ) \rightarrow { \mathbb v}_{{\mathcal c}}^u(\overline{{\mathbf{d}}^2}_{(-,v ) } \circ { \mathbf{d}}^2_{(+,v')})\ ] ] are isomorphisms for every degree @xmath1135 in @xmath2 and for every object @xmath85 of @xmath1545 and that they are natural with respect to morphisms of @xmath1545 .
both properties are checked by reproducing the proof of theorem [ description_universal_algebroids ] .
the surjectivity and injectivity of @xmath1669{bbeta.pdf}}}_{(-,v ) } ) ^u_{v'}$ ] is shown just like the fullness and faitfhulness of @xmath1682 .
indeed every degree @xmath1135 vector of @xmath1671 is of the form @xmath1683 $ ] for some morphism @xmath1684 and we have @xmath1685 = \left [ { \mathbf{z}}^u_{p_{v',-u } \circ f } \right ] = \left ( { \raisebox{-2.5pt}{\includegraphics[height=\heightof{$\gamma$}]{bbeta.pdf}}}_{(-,v ) } \right)^u_{v ' } \left ( r_{v',-u } \circ f \right).\ ] ] moreover , if @xmath1686 is non - zero then the non - degeneracy of the trace @xmath209 on @xmath197 guarantees the existence of some morphism @xmath1687 satisfying @xmath1640 .
this allows us to deduce the non - triviality of @xmath1676 $ ] .
finally , the naturality of @xmath1313{bbeta.pdf}}}_{(-,v)}$ ] is proved by establishing the equality @xmath1688 \right ) \left ( \left [ { \mathbf{z}}^u_{f^u_{g_0 } } \right ] \right ) \\ = \left [ { \mathbf{z}}^{u + u'}_{\left ( f'^{u ' } \star f^u \right)_{g_0 } } \right]\ ] ] of degree @xmath1359 vectors of @xmath1689 for all degrees @xmath1690 of @xmath2 , for all objects @xmath1691 of @xmath1545 , for every degree @xmath1135 vector @xmath1585 of @xmath1659 and for every degree @xmath1661 morphism @xmath1662 of @xmath1692 .
the result follows from lemma [ fundamental_equivalence_lemma ] , and the proof is completely analogous to the proof of the functoriality of @xmath1602 : thanks to lemma [ graded_triviality_tqft ] we reduce to compare two admissible closed @xmath4-colored ribbon graphs whose diagrams differ only in some small region where they look like the @xmath4-colored ribbon graphs represented in the left - hand part and in the right - hand part of figure [ fundamental_equivalence_figure ] respectively .
the evaluation of the geer - patureau - turaev renormalized invariant @xmath870 against the first admissible closed @xmath4-colored ribbon graph has then to be corrected with a @xmath1693 factor which takes into account the additional surgery component , and we can conclude . we define the _ 2-pant cobordism @xmath1694 _ as the 2-dimensional cobordism from @xmath1695 to @xmath141 whose support @xmath1696 is given by @xmath1697 where @xmath1698 is the open ball of center @xmath1699 and radius @xmath1170 in @xmath1700 , whose incoming boundary identification is given by @xmath1701 and whose outgoing boundary identification is given by @xmath1702 .
we also define the _ dual 2-pant cobordism @xmath1703 _ as the 2-dimensional cobordism from @xmath141 to @xmath1695 whose support is given by @xmath1704 , whose incoming boundary identification is given by @xmath1702 and whose outgoing boundary identification is given by @xmath1705 .
let us consider indices @xmath1550 .
@xmath1706-colored 2-pant _ @xmath1707 is the 1-morphism of @xmath17 given by @xmath1708 where @xmath1709 is the unique @xmath1-coloring of @xmath1710 extending @xmath1711 and @xmath1712 which evaluates to 0 every relative homology class that can be represented by some oriented arc contained in @xmath1713 .
see the left - hand part of figure [ 2-pants_figure ] for a graphical representation .
the _ dual @xmath1706-colored 2-pant _
@xmath1714 is the 1-morphism of @xmath17 given by @xmath1715 see the right - hand part of figure [ 2-pants_figure ] for a graphical representation .
[ 2-pants_figure ] -colored 2-pant @xmath1716 and the dual @xmath1706-colored 2-pant @xmath1717 .
the @xmath1-coloring @xmath1718 is uniquely determined by the specified evaluations against the homology classes represented by the oriented red curves.,title="fig : " ] we begin with the study of the 2-pant cobordism .
for all indices @xmath1719 in @xmath1 let @xmath1720 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{g , g ' } : { \mathbb{p}\mathrm{roj}}({\mathcal c}_g ) { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\mathbb{p}\mathrm{roj}}({\mathcal c}_{g ' } ) \rightarrow { \mathbb{p}\mathrm{roj}}({\mathcal c}_{g+g'})\ ] ] be the @xmath2-graded linear functor mapping every object @xmath1721 of @xmath1722 to the object @xmath191 of @xmath1723 and mapping every degree @xmath1359 morphism of the form @xmath1724 of @xmath1725 to the degree @xmath1359 morphism @xmath1726 of @xmath1727 .
[ identity_ribbon_graph ] for every object @xmath130 of @xmath129 we have a 2-morphism @xmath1728 of @xmath17 given by definition [ t^phi - cylinder ] for the obvious morphism @xmath1729 of @xmath129 . for every object @xmath1721 of @xmath1722
we define @xmath1669{bbeta.pdf}}}_{\nabla_{g , g ' } } ) ^0_{(v , v')}$ ] to be the degree 0 morphism @xmath1730\ ] ] of @xmath1273{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})}^0\left ( { \mathbf{d}}^2_{(+,v \otimes v')},{\mathbf{p}}^2_{g , g ' } \circ \left ( { \mathbf{d}}^2_{(+,v ) } \otimes { \mathbf{d}}^2_{(+,v ' ) } , \right ) \right),\ ] ] where we confuse the objects @xmath1731 and @xmath1732 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})$ ] by identifying them via the degree 0 isomorphism induced by the obvious positive diffeomorphism from @xmath1733 to @xmath1734 .
[ product ] the degree 0 morphisms @xmath1669{bbeta.pdf}}}_{\nabla_{g , g ' } } ) ^0_{(v , v')}$ ] define a @xmath2-graded natural isomorphism - graded linear functor @xmath1735 is given in theorem [ monoidality_tensor_product ] . ]
( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})$ ] ; ( p1 ) at ( 90:2 ) @xmath1736 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1737 ; ( p3 ) at ( 3 * 90:2 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_g \otimes { \mathbf{s}}^1_{g'})$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{\nabla_{g , g'}}$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1738 ( p0 ) ( p2 ) edge[- > ] node[above ] @xmath1643 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}}_{g , g'}$ ] ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath1739 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1740 ( p0 ) ; for all indices @xmath1719 in @xmath1 . for every object @xmath1721 of @xmath1722 the degree 0 morphism @xmath1669{bbeta.pdf}}}_{\nabla_{g , g ' } } ) ^0_{(v , v')}$ ] is clearly invertible . in order to show the naturality of @xmath1313{bbeta.pdf}}}_{\nabla_{g , g'}}$ ] we have to establish the equality @xmath1741 \otimes \left [ { \mathbf{y}}^{u'}_{f'^{u'}_{g_0 } } \right ] \right ) \right ) \ast \left [ ( { \mathbf{d}}^2 \times { \mathbf{i}})_{{\mathrm{t}}^{{\mathrm{id}}_{v \otimes v ' } } } \right ] \\ & = \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{-u - u ' } } \otimes ( { \mathbf{d}}^2 \times { \mathbf{i}})_{{\mathrm{t}}^{{\mathrm{id}}_{v '' \otimes v ' '' } } } \right ] \ast \left [ { \mathbf{y}}^{u+u'}_{\left({\reflectbox{\raisebox{\depth}{\scalebox{1}[-1 ] { \includegraphics[height=\heightof{\scriptsize $ \mathbf{\delta}$}]{bbdelta.pdf}}}}}_{g , g'}^{u+u'}(f^{u } \otimes f'^{u'})\right)_{g_0 } } \right ] \end{aligned}\ ] ] of degree @xmath1359 morphisms of @xmath1273{\scriptsize $ \mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})}^{u+u ' } \left ( { \mathbf{d}}^2_{(+,v \otimes
v')},{\mathbf{p}}^2_{g , g ' } \circ \left ( { \mathbf{d}}^2_{(+,v '' ) } \otimes { \mathbf{d}}^2_{(+,v ' '' ) } , \right ) \right)\ ] ] for every degree @xmath1742 morphism of the form @xmath1743 the proof essentially follows the proof of theorem [ description_universal_algebroids ] : thanks to lemma [ graded_triviality ] we can reduce to compare two admissible closed @xmath4-colored ribbon graphs whose diagrams differ only in some small region . then , since the definition of the @xmath2-graded linear functor @xmath1601{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] is actually independent of the choice of the section @xmath1744 of the epimorphism @xmath1745 which was made in remark [ notation_for_composition_with_section ] , we can make sure the region where they differ contains @xmath4-colored ribbon graphs which look like the ones represented in the left - hand part and in the right - hand part of figure [ fundamental_equivalence_figure ] respectively .
this allows us to conclude .
we choose now some fixed 3-dimensional cobordism with corners @xmath1746 from @xmath1747 to @xmath1748 whose support @xmath3 is given by @xmath1749 whose incoming horizontal boundary @xmath1750 is given by @xmath1751 whose outgoing horizontal boundary @xmath1752 is given by @xmath1753 and whose outgoing vertical boundary @xmath1754 is given by @xmath1755 we move on to the study of the dual 2-pant cobordism , and we have to distinguish two cases : either one of the two colors is generic or both of them are critical .
the two situations can be treated similarly , although the second one is much more complicated .
we begin by analysing the generic case , and the structure of the exposition is meant to highlight the analogies with the critical case . for all generic indices @xmath1756 of @xmath252 , for every arbitrary index @xmath1556 of @xmath1 , for all @xmath1757 and for every object @xmath64 of @xmath1723 we denote with @xmath1758 the @xmath4-colored ribbon graph represented in figure [ uccidetemi_figure_generic ] .
[ uccidetemi_figure_generic ] -colored ribbon graph @xmath1758.,title="fig : " ] [ uccidetemi_lemma_generic ] for all generic indices @xmath1756 of @xmath252 , for every arbitrary index @xmath1556 of @xmath1 , for all @xmath1757 and for every object @xmath64 of @xmath1723 we have the equality @xmath1759 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{g , g'}^{0}({\mathrm{id}}_{v_i}^0 \otimes { \mathrm{id}}_{v_i^ * \otimes v}^0)\ ] ] between degree 0 morphisms of @xmath1760 [ uccidetemi_remark_generic ] let us consider the object @xmath1761 of @xmath1762 .
then the degree 0 morphism @xmath1763 of @xmath1764 is the identity . for every generic index @xmath63 in @xmath252 and for every arbitrary index @xmath1556 in @xmath1 let @xmath1765{bbdelta.pdf}}_{g , g ' } : { \mathbb{p}\mathrm{roj}}({\mathcal c}_{g+g ' } ) \rightarrow { \mathbb{p}\mathrm{roj}}({\mathcal c}_g ) \hat{{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize } } } { \mathbb{p}\mathrm{roj}}({\mathcal c}_{g'})\ ] ] be the @xmath2-graded linear functor mapping every object @xmath64 of @xmath1723 to the object @xmath1766 of the completion of @xmath1722 and mapping every degree @xmath1135 morphism @xmath1767 of @xmath1768 to the degree @xmath1135 morphism @xmath1769 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{-g , g+g'}^{u}({\mathrm{id}}_{v_i^*}^0 \otimes f^u ) ) \right)_{i , j \in { \mathrm{i}}_g}\ ] ] @xmath1770 let us consider for every index @xmath457 in @xmath256 and for every object @xmath64 of @xmath1723 the 2-morphism @xmath1771 of @xmath17 given by @xmath1772 for the admissible @xmath4-skein @xmath1773 represented in figure [ coproduct_generic_figure_1 ] .
[ coproduct_generic_figure_1 ] of @xmath17.,title="fig : " ] then for every object @xmath64 of @xmath1723 we define @xmath1669{bbeta.pdf}}}_{\delta_{g , g ' } } ) ^0_{v}$ ] to be the degree 0 morphism @xmath1774 \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_g\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g } \otimes { \mathbf{s}}^1_{g'})}^0 \left ( \bigoplus_{j \in { \mathrm{i}}_g } { \mathbf{d}}^2_{(+,v_j ) } \otimes { \mathbf{d}}^2_{(+,v_j^ * \otimes v ) } , \overline{{\mathbf{p}}^2}_{g , g ' } \circ { \mathbf{d}}^2_{(+,v ) } \right),\ ] ] where we confuse the objects @xmath1776 and @xmath1777 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})$ ] by identifying them via the degree 0 isomorphism @xmath1778\ ] ] which was introduced in remark [ identity_ribbon_graph ] . [ coproduct_generic_proposition ] the degree 0 morphisms @xmath1669{bbeta.pdf}}}_{\delta_{g , g ' } } ) ^0_{v}$ ] define a @xmath2-graded natural isomorphism ( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_g \otimes { \mathbf{s}}^1_{g'})$ ] ; ( p1 ) at ( 90:2 ) @xmath1737 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1736 ; ( p3 ) at ( 3 * 90:2 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g+g'})$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{\delta_{g , g'}}$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1739 ( p0 ) ( p2 ) edge[- > ] node[above ] @xmath1779{bbdelta.pdf}}}_{g , g'}$ ] ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath1738 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1780 ( p0 ) ; for every generic index @xmath63 in @xmath252 and for every arbitrary index @xmath1556 in @xmath1 .
let us consider the 2-morphisms @xmath1781 of @xmath17 given by @xmath1782 for the ribbon graph @xmath1783 represented in figure [ coproduct_generic_figure_2 ] .
[ coproduct_generic_figure_2 ] of @xmath17.,title="fig : " ] then the inverse of the degree 0 morphism @xmath1774 \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_g\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g } \otimes { \mathbf{s}}^1_{g'})}^0 \left ( \bigoplus_{j \in { \mathrm{i}}_g } { \mathbf{d}}^2_{(+,v_j ) } \otimes { \mathbf{d}}^2_{(+,v_j^ * \otimes v ) } , \overline{{\mathbf{p}}^2}_{g , g ' } \circ { \mathbf{d}}^2_{(+,v ) } \right)\ ] ] is given by the degree 0 morphism @xmath1784 \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & \in { \mathrm{i}}_g \\ j & = 1\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g } \otimes { \mathbf{s}}^1_{g'})}^0 \left ( \overline{{\mathbf{p}}^2}_{g , g ' } \circ { \mathbf{d}}^2_{(+,v ) } , \bigoplus_{i \in { \mathrm{i}}_g } { \mathbf{d}}^2_{(+,v_i ) } \otimes { \mathbf{d}}^2_{(+,v_i^ * \otimes v ) } \right).\ ] ] indeed an admissible @xmath4-skein in @xmath1785 yielding the composition @xmath1786\ ] ] is represented in figure [ coproduct_generic_composition ] , so that the relative modularity condition of definition [ relative_modular_graded_categories_def ] yields the equality @xmath1787 = \left [ ( { \mathbf{d}}^3)^0 \otimes { \mathrm{id}}_{\overline{{\mathbf{p}}^2}_{g , g ' } \circ { \mathbf{d}}^2_{(+,v ) } } \right].\ ] ] [ coproduct_generic_composition ] $ ] .,title="fig : " ] the equality @xmath1788 \\ & = \delta_{ij } \cdot { \boldsymbol{\eta}}_{{\mathbf{s}}^1_g,{\mathbf{s}}^1_{g'}}^0 \left ( \left [ { \mathbf{y}}^0_{\left ( { \mathrm{id}}^0_{v_i } \right)_{g_0 } } \right ] \otimes \left [ { \mathbf{y}}^0_{\left ( { \mathrm{id}}^0_{v_i^ * \otimes v } \right)_{g_0 } } \right ] \right ) \\ & = \delta_{ij } \cdot \left [ ( { \mathbf{d}}^3)^0 \otimes { \mathrm{id}}_{{\mathbf{d}}^2_{(+,v_i ) } } \otimes { \mathrm{id}}_{{\mathbf{d}}^2_{(+,v_i^ * \otimes v ) } } \right ] \end{aligned}\ ] ] follows essentially from lemma [ uccidetemi_lemma_generic ] , and it boils down to the study of the @xmath4-colored ribbon graph @xmath1758 for every @xmath1789 thanks to lemma [ graded_triviality ] . in order to show the naturality of @xmath1313{bbeta.pdf}}}_{\delta_{g , g'}}$ ] we have to establish the equality @xmath1790 \right ) \ast \left [ { \mathbf{x}}_{j , v } \right ] \\ & = \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{-u } } \otimes { \mathbf{x}}_{k , v '' } \right ] \ast { \boldsymbol{\eta}}_{{\mathbf{s}}^1_g,{\mathbf{s}}^1_{g'}}^{u } \left ( \left [ { \mathbf{y}}^{0}_{\left ( { \mathrm{id}}^0_{v_j } \right)_{g_0 } } \right ] \otimes \left [ { \mathbf{y}}^{u}_{\left ( { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1 ] { \includegraphics[height=\heightof{\scriptsize $ \mathbf{\delta}$}]{bbdelta.pdf}}}}}_{-g , g+g'}^{u}({\mathrm{id}}_{v_j^*}^0 \otimes f^u ) ) \right)_{g_0 } } \right ] \right ) \end{aligned}\ ] ] of degree @xmath1135 morphisms of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g } \otimes { \mathbf{s}}^1_{g'})}^0 \left ( \bigoplus_{j \in { \mathrm{i}}_g } { \mathbf{d}}^2_{(+,v_j ) } \otimes { \mathbf{d}}^2_{(+,v_j^ * \otimes v ) } , \overline{{\mathbf{p}}^2}_{g , g ' } \circ { \mathbf{d}}^2_{(+,v '' ) } \right)\ ] ] for every @xmath1791 , for every @xmath242 and for every @xmath1792 . this is done by using lemmas [ graded_triviality ] and [ fundamental_equivalence_lemma ] .
we move on to analyse the critical case .
since the statements are much heavier , we should bear in mind the analogue but simpler generic situation in order to interpet them . for all critical indices @xmath1793 of @xmath3 , for every generic index @xmath1794 of @xmath252 , for all @xmath1795 and for every object @xmath64 of @xmath1796
we denote with @xmath1797 the @xmath4-colored ribbon graph represented in figure [ uccidetemi_figure_generic ] .
[ uccidetemi_figure_critical ] -colored ribbon graph @xmath1797.,title="fig : " ] [ uccidetemi_lemma_critical ] for all critical indices @xmath1793 of @xmath3 , for every generic index @xmath1794 of @xmath252 , for all @xmath1795 and for every object @xmath64 of @xmath1796 there exist degree @xmath1798 morphisms @xmath1799 and degree @xmath1800 morphisms @xmath1801 for @xmath1802 that yield the equality @xmath1803 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{x , x'}^{0}(s^{-u_n}_{n , i , j } \otimes r^{u_n}_{n , i , j , v})\ ] ] between degree 0 morphisms of @xmath1804 [ uccidetemi_remark_critical ] let us consider the object @xmath1805 of @xmath1806 .
then the morphism @xmath1807 of @xmath1808 is an idempotent . indeed for every @xmath1789
the degree 0 morphism @xmath1809 of @xmath1810 is clearly an idempotent thanks to the relative modularity condition of definition [ relative_modular_graded_categories_def ] .
this means @xmath1811 for all @xmath1795 . but
this implies @xmath1812 for every @xmath242 , so that the claim follows from the properties of @xmath1813 .
it will be useful to introduce the following notation : for all @xmath1814 , for all @xmath1815 and for every @xmath1816 we denote with @xmath1817 the degree @xmath1818 morphism @xmath1819 and we denote with @xmath1820 the degree @xmath1821 morphism @xmath1822 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{-x , x+x'}^u \left ( { \mathrm{id}}_{v_k^ * \otimes v_{g_x}}^0 \otimes f^u \right ) \star r^{u_m}_{m , k , i , v}.\ ] ] for all critical indices @xmath1793 in @xmath3 let @xmath1765{bbdelta.pdf}}_{x , x ' } : { \mathbb{p}\mathrm{roj}}({\mathcal c}_{x+x ' } ) \rightarrow { \mathbb{p}\mathrm{roj}}({\mathcal c}_x ) \hat{{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize } } } { \mathbb{p}\mathrm{roj}}({\mathcal c}_{x'})\ ] ] be the @xmath2-graded linear functor mapping every object @xmath64 of @xmath1796 to the object @xmath1823 of the completion of @xmath1824 and mapping every degree @xmath1135 morphism @xmath1767 of @xmath1825 to the degree @xmath1135 morphism @xmath1826 of @xmath1827 let us consider for every index @xmath103 in @xmath1828 and for every object @xmath64 of @xmath1796 the 2-morphism @xmath1829 of @xmath17 given by @xmath1830 for the admissible @xmath4-skein @xmath1831 represented in figure [ coproduct_critical_figure_1 ] .
[ coproduct_critical_figure_1 ] of @xmath17.,title="fig : " ] then for every object @xmath64 of @xmath1796 we define @xmath1669{bbeta.pdf}}}_{\delta_{x , x ' } } ) ^0_{v}$ ] to be the degree 0 morphism @xmath1832 \ast { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{k , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{k , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_{g_x + x}\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x } \otimes { \mathbf{s}}^1_{x'})}^0 \left ( { \mathrm{im}}\left ( { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{i , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{i , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) , \overline{{\mathbf{p}}^2}_{x , x ' } \circ { \mathbf{d}}^2_{(+,v ) } \right),\ ] ] where we confuse the objects @xmath1833 and @xmath1834 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x+x'})$ ] by identifying them via the degree 0 isomorphism @xmath1835\ ] ] and where we confuse the objects @xmath1836 and @xmath1837 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x+x'})$ ] by identifying them via the degree 0 isomorphism @xmath1838.\ ] ] [ coproduct_critical_proposition ] the degree 0 morphisms @xmath1669{bbeta.pdf}}}_{\delta_{x , x ' } } ) ^0_{v}$ ] define a @xmath2-graded natural isomorphism ( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_x \otimes { \mathbf{s}}^1_{x'})$ ] ; ( p1 ) at ( 90:2 ) @xmath1839 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1840 ; ( p3 ) at ( 3 * 90:2 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x+x'})$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{\delta_{x , x'}}$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1841 ( p0 ) ( p2 ) edge[- > ] node[above ] @xmath1779{bbdelta.pdf}}}_{x , x'}$ ] ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath1842 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1843 ( p0 ) ; for all critical indices @xmath1793 in @xmath3 . let us consider the 2-morphisms @xmath1844 of @xmath17 given by @xmath1845 for the ribbon graph @xmath1846 represented in figure [ coproduct_critical_figure_2 ] . [ coproduct_critical_figure_2 ] of @xmath17.,title="fig : " ] then the inverse of the degree 0 morphism @xmath1832 \ast { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{k , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{k , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_{g_x + x}\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x } \otimes { \mathbf{s}}^1_{x'})}^0 \left ( { \mathrm{im}}\left ( { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{i , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{i , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) , \overline{{\mathbf{p}}^2}_{x , x ' } \circ { \mathbf{d}}^2_{(+,v ) } \right),\ ] ] is given by the degree 0 morphism @xmath1847 \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{i , k , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \ast \left [ \overline{{\mathbf{x}}}_{i , v } \right ] \right ) \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_{g_x + x}\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x } \otimes { \mathbf{s}}^1_{x'})}^0 \left ( \overline{{\mathbf{p}}^2}_{x , x ' } \circ { \mathbf{d}}^2_{(+,v ) } , { \mathrm{im}}\left ( { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{i , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^{u_n}_{i , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) \right).\ ] ] indeed an admissible @xmath4-skein in @xmath1785 yielding the composition @xmath1848\ ] ] is represented in figure [ coproduct_critical_composition ] , so that the relative modularity condition of definition [ relative_modular_graded_categories_def ] yields the equality @xmath1849 = \left [ ( { \mathbf{d}}^3)^0 \otimes { \mathrm{id}}_{\overline{{\mathbf{p}}^2}_{x , x ' } \circ { \mathbf{d}}^2_{(+,v ) } } \right].\ ] ] [ coproduct_critical_composition ] $ ] .,title="fig : " ] the equality @xmath1788 \\ & = \sum_{n=1}^n { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}^0_{\left ( s_{n , i , j}^{-u_n } \right)_{g_0 } } \right ] \otimes \left [ { \mathbf{y}}^0_{\left ( r_{n , i , j , v}^{u_n } \right)_{g_0 } } \right ] \right ) \end{aligned}\ ] ] follows essentially from lemma [ uccidetemi_lemma_generic ] , and it boils down to the study of the @xmath4-colored ribbon graph @xmath1797 for every @xmath1789 thanks to lemma [ graded_triviality ] . in order to show the naturality of @xmath1313{bbeta.pdf}}}_{\delta_{x , x'}}$ ] we have to establish the equality @xmath1850 \right ) \ast \left [ { \mathbf{x}}_{j , v } \right ] \\ & = \sum_{\substack{1 { \leqslant}\ell , m { \leqslant}n \\ k \in { \mathrm{i}}_{g_x + x } } } \left [ { \mathrm{id}}_{({\mathbf{s}}^2)^{-u } } \otimes { \mathbf{x}}_{k , v '' } \right ] \ast { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{u } \left ( \left [ { \mathbf{y}}^{-u_{\ell}-u_m}_{\left ( s_{\ell , m , k , j}^{-u_{\ell}-u_m } \right)_{g_0 } } \right ] \otimes \left [ { \mathbf{y}}^{u_{\ell } + u + u_m}_{\left ( r_{\ell , m , k , j , f^u}^{u_{\ell } + u + u_m } \right)_{g_0 } } \right ] \right ) \end{aligned}\ ] ] of degree @xmath1135 morphisms of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{x } \otimes { \mathbf{s}}^1_{x'})}^0 \left ( { \mathrm{im}}\left ( { \boldsymbol{\eta}}_{{\mathbf{s}}^1_x,{\mathbf{s}}^1_{x'}}^{0 } \left ( \left [ { \mathbf{y}}_{\left ( s^{-u_n}_{i , j } \right)_{g_0}}^{-u_n } \right ] \otimes \left [ { \mathbf{y}}_{\left ( r^ { * u_n}_{i , j , v } \right)_{g_0}}^{{u_n } } \right ] \right ) \right ) , \overline{{\mathbf{p}}^2}_{x , x ' } \circ { \mathbf{d}}^2_{(+,v '' ) } \right)\ ] ] for every @xmath1851 , for every @xmath242 and for every @xmath1852
. this is done by using lemmas [ graded_triviality ] and [ fundamental_equivalence_lemma ] .
the last cobordisms we are left to analyse are somewhat simpler than the preceding ones .
the _ @xmath1706-colored 2-cylinder _
@xmath1853 is the 1-morphism of @xmath17 given by @xmath1854 where @xmath1855 is the unique @xmath1-coloring of @xmath1856 extending @xmath1857 and @xmath1857 which satisfies @xmath1858 .
see figure [ 2-cylinder_figure ] for a graphical representation .
[ 2-cylinder_figure ] .,title="fig : " ] for all indices @xmath1719 in @xmath1 let @xmath1859 be the @xmath2-graded linear functor given by the identity on objects of @xmath1545 and by a factor @xmath1860 rescaling on degree @xmath1135 morphism vector spaces of @xmath1545 . for all indices @xmath1719 in @xmath1 and for every object @xmath64 of @xmath1545
we define @xmath1669{bbeta.pdf}}}_{i_{g , g ' } } ) ^0_{v}$ ] to be the degree 0 morphism @xmath1861 \in { \mathrm{hom}}_{\hat{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g})}^0 \left ( { \mathbf{d}}^2_{(+,v ) } , \left ( { \mathbf{i}}_{g ' } \times { \mathbf{s}}^1_{g } \right ) \circ { \mathbf{d}}^2_{(+,v ) } \right)\ ] ] determined by the 2-morphism @xmath1566 of @xmath17 which was introduced in remark [ shift_isomorphism_remark ] and which is represented in figure [ shift_isomorphism_remark ] , where we confuse the objects @xmath1565 and @xmath1862 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_g)$ ] by identifying them via the degree 0 isomorphism induced by any positive diffeomorphism from @xmath1733 to @xmath1863 .
[ 2-cylinder_isomorphism_figure ] the degree 0 morphisms @xmath1669{bbeta.pdf}}}_{i_{g , g ' } } ) ^0_{v}$ ] define a @xmath2-graded natural isomorphism ( p0 ) at ( atan(-0.5):sqrt((1.9)^2+(2 * 1.9)^2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g})$ ] ; ( p1 ) at ( 90:1.9 ) @xmath1864 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt((1.9)^2+(2 * 1.9)^2 ) ) @xmath1864 ; ( p3 ) at ( 3 * 90:1.9 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g})$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{i_{g , g'}}$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1646 ( p0 ) ( p2 ) edge[- > ] node[above ] @xmath1865 ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath1646 ( p3 ) ( p3 ) edge[- > ] node[below ] @xmath1866 ( p0 ) ; for all indices @xmath1719 in @xmath1 .
the degree 0 morphism @xmath1669{bbeta.pdf}}}_{i_{g , g ' } } ) ^0_{v}$ ] is clearly invertible for every object @xmath64 of @xmath1545 and the equality @xmath1867 \\ & = \psi(-g',u ) \cdot \left [ \left ( { \mathrm{id}}_{({\mathbf{s}}^2)^{-u } } \otimes \left ( { \mathbf{d}}^2_{(+,v ) } \times { \mathbf{i}}_{g ' } \right ) \right ) \ast { \mathbf{y}}^u_{f^u_{g_0 } } \right ] \end{aligned}\ ] ] of degree @xmath1135 morphisms of @xmath1433{\scriptsize $ \mathbb v$}}}}({\mathbf{s}}^1_g)}({\mathbf{d}}^2_{(+,v)},{\mathbf{d}}^2_{(+,v'')})$ ] , which holds for every degree @xmath1135 morphism @xmath1585 of @xmath1868 , follows from remarks which were made during the proof of theorem [ description_universal_algebroids ] .
we are missing one last family of generating 1-morphisms of @xmath17 from our description : indeed for every generic index @xmath63 of @xmath252 the @xmath63-colored 1-sphere @xmath765 of @xmath17 is dualizable , but the left counit given by the @xmath63-colored 2-cylinder @xmath1869 , which is admissible , can not be obtained from the previous 1-morphisms because @xmath1870 is not admissible .
however we do not need any extra work in order to figure out the covariant universal @xmath2-graded linear functor @xmath1871 .
let us consider the object @xmath1765{bbdelta.pdf}}_{g ,- g}({\mathbbm 1 } ) : = \bigoplus_{i \in { \mathrm{i}}_g } ( v_i , v_i^*)\ ] ] of @xmath1872 . for every index @xmath457 in @xmath256 , let us consider the 2-morphism @xmath1873 of @xmath17 given by @xmath1874 for the admissible @xmath4-skein @xmath1875 represented in figure [ counit_generic_figure ] .
[ counit_generic_figure ] of @xmath17.,title="fig : " ] then we define @xmath1313{bbeta.pdf}}}_{\delta_{g ,- g}({\mathbbm 1})}^0 $ ] to be the degree 0 morphism @xmath1876 \right ) _ { { \vcenter { \let@ \restore@math@cr \default@tag \baselineskip\fontdimen10 \scriptfont\tw@ \advance\baselineskip\fontdimen12 \scriptfont\tw@ \lineskip\thr@@\fontdimen8 \scriptfont\thr@@ \lineskiplimit\lineskip \ialign{\hfil$\m@th\scriptstyle##$&$\m@th\scriptstyle{}##$\crcr i & = 1 \\ j & \in { \mathrm{i}}_g\crcr } } } } \ ] ] of @xmath1775{\scriptsize $ \mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_{g } \otimes { \mathbf{s}}^1_{-g})}^0 \left ( \bigoplus_{j \in { \mathrm{i}}_g } { \mathbf{d}}^2_{(+,v_j ) } \otimes { \mathbf{d}}^2_{(+,v_j ) } , \overline{{\mathbf{p}}^2}_{g ,- g } \circ { \mathbf{d}}^2 \right),\ ] ] where we confuse the objects @xmath1877 of @xmath1271{$\mathbb v$}}}}_{{\mathcal c}}({\mathbf{s}}^1_{-g})$ ] and @xmath1878 by identifying them via the degree 0 isomorphism @xmath1879\ ] ] which was introduced in remark [ identity_ribbon_graph ] .
therefore we get a @xmath2-graded natural isomorphism ( p0 ) at ( atan(-0.5):sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}({\mathbf{s}}^1_g \otimes { \mathbf{s}}^1_{-g})$ ] ; ( p1 ) at ( 90:2 ) @xmath1880 ; ( p2 ) at ( atan(-0.5 ) + 2 * 90:sqrt(2 ^ 2 + 4 ^ 2 ) ) @xmath22 ; ( p3 ) at ( 3 * 90:2 ) @xmath1643{$\mathbb v$}}}}}_{{\mathcal c}}(\varnothing)$ ] ; at ( 0,0 ) @xmath1540 ; at ( 0,0 ) @xmath1678{bbeta.pdf}}}}_{\delta_{g ,- g}({\mathbbm 1})}$ ] ; ( p1 ) edge[- > ] node[right , xshift=2ex ] @xmath1881 ( p0 ) ( p2 ) edge[- > ] node[above , yshift=1ex ] @xmath1779{bbdelta.pdf}}}_{g , g'}({\mathbbm 1})$ ] ( p1 ) ( p2 ) edge[- > ] node[left , xshift=-2ex ] @xmath547 ( p3 ) ( p3 ) edge[- > ] node[below , yshift=-1ex ] @xmath1882 ( p0 ) ; for every generic index @xmath63 in @xmath252 . for every @xmath1883 we fix a standard closed surface @xmath1884 of genus @xmath574 and a standard ordered basis @xmath1885 of @xmath1886 satisfying @xmath1887 for all @xmath1888 .
if @xmath1889 is an element of @xmath1890 then we denote with @xmath1891 the element of @xmath1892 given by @xmath1893 . for all @xmath1894
we denote with @xmath1895 the 1-morphism of @xmath26 given by @xmath1896 where the base set @xmath1897 is composed of a single point , where the cohomology class @xmath1898 is determined by the evaluations @xmath1899 and @xmath1900 for every @xmath1901 and where the lagrangian subspace @xmath1902 is generated by the homology classes @xmath1903 .
see figure [ generic_surface_figure ] for a graphical representation .
[ generic_surface_figure ] of @xmath26.,title="fig : " ] we denote with @xmath1904 the finite set @xmath1905 an element of @xmath1904 is denoted @xmath1906 and @xmath1907 denotes the subset of @xmath1904 given by @xmath1908 if @xmath1909 and @xmath1910 then @xmath1911 is isomorphic to @xmath1912 the 1-morphism @xmath1913 of @xmath17 can be decomposed , up to isomorphism , as @xmath1914 for the 1-morphisms @xmath1915 of @xmath17 represented figure [ xi ] .
[ xi ] of @xmath17.,title="fig : " ] then thanks to proposition [ coproduct_generic_proposition ] we can conclude .
let @xmath64 be an object of @xmath1916 and let @xmath1917 denote the 1-morphism of @xmath17 given by @xmath1918 where the @xmath4-colored ribbon set @xmath1919 is given by a single positive point with color @xmath64 .
if @xmath1129 is a critical index in @xmath3 , if @xmath1794 is an arbitrary index in @xmath1 and if @xmath64 is an object of @xmath1916 then for all indices @xmath1920 in @xmath1921 we denote with @xmath1922 the @xmath2-graded linear map whose degree @xmath1135 component @xmath1923 maps every vector @xmath1924 of @xmath1925 to the vector @xmath1926 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{x ,- x}^{0}(s^{-u_n}_{n , i , j } \otimes r^{u_n}_{n , i , j , v } ) \star f^u_{j , v } \right)\ ] ] of @xmath1927 where , if @xmath1789 , we have @xmath1928 { \includegraphics[height=\heightof{$\mathbf{\delta}$}]{bbdelta.pdf}}}}}_{x , x'}^{0}(s^{-u_n}_{n , i , j } \otimes r^{u_n}_{n , i , j , v})\ ] ] for the @xmath4-colored ribbon graph @xmath1797 represented in figure [ uccidetemi_figure_critical ]
. if @xmath1793 are critical indices in @xmath3 and if @xmath64 is an object of @xmath1916 then @xmath1929 is isomorphic to @xmath1930 the 1-morphisms @xmath1931 of @xmath17 can be decomposed , up to isomorphism , as @xmath1932 then thanks to proposition [ coproduct_critical_proposition ] we can conclude .
let @xmath45 be a natural number .
[ def_cobordism ] if @xmath329 and @xmath336 are oriented @xmath1933-dimensional closed manifolds a _ @xmath1934-dimensional cobordism @xmath67 from @xmath329 to @xmath336 _ is a 5-tuple @xmath1935 where : 1 .
@xmath345 is a smooth oriented @xmath1934-dimensional compact manifold with boundary called the _ support _ whose boundary is decomposed as @xmath1936 with @xmath1937 ; 2 .
@xmath1938 is a positive diffeomorphism called the _ incoming boundary identification _ and @xmath1939 is a positive diffeomorphisms called the _ outgoing boundary identification _ ; 3 .
@xmath1940 is a collar for @xmath350 ; 4 .
@xmath1941 is a collar for @xmath351 .
two cobordisms @xmath67 and @xmath75 from @xmath329 to @xmath336 are _ isomorphic _ if there exists a positive diffeomorphism @xmath1942 such that : 1 . @xmath1943
@xmath1944 agrees with @xmath1945 in a neighborhood of @xmath1946 ; 3 .
@xmath1947 agrees with @xmath1948 in a neighborhood of @xmath1949 .
such a diffeomorphism @xmath205 is called an _ isomorphism of cobordisms _ and is denoted @xmath1950 .
if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath336 and @xmath1951 is a smooth oriented @xmath1934-dimensional compact manifold with boundary then every positive diffeomorphism @xmath1942 induces an isomorphism of cobordisms between @xmath67 and @xmath1952 [ trivial_cobordism ] every oriented smooth @xmath1933-dimensional closed manifold @xmath329 determines a @xmath1934-dimensional cobordism @xmath353 from @xmath329 to itself called the _
trivial cobordism of @xmath329 _ and given by @xmath1953 where @xmath1954 and @xmath1955 project to the identity on the second component and where @xmath1956 and @xmath145 are some fixed embeddings satisfying @xmath1957 for every @xmath1958 and @xmath1959 for every @xmath1960 .
[ opposite_cobordism ] if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath336 then we will always denote with @xmath1009 the _ opposite _ cobordism from @xmath336 to @xmath329 given by @xmath1961 where : 1 .
@xmath1962 and @xmath1963 ; 2 .
@xmath1964 for all @xmath1965 and all @xmath1966 ; 3 .
@xmath1967 for all @xmath1968 and all @xmath1969 .
whenever we have an oriented @xmath1934-dimensional closed manifold @xmath345 we will denote with @xmath67 the only @xmath1934-dimensional cobordism from @xmath140 to @xmath140 whose support is given by @xmath345 .
if @xmath67 is a cobordism then we will speak of topological properties of its support @xmath345 as if they were properties of @xmath67 itself .
for instance if @xmath345 is connected then we will say @xmath67 is connected . if @xmath329 and @xmath336 are oriented @xmath1933-dimensional closed manifolds and @xmath67 and @xmath75 are @xmath1934-dimensional cobordisms from @xmath329 to @xmath336 a _
@xmath45-dimensional cobordism with corners @xmath80 from @xmath67 to @xmath75 _ is a 9-tuple @xmath1970 where : 1 .
@xmath7 is an oriented @xmath45-dimensional compact @xmath1971-manifold whose underlying manifold with faces is called the _ support _ , whose horizontal boundary is decomposed as @xmath1972 with @xmath1973 and whose vertical boundary is decomposed as @xmath1974 with @xmath1975 ; 2 .
@xmath1976 is a positive diffeomorphism called the _ incoming horizontal boundary identification _ , @xmath1977 is a positive diffeomorphism called the _ outgoing horizontal boundary identification _ , @xmath1978 is a positive diffeomorphism called the _ incoming vertical boundary identification _ , @xmath1979 is a positive diffeomorphism called the _ outgoing vertical boundary identification _ such that @xmath1980 for all @xmath1969 and @xmath1981 for all @xmath1966 ; 3 .
@xmath1982 is a collar for @xmath1983 agreeing with @xmath1984 and @xmath1985 in a neighborhood of @xmath1986 ; 4 .
@xmath1987 is a collar for @xmath1988 agreeing with @xmath1945 and @xmath1989 in a neighborhood of @xmath1986 ; 5 .
@xmath1990 is a collar for @xmath1991 agreeing with @xmath1992 and @xmath1993 in a neighborhood of @xmath1994 ; 6 .
@xmath1995 is a collar for @xmath1996 agreeing with @xmath1948 and @xmath1997 in a neighborhood of @xmath1994 . two cobordisms with corners @xmath80 and @xmath117 from @xmath67 to @xmath75 are _ isomorphic _ if there exists an isomorphism of oriented @xmath1971-manifolds @xmath1998 such that : 1 . @xmath1999 ; 2 . @xmath2000
@xmath2001 agrees with @xmath2002 in a neighborhood of @xmath1986 ; 4 .
@xmath2003 agrees with @xmath2004 in a neighborhood of @xmath1994 .
such an @xmath205 is called an _ isomorphism of cobordisms with corners _ and is denoted @xmath380 .
if @xmath80 is a @xmath45-dimensional cobordism with corners from @xmath67 to @xmath75 and @xmath2005 is an oriented @xmath45-dimensional compact @xmath1971-manifold then every isomorphism of oriented @xmath1971-manifolds @xmath1998 induces an isomorphism of cobordisms with corners between @xmath80 and @xmath2006 every @xmath1934-dimensional cobordism @xmath67 from @xmath329 to @xmath336 determines a @xmath45-dimensional cobordism with corners @xmath2007 from @xmath67 to itself called the _
trivial cobordism with corners of @xmath67 _ and given by @xmath2008 whenever we have a @xmath45-dimensional compact manifold @xmath7 with boundary @xmath2009 where @xmath2010 we denote with @xmath80 the only @xmath45-dimensional cobordism with corners from @xmath2011 to @xmath2012 whose support is given by @xmath7 with incoming horizontal boundary identification given by @xmath2013 and with outgoing horizontal boundary identification given by @xmath2014 .
since we use the same notation for all possible decompositions of the boundary , we will specify incoming and outgoing horizontal boundaries whenever our choice will not be evident .
if @xmath80 is a cobordism with corners then we will speak of topological properties of its support @xmath7 as if they were properties of @xmath80 itself .
for instance if @xmath7 is connected then we will say @xmath80 is connected .
if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath336 and if @xmath75 is a @xmath1934-dimensional cobordism from @xmath336 to @xmath2015 then we denote with @xmath2016 the @xmath1934-dimensional cobordism from @xmath329 to @xmath2015 given by @xmath2017 with gluing data @xmath2018 . if @xmath67 and @xmath116 are @xmath1934-dimensional cobordisms from @xmath329 to @xmath336 , if @xmath75 and @xmath118 are @xmath1934-dimensional cobordisms from @xmath336 to @xmath2015 , if @xmath80 is a @xmath45-dimensional cobordism with corners from @xmath67 to @xmath116 and if @xmath117 is a @xmath45-dimensional cobordism with corners from @xmath75 to @xmath118 then we denote with @xmath2019 the @xmath45-dimensional cobordism with corners from @xmath2016 to @xmath2020 given by @xmath2021 where : 1
. the gluing data is given by the @xmath2022-dimensional manifold with boundary @xmath376 together with the diffeomorphisms @xmath373 and @xmath2023 and with any pair of collars @xmath2024 for @xmath373 and @xmath2023 respectively such that @xmath2025 agrees with @xmath2026 in a neighborhood of @xmath2027 and with @xmath2028 in a neighborhood of @xmath2029 and such that @xmath2030 agrees with @xmath2031 in a neighborhood of @xmath2032 and with @xmath2033 in a neighborhood of @xmath2034 ; 2 .
the boundary of @xmath2035 is decomposed as @xmath2036 [ vertical_gluing ] if @xmath67 , @xmath75 and @xmath116 are @xmath1934-dimensional cobordisms from @xmath329 to @xmath336 , if @xmath80 is a @xmath45-dimensional cobordism with corners from @xmath67 to @xmath75 and if @xmath117 is a @xmath45-dimensional cobordism with corners from @xmath75 to @xmath116 then we denote with @xmath2037 the @xmath45-dimensional cobordism with corners from @xmath67 to @xmath116 given by @xmath2038 where : 1 .
the gluing data is given by the @xmath2022-dimensional manifold with boundary @xmath1951 together with the diffeomorphisms @xmath371 and @xmath2039 and with any pair of collars @xmath2040 for @xmath371 and @xmath2039 respectively such that @xmath2041 agrees with @xmath2042 in a neighborhood of @xmath2043 and with @xmath2028 in a neighborhood of @xmath2029 and such that @xmath2044 agrees with @xmath2031 in a neighborhood of @xmath2045 and with @xmath2046 in a neighborhood of @xmath2027 ; 2 .
the boundary of @xmath2047 is decomposed as @xmath2048 3 .
the diffeomorphisms @xmath2049 and @xmath2050 are defined as @xmath2051 & t { \leqslant}\frac 12 \\ [ ( x,2\psi(t)-1),1 ] & t { \geqslant}\frac 12 \end{cases } \end{array } \\ \begin{array}{rccc } u_{\vargamma ' \times i } : & \vargamma ' \times i & \rightarrow & ( \vargamma ' \times i ) \cup_{\vargamma ' } ( \vargamma ' \times i ) \\ & ( x',t ) & \mapsto & \begin{cases } [ ( x',2\psi(t)),0 ] & t { \leqslant}\frac 12 \\ [ ( x',2\psi(t)-1),1 ] & t { \geqslant}\frac 12 \end{cases } \end{array } \end{gathered}\ ] ] where @xmath2052 is a diffeomorphism satisfying @xmath2053 for all @xmath2054 , @xmath2055 for all @xmath2056 and @xmath2057 . [ left_wedge_cobordism ]
if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath336 then the _ left wedge cobordism with corners @xmath2058 _ is the @xmath45-dimensional cobordism with corners from @xmath2059 to @xmath67 given by @xmath2060 where @xmath2061 coincides with @xmath2062 on @xmath2063 and with @xmath2064 & \mapsto & f_{\varsigma_-}(\psi_+(t_+),x ) \\ \left [ t , x \right ] & \mapsto & f_{\varsigma_-}(t , x ) , \end{array}\ ] ] elsewhere for some fixed embedding @xmath2065 satisfying @xmath2066 for all @xmath2067 and @xmath2068 for all @xmath2069 .
[ right_wedge_cobordism ] if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath336 then the _ right wedge cobordism with corners @xmath2070 _ is the @xmath45-dimensional cobordism with corners from @xmath2071 to @xmath67 given by @xmath2072 where @xmath2073 coincides with @xmath2062 on @xmath2074 and with @xmath2075 & \mapsto & f_{\varsigma_+}(\psi_-(t_-),x ' ) \\
\left [ t , x ' \right ] & \mapsto & f_{\varsigma_+}(t-1,x ' ) , \end{array}\ ] ] elsewhere for some fixed embedding @xmath2076 satisfying @xmath2077 for all @xmath2078 and @xmath2079 for all @xmath2080 . if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath2015 and if @xmath75 is a @xmath1934-dimensional cobordism from @xmath336 to @xmath2081 then we denote with @xmath2082 the @xmath1934-dimensional cobordism and @xmath2083 are sets we denote with @xmath2084 the set @xmath2085 , although most of the time we will suppress references to the first coordinate if there is no risk for confusion . ] from @xmath2086 to @xmath2087 given by @xmath2088 if @xmath80 is a @xmath45-dimensional cobordism with corners from @xmath67 to @xmath116 and if @xmath117 is a @xmath45-dimensional cobordism from @xmath75 to @xmath118 then we denote with @xmath2089 the @xmath45-dimensional cobordism with corners from @xmath2082 to @xmath2090 given by @xmath2091 [ flip_cobordism ] if @xmath329 and @xmath336 are oriented smooth @xmath1933-dimensional closed manifolds then the _ flip cobordism @xmath508 of @xmath329 and @xmath336 _ is the @xmath1934-dimensional cobordism from @xmath2086 to @xmath2092 given by @xmath2093 where we use the notation of remark [ trivial_cobordism ] and @xmath2094 [ flip_cobordism_with_corners ] if @xmath67 is a @xmath1934-dimensional cobordism from @xmath329 to @xmath2015 and if @xmath75 is a @xmath1934-dimensional cobordism from @xmath336 to @xmath2081 then the _ flip cobordism with corners @xmath511 of @xmath67 and @xmath75 _ is the @xmath45-dimensional cobordism from @xmath2095 to @xmath2096 given by @xmath2097 where we use the notation of definitions [ left_wedge_cobordism ] , [ right_wedge_cobordism ] and [ flip_cobordism ] .
our notation will differ from the one used in @xcite for what concerns the orientations of edges .
this is done in order to have a coherent notation for the orientation induced on the boundary throughout the exposition . if @xmath67 is a 2-dimensional cobordism then an _ oriented vertex set @xmath1011 inside @xmath67 _ , denoted @xmath2098 , is an embedded oriented discrete set inside @xmath345 disjoint from @xmath2099 whose points are called _
vertices_. the empty set is admitted as an oriented vertex set .
let @xmath2100 denote the trivial cobordism of the oriented 0-dimensional manifold given by the set @xmath314 with positive orientation and let @xmath2101 denote the trivial cobordism with corners of @xmath2100 .
if @xmath7 is a 3-dimensional manifold with corners then a _ coupon @xmath91 inside @xmath7 of type @xmath98 _ is the given of : 1 .
an embedded oriented 2-dimensional manifold with faces @xmath108 inside @xmath7 disjoint from @xmath834 ; 2 . a diffeomorphism of manifolds with faces @xmath2102 specifying an orientation for @xmath108 and a structure @xmath91 of cobordism with corners from @xmath2100 to @xmath2100 on @xmath108 distinguishing an incoming horizontal boundary @xmath2103 and an outgoing horizontal boundary @xmath2104 called the _ bottom base of @xmath108 _ and the _ top base of @xmath108 _ respectively ; 3 . a set @xmath2105 of @xmath720 oriented points lying along the bottom base at @xmath2106 called the _ input set _ ; 4 . a set @xmath2107 of @xmath2108 oriented points lying along the top base at @xmath2109 called the _ output set_. the empty set is admitted both as an input and as an output set .
if @xmath7 and @xmath2005 are 3-dimensional manifolds with faces and if @xmath91 is a coupon of type @xmath98 inside @xmath7 then every positive embedding @xmath2110 naturally induces a coupon of type @xmath98 inside @xmath2005 which will be denoted @xmath2111 .
if @xmath7 is a 3-dimensional manifold with faces we will say an embedded 1-dimensional manifold with boundary @xmath2112 intersects nicely an embedded 2-dimensional manifold with faces @xmath2113 if @xmath2114 and if for every point in this intersection the tangent vector to @xmath3 is also tangent to @xmath427 but not to @xmath2115 . if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 and if @xmath2098 and @xmath2116 are oriented vertex sets then an _ oriented graph @xmath101 inside @xmath80 from @xmath1011 to @xmath2117 _ , denoted @xmath2118 , is the given of : 1
a finite set of oriented embedded smooth arcs inside @xmath7 called the _ edges of @xmath101 _ ; 2 . a finite set of coupons inside @xmath7 .
these data satisfy : 1 .
the union of all edges @xmath2119 intersects @xmath834 transversely along the union of @xmath2120 and @xmath2121 ; 2 .
the union of all edges @xmath2119 intersects @xmath108 nicely along the union of @xmath2122 and @xmath2123 for every coupon @xmath91 ; 3 .
the orientation obtained by placing a positive tangent vector to @xmath2119 before a positive frame of @xmath2124 agrees with the orientation of @xmath7 at positive vertices and disagrees at negative vertices of @xmath2120 and @xmath2121 .
4 . the orientation obtained by placing a positive tangent vector to @xmath2119 before a positive tangent vector to @xmath2125 agrees with the orientation of @xmath108 at positive points and disagrees at negative points of @xmath2122 and @xmath2123 for every coupon @xmath91 . we will use the notation @xmath89 for edges of @xmath101 and @xmath92 for coupons of @xmath101 . if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath116 , if @xmath117 is a 3-dimensional cobordism with corners from @xmath75 to @xmath118 and if @xmath2118 is an oriented graph from @xmath2098 to @xmath2116 then every time we have positive embedding of surfaces @xmath76 and @xmath119 and a positive embedding of manifolds with faces @xmath120 satisfying @xmath121 we obtain an oriented graph @xmath2126 from @xmath2127 to @xmath2128 with an edge @xmath126 for every edge @xmath89 and with a coupon @xmath127 for every coupon @xmath92 .
every oriented vertex set @xmath2098 naturally determines an oriented graph @xmath2129 from @xmath1011 to itself inside @xmath2007 called the _ trivial oriented graph on @xmath1011 _ which features an edge @xmath2130 for every vertex @xmath73 with orientation determined by the orientation of @xmath72 . if @xmath67 is a 2-dimensional cobordism then a _ ribbon set @xmath69 inside @xmath67 _ , denoted @xmath68 , is the given of : 1 .
an oriented vertex set @xmath2098 ; 2 . a framing given by a non - zero tangent vector @xmath2131 for all @xmath73 .
a vertex together with its framing is called a _ ribbon vertex_. the empty set is admitted as a ribbon set .
if @xmath67 and @xmath75 are 2-dimensional cobordisms and if @xmath68 is a ribbon set then every positive embedding of surfaces @xmath76 induces a ribbon set @xmath2132 given by @xmath2127 with framing @xmath2133 at @xmath79 for every vertex @xmath73 .
if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath75 and if @xmath68 and @xmath358 are ribbon sets then a _ ribbon graph @xmath84 inside @xmath80 from @xmath69 to @xmath83 _ , denoted @xmath82 , is the given of : 1 .
an oriented graph @xmath2118 ; 2 . a framing given by a nowhere - vanishing normal vector field @xmath2134 along every edge @xmath89 .
these data satisfy : 1 .
the framing of @xmath101 equals the framing of @xmath2120 and @xmath2121 at their intersection .
the framing of @xmath101 equals the positive unit tangent vector of @xmath2125 at positive input and output points and equals the negative unit tangent vector of @xmath2125 at negative input and output points of @xmath108 for every coupon @xmath91 .
the empty set is considered to be a ribbon graph .
a ribbon graph between empty ribbon sets will be called a _ closed ribbon graph _ , a ribbon graph without coupons will be called a _ ribbon tangle _ , a ribbon tangle between empty ribbon sets will be called a _ framed link _ and a connected framed link will be called a _ framed knot_. if @xmath80 is a 3-dimensional cobordism with corners from @xmath67 to @xmath116 , if @xmath117 is a 3-dimensional cobordism with corners from @xmath75 to @xmath118 and if @xmath82 is a ribbon graph from @xmath68 to @xmath358 then every time we have positive embedding of surfaces @xmath76 and @xmath119 and a positive embedding of manifolds with faces @xmath120 satisfying @xmath121 we obtain a ribbon graph @xmath2135 from @xmath78 to @xmath2136 given by @xmath2137 with framing @xmath2138 along @xmath126 for every edge @xmath89 .
every ribbon set @xmath68 naturally determines a ribbon tangle @xmath2139 from @xmath69 to itself inside @xmath2007 called the _
trivial ribbon tangle on @xmath69 _ whose oriented graph is given by @xmath2140 with framing @xmath2141 along each edge @xmath2130 for the natural projection @xmath2142 .
let @xmath551 denote the monoidal category of complex vector spaces and complex linear maps .
a _ linear category _ , also called _ algebroid _ , is a @xmath551-enriched category @xmath1263 .
a _ linear functor _ is a @xmath551-enriched functor @xmath2143 between linear categories . if @xmath1263 and @xmath2144 are linear categories then the category @xmath2145 whose objects are linear functors @xmath2146 and whose morphisms are natural transformations @xmath2147 naturally admits the structure of a linear category .
if @xmath4 is a category then the _ free linear category generated by @xmath4 _ is defined as the linear category having the same set of objects as @xmath4 and having the free complex vector space generated by the set @xmath2148 as vector space of morphisms between objects @xmath1129 and @xmath2149 . if @xmath2146 is a linear functor between linear categories @xmath1263 and @xmath2144 then the _ quotient linear category _
@xmath2150 is defined as the linear category whose set of objects is given by @xmath2151 and whose vector spaces of morphisms are given by is the linear map determined by the linear functor @xmath2143 . ]
@xmath2152 for all @xmath2153 .
the _ unit algebroid @xmath22 _ is the @xmath551-enriched unit category is a monoidal category then the @xmath2154-enriched unit category is the @xmath2154-enriched category having a single object with @xmath2154-object of morphisms given by the unit object @xmath38 of @xmath2154 . ] .
the _ tensor product @xmath2155 of linear categories @xmath1263 and @xmath2144 _ is the @xmath551-enriched tensor product of linear categories .
the _ trivial braiding functor @xmath2156 associated with linear categories @xmath1263 and @xmath2144 _ is the linear functor @xmath2157 induced by the trivial braiding of vector spaces .
the symmetric monoidal 2-category @xmath553 of linear categories is the 2-category whose objects @xmath1263 are linear categories , whose 1-morphisms @xmath2146 are linear functors and whose 2-morphisms @xmath2147 are natural transformations , with monoidal structure given by the unit algebroid @xmath22 , by the tensor product of linear categories @xmath2158 and by the trivial braiding @xmath51 of linear categories .
a linear category is _ complete _ if it is closed under finite direct sums and if all its idempotents split .
sometimes complete linear categories are called cauchy complete .
the _ additive completion @xmath2159 of a linear category @xmath1263 _ is the linear category whose set of objects @xmath2160 is given by @xmath2161 and whose morphism vector spaces @xmath2162 are given by @xmath2163 for all @xmath2164 with @xmath2165 and @xmath2166 .
the identity morphism of an object @xmath2167 is given by the morphism @xmath2168 and the composition of a morphism @xmath2169 with a morphism @xmath2170 is given by the morphism @xmath2171 if @xmath1129 is an object of a linear category @xmath1263 then we will denote with @xmath1129 the object @xmath2172 of @xmath2159 too .
the _ idempotent completion @xmath2173 of a linear category @xmath1263 _ is the linear category whose set of objects @xmath2174 is given by @xmath2175 and whose morphism vector spaces @xmath2176 are given by @xmath2177 for all @xmath2178 .
the identity of an object @xmath2179 is given by the idempotent morphism @xmath72 and compositions are directly inherited from compositions in @xmath1263 .
if @xmath1129 is an object of a linear category @xmath1263 then we will denote with @xmath1129 the object @xmath2180 of @xmath2173 too .
the _ completion @xmath2181 of a linear category @xmath1263 _ is defined as the linear category @xmath2182 .
[ completion_remark ] every object of the completion @xmath2181 of a linear category @xmath1263 is of the form @xmath2183 for some natural number @xmath2184 , for some objects @xmath2185 of @xmath1271{$\mathbb v$}}}}$ ] and for some morphisms @xmath2186 satisfying @xmath2187 for all @xmath2188 .
analogously every morphism of @xmath2189 is of the form @xmath2190 for some morphisms @xmath2191 satisfying @xmath2192 for all @xmath2165 and @xmath2166 . in practice
most of the time we will drop all references to the sets of indices , either because they can be clearly deduced from the context or because it is unnecessary to choose name for them , so that objects will often simply be denoted by @xmath2193 and morphisms will often simply be denoted by @xmath2194 every linear functor @xmath2146 admits an extension @xmath2195 , called the _ completion of @xmath2143 _ , mapping every object @xmath2196 to the object @xmath2197 and mapping every morphism @xmath2198 to the morphism @xmath2199 analogously every natural transformation @xmath2200 admits a natural extension @xmath2201 , called the _ completion of @xmath1369 _ , associating with every object @xmath2202 of @xmath2181 the morphism @xmath2203 of @xmath2204 the _ complete unit algebroid @xmath1645 _ is the completion of the unit algebroid @xmath22 .
the _ complete tensor product @xmath2205 of linear categories @xmath1263 and @xmath2144 _ is defined as @xmath2206 the _ complete braiding functor @xmath2207 associated with linear categories @xmath1263 and @xmath2144 _ is the completion of the braiding functor @xmath2156 associated with @xmath1263 and @xmath2144 .
the symmetric monoidal 2-category @xmath27 of complete linear categories is the 2-category whose objects @xmath1263 are complete linear categories , whose 1-morphisms @xmath2146 are linear functors and whose 2-morphisms @xmath2147 are natural transformations , with identities and compositions inherited by @xmath553 and with monoidal structure given by the complete unit algebroid @xmath1645 , by the complete tensor product of linear categories @xmath2208 and by the complete braiding @xmath2209 of linear categories .
the _ completion _
@xmath2210 of a functor @xmath2211 is the functor of 2-categories mapping every object @xmath1129 of @xmath2212 to the completion @xmath2213 of the linear category @xmath2214 , mapping every 1-morphism @xmath205 of @xmath2212 to the completion @xmath2215 of the linear functor @xmath2216 and mapping every 2-morphism @xmath2217 of @xmath2212 to the completion @xmath2218 of the natural transformation @xmath2219 .
two linear categories @xmath1263 and @xmath2144 are _ morita equivalent _ if their completions are equivalent as objects of @xmath27 . a _ morita equivalence between linear categories @xmath1263 and @xmath2144 _ is a linear functor @xmath2146 whose completion @xmath2220 is part of an equivalence adjunction system in @xmath27 .
[ dominating set ] a set @xmath245 of objects of a linear category @xmath1263 is a _ dominating set for @xmath1263 _ if for every object @xmath2221 there exist objects @xmath2222 and morphisms @xmath2223 satisfying @xmath2224 .
if @xmath245 is a dominating set for @xmath1263 we say that _
@xmath245 dominates @xmath1263_. [ cauchy_completion_equivalence ] let @xmath1263 and @xmath2144 be linear categories and let @xmath2146 be a fully faithful linear functor . if @xmath2225 dominates @xmath2144 then @xmath2143 is a morita equivalence .
let @xmath2 be an abelian group , let @xmath223 be a symmetric @xmath10-bilinear map and let @xmath2226 denote the discrete category whose set of objects is given by the set of elements of @xmath2 .
a _ @xmath2-graded vector space @xmath1280 _ is a functor @xmath2227 if @xmath1280 is a @xmath2-graded vector space then for every @xmath242 the vector space @xmath2228 is called the _ space of degree @xmath1135 vectors_. a @xmath2-graded vector space @xmath1280 is _ locally finitely generated _ if @xmath2229 is finite dimensional for every @xmath242 .
the _ dimension of a locally finitely generated @xmath2-graded vector space @xmath1280 _ is the function @xmath2230 a locally finitely generated @xmath2-graded vector space @xmath1280 is _ finitely generated _ if @xmath2231 a _ @xmath2-graded linear map @xmath1424{bbf.pdf } } : { \mathbb v}\rightarrow { \mathbb v}'$ ] _ for @xmath2-graded vector spaces @xmath1280 and @xmath2232 is a natural transformation @xmath1422{bbf.pdf } } : { \mathbb v}\rightarrow { \mathbb v}'.\ ] ] if @xmath1424{bbf.pdf } } : { \mathbb v}\rightarrow { \mathbb v}'$ ] is a @xmath2-graded linear map then for every @xmath242 the vector space linear map @xmath1424{bbf.pdf}}^u : = { \includegraphics[height=\heightof{$\mathbf{f}$}]{bbf.pdf}}_u \in { \mathrm{hom}}_{{\mathbb c}}({\mathbb v}^u,{\mathbb v}'^u)$ ] is called the _ degree @xmath1135 component of @xmath1424{bbf.pdf}}$]_. [ ker_im_graded_linear_function ] if @xmath1424{bbf.pdf } } : { \mathbb v}\rightarrow { \mathbb v}'$ ] is a @xmath2-graded linear map then we can define @xmath2-graded vector spaces @xmath2233{bbf.pdf}}$ ] and @xmath2234{bbf.pdf}}$ ] as @xmath2235{bbf.pdf}})^u : = { \mathrm{ker}}({\includegraphics[height=\heightof{$\mathbf{f}$}]{bbf.pdf}}^u ) , \quad ( { \mathrm{im}}{\includegraphics[height=\heightof{$\mathbf{f}$}]{bbf.pdf}})^u : = { \mathrm{im}}({\includegraphics[height=\heightof{$\mathbf{f}$}]{bbf.pdf}}^u)\ ] ] respectively . the _ unit @xmath2-graded vector space
@xmath22 is the @xmath2-graded vector space defined by @xmath2236 the _ tensor product @xmath2237 of @xmath2-graded vector spaces _
@xmath1280 and @xmath2232 is the @xmath2-graded vector space defined by @xmath2238 the _ @xmath2239-braiding morphism @xmath2240 associated with @xmath2-graded vector spaces @xmath1280 and @xmath2232 _ is the degree 0 linear map @xmath2241 defined by @xmath2242 the monoidal category @xmath2243 of @xmath2-graded vector spaces is the category whose objects are @xmath2-graded vector spaces , whose morphisms are @xmath2-graded linear maps , whose identities are given by identity natural transformations , whose compositions are given by vertical compositions of natural transformations , whose unit is the unit @xmath2-graded vector space and whose tensor product is given by the tensor product of @xmath2-graded vector spaces .
it can be made into a symmetric monoidal category by equipping it with the braiding @xmath2244 defined by the @xmath2239-braiding morphism @xmath2240 for every pair of @xmath2-graded vector spaces @xmath1280 and @xmath2232 .
a _ @xmath2-graded linear category _ , also called _
@xmath2-graded algebroid _ , is a @xmath2243-enriched category @xmath1271{$\mathbb v$}}}}$ ] .
a _ @xmath2-graded linear functor _ is a @xmath2243-enriched functor @xmath2245 between @xmath2-graded linear categories .
a _ @xmath2-graded natural transformation _ is a @xmath2243-enriched natural transformation @xmath1313{bbeta.pdf}}}$ ] between @xmath2-graded linear functors .
if @xmath1271{$\mathbb v$}}}}$ ] is a @xmath2-graded linear category , if @xmath2246 are objects of @xmath1271{$\mathbb v$}}}}$ ] and if @xmath1135 is an element of @xmath2 then the vector space @xmath2247{\scriptsize $ \mathbb v$}}}}}(x , y))^u$ ] will be called the _ vector space of degree @xmath1135 morphism of @xmath1271{$\mathbb v$}}}}$ ] from @xmath1129 to @xmath2149 _ and it will be denoted @xmath1433{\scriptsize $ \mathbb v$}}}}}^u(x , y)$ ] .
the underlying category @xmath1271{$\mathbb v$}}}}_0 $ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is the linear category having @xmath2248{$\mathbb v$}}}})$ ] as set of objects and having @xmath1433{\scriptsize $ \mathbb v$}}}}}^0(x , y)$ ] as vector space of morphisms from @xmath1129 to @xmath2149 for all @xmath2249{$\mathbb v$}}}})$ ] .
the _ unit @xmath2-graded algebroid @xmath22 _ is the @xmath2243-enriched unit category .
the _ tensor product @xmath1271{$\mathbb v$}}}}{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] of @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] _ is the @xmath2243-enriched tensor product of @xmath2-graded linear categories .
[ composition_in_enriched_tensor_product ] if @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] are @xmath2-graded linear categories then @xmath1273{\scriptsize $ \mathbb v$}}}}{\mbox{\tiny$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}((x , x'),(y , y ' ) ) : = { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}}(x , y ) \otimes { \mathrm{hom}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}(x',y')\ ] ] for all objects @xmath2246 of @xmath1271{$\mathbb v$}}}}$ ] and @xmath2250 of @xmath1271{$\mathbb v$}}}}'$ ] . if @xmath2251{\scriptsize $ \mathbb v$}}}}}(x , y)$ ] , @xmath2252{\scriptsize $ \mathbb v$}}}}}(y ,
z)$ ] are morphisms of @xmath1271{$\mathbb v$}}}}$ ] and if @xmath2253{\scriptsize $ \mathbb v$}}}}'}(x',y')$ ] , @xmath2254{\scriptsize $ \mathbb v$}}}}'}(y',z')$ ] are morphisms of @xmath1271{$\mathbb v$}}}}'$ ] then @xmath2255{\scriptsize $ \mathbb v$}}}}{\mbox{\tiny$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}((x , x'),(y , y ' ) ) , \quad g^{v } \otimes g'^{v ' } \in
{ \mathrm{hom}}^{v+v'}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}{\mbox{\tiny$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}((y , y'),(z , z'))\ ] ] are morphisms of @xmath1271{$\mathbb v$}}}}{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] and their composition is defined as @xmath2256 remark we are actually using a non - standard convention , as the standard choice would have been to use @xmath2257 as a multiplying factor instead of @xmath2258 .
of course both choices produce a coherent compositon laws on @xmath1271{$\mathbb v$}}}}{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] .
the _ @xmath2239-braiding functor @xmath2259{\scriptsize $ \mathbb v$}}}},{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}$ ] associated with @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] _ is the @xmath2-graded linear functor @xmath2260{\scriptsize $ \mathbb v$}}}},{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$ } } } } ' } : { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$ } } } } ' \rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$ } } } } ' { \mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}\ ] ] induced by the @xmath2239-braiding of @xmath2-graded vector spaces .
the symmetric monoidal 2-category @xmath2261 of @xmath2-graded linear categories with @xmath2239-braiding is the 2-category whose objects @xmath1271{$\mathbb v$}}}}$ ] are complete @xmath2-graded linear categories , whose 1-morphisms @xmath2262{$\mathbb v$}}}}\rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] are @xmath2-graded linear functors and whose 2-morphisms @xmath1313{bbeta.pdf } } } : { \mathbb f}\rightarrow { \mathbb f}'$ ] are @xmath2-graded natural transformations , with monoidal structure given by the unit @xmath2-graded algebroid @xmath22 , by the tensor product of @xmath2-graded linear categories @xmath2158 and by the @xmath2239-braiding @xmath2263 of @xmath2-graded linear categories .
an object @xmath1129 of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is said to be the _
@xmath1135-suspension of an object @xmath2149 of @xmath1271{$\mathbb v$}}}}$ ] _ for some @xmath242 if there exists a degree @xmath1135 invertible morphism @xmath2264{\scriptsize $ \mathbb v$}}}}}^{u}(x , y)$ ] .
an object @xmath1129 of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is said to be the _ direct sum in @xmath1271{$\mathbb v$}}}}$ ] of a family of objects @xmath2265 of @xmath1271{$\mathbb v$}}}}$ ] for @xmath2266 _ if it is the direct sum of @xmath2265 for @xmath2266 in the underlying category @xmath1271{$\mathbb v$}}}}_0 $ ] of @xmath1271{$\mathbb v$}}}}$ ] .
a degree 0 morphism of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is said to be an _ idempotent of @xmath1271{$\mathbb v$}}}}$ ] _ if it is an idempotent morphism of the underlying category @xmath1271{$\mathbb v$}}}}_0 $ ] .
a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is _ complete _ if it is closed under suspensions and finite direct sums and if all of its idempotents split .
the _ suspension completion @xmath2267{$\mathbb v$}}}})$ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] _ is the @xmath2-graded linear category whose set of objects @xmath2268{$\mathbb v$}}}}))$ ] is given by @xmath2269{$\mathbb v$ } } } } ) , v \in \varpi \right\}\ ] ] and whose degree @xmath1135 morphism vector spaces @xmath2270{\scriptsize $ \mathbb v$}}}})}^u\left ( ( x , v),(y , w ) \right)$ ] are given by @xmath1433{\scriptsize $ \mathbb v$}}}}}^{u + v - w}\left ( x , y \right)$ ] for all @xmath242 and all @xmath2271{$\mathbb v$}}}}))$ ] .
identities and compositions are directly inherited from @xmath1271{$\mathbb v$}}}}$ ] .
if @xmath1129 is an object of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] then we will denote with @xmath1129 the object @xmath2272 of @xmath2267{$\mathbb v$}}}})$ ] too .
the _ additive completion @xmath2273{$\mathbb v$}}}})$ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] _ is the @xmath2-graded linear category whose set of objects @xmath2274{$\mathbb v$}}}}))$ ] is given by @xmath2275{$\mathbb v$ } } } } ) , \ell \in { \mathbb n}\right\}\ ] ] and whose degree @xmath1135 morphism vector spaces @xmath2276{\scriptsize $ \mathbb v$}}}})}^u \left ( \bigoplus_{j=1}^{\ell } x_j , \bigoplus_{i=1}^{m } y_i \right)\ ] ] are given by @xmath2277{\scriptsize $ \mathbb v$}}}}}^u(x_j , y_i ) \right\}\ ] ] for every @xmath242 and for all @xmath2278{$\mathbb v$}}}})$ ] with @xmath2165 and @xmath2166 .
the identity morphism of an object @xmath2279{$\mathbb v$}}}}))$ ] is given by the degree 0 morphism @xmath2280{\scriptsize $ \mathbb v$}}}})}^0 \left ( \bigoplus_{i=1}^{\ell } x_i \right)\ ] ] and the composition of a degree @xmath1135 morphism @xmath2281{\scriptsize $ \mathbb v$}}}})}^u \left ( \bigoplus_{j=1}^{\ell } x_j , \bigoplus_{i=1}^{m } y_i \right)\ ] ] with a degree @xmath1496 morphism @xmath2282{\scriptsize $ \mathbb v$}}}})}^v \left ( \bigoplus_{j=1}^{m } y_j , \bigoplus_{i=1}^{n } z_i \right)\ ] ] is given by the degree @xmath2283 morphism @xmath2284{\scriptsize $ \mathbb v$}}}})}^{u+v } \left ( \bigoplus_{j=1}^{\ell } x_j , \bigoplus_{i=1}^{n } z_i \right)\ ] ] if @xmath1129 is an object of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] then we will denote with @xmath1129 the object @xmath2172 of @xmath2273{$\mathbb v$}}}})$ ] too . the _ idempotent completion @xmath2285{$\mathbb v$}}}})$ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] _ is the @xmath2-graded linear category whose set of objects @xmath2286{$\mathbb v$}}}}))$ ] is given by @xmath2287{$\mathbb v$ } } } } ) , \quad p^0 \in { \mathrm{end}}_{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}}^0(x ) , \quad p^0 \circ p^0 = p^0 \}\ ] ] and whose degree @xmath1135 morphism vector spaces @xmath2288{\scriptsize $ \mathbb v$}}}})}^u((x , p^0),(y , q^0))$ ] are given by @xmath2289{\scriptsize $ \mathbb v$}}}}}^u(x ,
y ) \mid f^u \circ p^0 = f^u = q^0 \circ f^u \}\ ] ] for all @xmath242 and all @xmath2290{$\mathbb v$}}}}))$ ] .
the identity of an object @xmath2291 is given by the idempotent morphism @xmath72 and compositions are directly inherited from compositions in @xmath1271{$\mathbb v$}}}}$ ] .
if @xmath1129 is an object of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] then we will denote with @xmath1129 the object @xmath2292 of @xmath2285{$\mathbb v$}}}})$ ] too .
[ graded_completion_def ] the _ completion @xmath1643{$\mathbb v$}}}}}$ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] _ is defined as the @xmath2-graded linear category @xmath2293 .
[ graded_completion_remark ] every object of the completion @xmath1643{$\mathbb v$}}}}}$ ] of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is of the form @xmath2294 for some natural number @xmath2184 , for some objects @xmath2185 of @xmath1271{$\mathbb v$}}}}$ ] , for some degrees @xmath2295 and for some morphisms @xmath2296{\scriptsize $ \mathbb v$}}}}}(x_j , x_i)$ ] satisfying @xmath2297 for all @xmath2188 .
analogously every degree @xmath1135 morphism of @xmath2298{\scriptsize $ \mathbb v$ } } } } } } \left ( { \mathrm{im}}\left ( p^{v_j - v_i}_{ij } \right)_{1 { \leqslant}i , j { \leqslant}\ell } , { \mathrm{im}}\left ( q^{w_j - w_i}_{ij } \right)_{1 { \leqslant}i , j { \leqslant}m } \right)\ ] ] is of the form @xmath2299 for some morphisms @xmath2300{\scriptsize $ \mathbb v$}}}}}(x_j , y_i)$ ] satisfying @xmath2301 for all @xmath2165 and @xmath2166 . in practice
most of the time we will drop all references to the sets of indices , either because they can be clearly deduced from the context or because it is unnecessary to choose name for them , so that objects will often simply be denoted by @xmath2302{$\mathbb v$}}}}})\ ] ] and degree @xmath1135 morphisms will often simply be denoted by @xmath2303{\scriptsize $ \mathbb v$ } } } } } } \left ( { \mathrm{im}}\left ( p^{v_j - v_i}_{ij } \right ) , { \mathrm{im}}\left ( q^{w_j - w_i}_{ij } \right ) \right).\ ] ] every @xmath2-graded linear functor @xmath2262{$\mathbb v$}}}}\rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] admits an extension @xmath2304{$\mathbb v$ } } } } } \rightarrow \hat{{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}}'$ ] , called the _ completion of @xmath2245 _ , mapping every object @xmath2305{$\mathbb v$}}}}})$ ] to the object @xmath2306{$\mathbb v$}}}}}')\ ] ] and mapping every degree @xmath1135 morphism @xmath2307{\scriptsize $ \mathbb v$}}}}}}({\mathrm{im } } ( p^{v_j - v_i}_{ij } ) , { \mathrm{im } } ( q^{w_j - w_i}_{ij } ) ) $ ] to the degree @xmath1135 morphism @xmath2308{\scriptsize $ \mathbb v$ } } } } } ' } \left ( { \mathrm{im}}\left ( { \mathbb f}^{v_j - v_i } \left ( p^{v_j - v_i}_{ij } \right ) \right ) , { \mathrm{im}}\left ( { \mathbb f}^{w_j - w_i } \left ( q^{w_j - w_i}_{ij } \right ) \right ) \right)\ ] ] analogously every @xmath2-graded natural transformation @xmath1313{bbeta.pdf } } } : { \mathbb f}\rightarrow { \mathbb g}$ ] admits a natural extension @xmath1678{bbeta.pdf } } } } : \hat{{\mathbb f } } \rightarrow \hat{{\mathbb g}}$ ] , called the _ completion of @xmath1313{bbeta.pdf}}}$ ] _ , associating with every object @xmath2309 of @xmath1643{$\mathbb v$}}}}}$ ] the degree 0 morphism @xmath2310{bbeta.pdf}}}^0_{x_k } \circ { \mathbb f}^{v_j - v_k}(p^{v_j - v_k}_{kj } ) \right)_{1 { \leqslant}i , j { \leqslant}\ell}\ ] ] of @xmath2311{\scriptsize $ \mathbb v$ } } } } } ' } \left ( { \mathrm{im}}\left ( { \mathbb f}^{v_j - v_i } \left ( p^{v_j - v_i}_{ij } \right ) \right)_{1 { \leqslant}i , j { \leqslant}\ell } , { \mathrm{im}}\left ( { \mathbb g}^{v_j - v_i } \left ( p^{v_j - v_i}_{ij } \right ) \right)_{1 { \leqslant}i , j { \leqslant}\ell } \right).\ ] ] the _ complete unit @xmath2-graded algebroid @xmath1645 _ is the completion of the unit @xmath2-graded algebroid @xmath22 . the _ complete tensor product @xmath1271{$\mathbb v$}}}}\hat{{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize } } } { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] of @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] _ is defined as @xmath2312{$\mathbb v$}}}}{\mbox{\footnotesize$\ ; \boxtimes \;$\normalsize}}{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'}.\ ] ] the _ complete @xmath2239-braiding functor @xmath2313{\scriptsize $ \mathbb v$}}}},{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}$ ] associated with @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] _ is the completion of the @xmath2239-braiding functor @xmath2259{\scriptsize $ \mathbb v$}}}},{\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{\scriptsize $ \mathbb v$}}}}'}$ ] associated with @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] .
the symmetric monoidal 2-category @xmath18 of complete @xmath2-graded linear categories with @xmath2239-braiding is the 2-category whose objects @xmath1271{$\mathbb v$}}}}$ ] are complete @xmath2-graded linear categories , whose 1-morphisms @xmath2262{$\mathbb v$}}}}\rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] are @xmath2-graded linear functors and whose 2-morphisms @xmath1313{bbeta.pdf } } } : { \mathbb f}\rightarrow { \mathbb f}'$ ] are @xmath2-graded natural transformations , with monoidal structure given by the complete unit @xmath2-graded algebroid @xmath1645 , by the complete tensor product of @xmath2-graded linear categories @xmath2208 and by the complete @xmath2239-braiding @xmath2314 of @xmath2-graded linear categories .
the _ completion _
@xmath2315 of a functor @xmath2316 is the functor of 2-categories mapping every object @xmath1129 of @xmath2212 to the completion @xmath2213 of the @xmath2-graded linear category @xmath2214 , mapping every 1-morphism @xmath205 of @xmath2212 to the completion @xmath2215 of the @xmath2-graded linear functor @xmath2216 and mapping every 2-morphism @xmath2217 of @xmath2212 to the completion @xmath2218 of the @xmath2-graded natural transformation @xmath2219 .
two @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] are _
@xmath2-morita equivalent _ if their completions are equivalent as objects of @xmath18 .
@xmath2-morita equivalence between @xmath2-graded linear categories @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] _ is a @xmath2-graded linear functor @xmath2262{$\mathbb v$}}}}\rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] whose completion @xmath2317 is part of an equivalence adjunction system in @xmath18 .
a set @xmath245 of objects of a @xmath2-graded linear category @xmath1271{$\mathbb v$}}}}$ ] is a _ dominating set for @xmath1271{$\mathbb v$}}}}$ ] _ if for every object @xmath2318{$\mathbb v$}}}})$ ] there exist objects @xmath2222 , elements @xmath952 and morphisms @xmath2319 satisfying @xmath2320 .
if @xmath245 is a dominating set for @xmath1271{$\mathbb v$}}}}$ ] we say that _
@xmath245 dominates @xmath1271{$\mathbb v$}}}}$]_. [ graded_cauchy_completion_equivalence ] let @xmath1271{$\mathbb v$}}}}$ ] and @xmath1271{$\mathbb v$}}}}'$ ] be @xmath2-graded linear categories and let @xmath2262{$\mathbb v$}}}}\rightarrow { \reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathbb v$}}}}'$ ] be a fully faithful @xmath2-graded linear functor . if @xmath2321{$\mathbb v$}}}}))$ ] dominates @xmath1271{$\mathbb v$}}}}'$ ] then @xmath2245 is a @xmath2-morita equivalence .
let @xmath91 be a set , let @xmath4 be a category and let @xmath2212 be a 2-category . 1
. an _ invariant on @xmath91 _ is a function from @xmath91 to @xmath22 .
2 . a _ covariant quantization functor on @xmath4 _ is a functor from @xmath4 to @xmath551 .
3 . a _ contravariant quantization functor on @xmath4 _ is a functor from @xmath32 to @xmath551 .
4 . a _ covariant extended quantization functor on @xmath2212 _ is a functor from @xmath2212 to @xmath553 .
a _ contravariant extended quantization functor on @xmath2212 _ is a functor from @xmath2322 to @xmath553 .
a _ pointed category _
is the given of a category @xmath4 together with a specified object called the _ base point_. a _ pointed 2-category _ is the given of a 2-category @xmath2212 together with a specified object called the _
base point_. by abuse of notation we will often denote pointed categories and pointed 2-categories without specifying base points , especially when we have natural choices for base points , as it is the case with monoidal categories and monoidal 2-categories , for which the base point will always be tacitly taken to be the unit for the tensor product . let us start by recalling the universal construction , which was introduced in @xcite . in order to do so let us fix a pointed category @xmath4 with base point @xmath2323 , let us consider the set @xmath2324 and an invariant @xmath1813 on @xmath91 .
let @xmath1129 be an object of @xmath4 , let @xmath2325 be the free complex vector space generated by the set @xmath2326 and let @xmath2327 be the free complex vector space generated by the set @xmath2328 .
we define the linear map @xmath2329 which maps every vector @xmath2330 in the basis of @xmath2325 to the linear map @xmath2331 mapping every vector @xmath2332 in the basis of @xmath2327 to the evaluation @xmath2333 .
analogously we define the linear map @xmath2334 which maps every vector @xmath2332 in the basis of @xmath2327 to the linear map @xmath2335 mapping every vector @xmath2330 in the basis of @xmath2325 to the evaluation @xmath2333 .
[ uc_vector_space ] the _ covariant universal vector space of @xmath1129 with respect to @xmath1813 _ is the complex vector space @xmath2336 the _ contravariant universal vector space of @xmath1129 with respect to @xmath1813 _ is the complex vector space @xmath2337 [ uc_linear_map ] the _ covariant universal linear map of @xmath205 with respect to @xmath1813 _ is the linear map @xmath2339 & \mapsto & [ f \circ f_x : x_0 \rightarrow x ' ] \end{array}\ ] ] the _ contravariant universal linear map of @xmath205 with respect to @xmath1813 _ is the linear map @xmath2340 & \mapsto & [ f'_{x ' } \circ f : x \rightarrow x_0 ] \end{array}\ ] ] 1 . a covariant quantization functor @xmath2341 mapping each object @xmath1129 of @xmath4 to @xmath2342 and each morphism @xmath205 of @xmath4 to @xmath2343 ; 2 .
a contravariant quantization functor @xmath2344 mapping each object @xmath1129 of @xmath32 to @xmath2345 and each morphism @xmath205 of @xmath32 to @xmath2346 .
let us now describe the extended universal construction , which builds on the universal construction we just introduced . in order to do so
let us fix a pointed 2-category @xmath2212 with base point @xmath2348 , let us consider the pointed category @xmath2349 with base point @xmath2350 , the set @xmath2351 and an invariant @xmath1813 on @xmath91 .
then the universal construction extends @xmath1813 to a covariant quantization functor @xmath2341 and to a contravariant quantization functor @xmath2344 .
let @xmath1129 be an object of @xmath2212 , let @xmath1271{$\mathcal v$}}}}(x)$ ] be the free linear category generated by @xmath2352 and let @xmath1271{$\mathcal v$}}}}'(x)$ ] be the free linear category generated by @xmath2353 .
we define the linear functor @xmath2354{$\mathcal v$}}}}(x ) \rightarrow { \mathbf{cat}}_{{\mathbb c}}({\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathcal v$}}}}'(x),{\mathrm{vect}}_{{\mathbb c}})\ ] ] which maps every object @xmath2330 of @xmath1271{$\mathcal v$}}}}(x)$ ] to the linear functor @xmath2355{$\mathcal v$}}}}'(x ) \rightarrow { \mathrm{vect}}_{{\mathbb c}}\ ] ] mapping every object @xmath2356 of @xmath1271{$\mathcal v$}}}}'(x)$ ] to the universal vector space @xmath2357 and every vector @xmath2358 in the basis of @xmath1433{\scriptsize $ \mathcal v$}}}}'(x)}(f'_{x},f'''_{x})$ ] to the universal linear map @xmath2359 and which maps every vector @xmath2360 in the basis of @xmath1433{\scriptsize $ \mathcal v$}}}}(x)}(f_{x},f''_{x})$ ] to the natural transformation @xmath2361 associating with every object @xmath2356 of @xmath1271{$\mathcal v$}}}}'(x)$ ] the linear map @xmath2362 analogously we define the linear functor @xmath2363{$\mathcal v$}}}}'(x ) \rightarrow { \mathbf{cat}}_{{\mathbb c}}({\reflectbox{\raisebox{\depth}{\scalebox{1}[-1]{$\mathcal v$}}}}(x),{\mathrm{vect}}_{{\mathbb c}})\ ] ] which maps every object @xmath2356 of @xmath1271{$\mathcal v$}}}}'(x)$ ] to the linear functor @xmath2364{$\mathcal v$}}}}(x ) \rightarrow { \mathrm{vect}}_{{\mathbb c}}\ ] ] mapping every object @xmath2330 of @xmath1271{$\mathcal v$}}}}(x)$ ] to the universal vector space @xmath2357 and every vector @xmath2360 in the basis of @xmath1433{\scriptsize $ \mathcal v$}}}}(x)}(f_{x},f''_{x})$ ] to the universal linear map @xmath2365 and which maps every vector @xmath2358 in the basis of @xmath1433{\scriptsize $ \mathcal v$}}}}'(x)}(f'_{x},f'''_{x})$ ] to the natural transformation @xmath2366 associating with every object @xmath2330 of @xmath1271{$\mathcal v$}}}}(x)$ ] the linear map @xmath2367 [ euc_algebroid ] the _ covariant universal algebroid of @xmath1129 with respect to @xmath1813 _ is the linear category @xmath2368{$\mathcal v$}}}}(x)/{\mathrm{ker}}\psi_{x}.\ ] ] the _ contravariant universal algebroid of @xmath1129 with respect to @xmath1813 _ is the linear category @xmath2369{$\mathcal v$}}}}'(x)/{\mathrm{ker}}\psi'_{x}.\ ] ] [ euc_functor ] the _ covariant universal linear functor of @xmath205 with respect to @xmath1813 _ is the linear functor @xmath2370 mapping every object @xmath2330 of @xmath2371 to the object @xmath2372 of @xmath2373 and mapping every morphism @xmath2374 $ ] of @xmath2371 to the morphism @xmath2375 $ ] of @xmath2373 .
the _ contravariant universal linear functor of @xmath205 with respect to @xmath1813 _ is the linear functor @xmath2376 mapping every object @xmath2377 of @xmath2378 to the object @xmath2379 of @xmath2380 and mapping every morphism @xmath2381 $ ] of @xmath2378 to the morphism @xmath2382 $ ] of @xmath2380 .
[ euc_natural_transformation ] the _ covariant universal natural transformation of @xmath2217 with respect to @xmath1813 _ is the natural transformation @xmath2385 which associates with every object @xmath2386 of @xmath2371 the morphism @xmath2387\ ] ] of @xmath2373 .
the _ contravariant universal natural transformation of @xmath2217 with respect to @xmath1813 _ is the natural transformation @xmath2388 which associates with every object @xmath2377 of @xmath2378 the morphism @xmath2389\ ] ] of @xmath2380 . 1
. a covariant extended quantization functor @xmath2390 mapping each object @xmath1129 of @xmath2212 to @xmath2371 , each 1-morphism @xmath205 of @xmath2212 to @xmath2391 and each 2-morphism @xmath2217 of @xmath2212 to @xmath2392 ; 2 . a contravariant extended quantization functor @xmath2393 mapping each object @xmath1129 of @xmath2322 to @xmath2380 , each 1-morphism @xmath205 of @xmath2322 to @xmath2394 and each 2-morphism @xmath2217 of @xmath2322 to @xmath2395 . for every object @xmath1129 of @xmath2212 we will denote with @xmath2396 the completion of the universal algebroid @xmath2371 , for every 1-morphism @xmath205 of @xmath2212 we will denote with @xmath2397 the completion of the universal linear functor @xmath2391 and for every 2-morphism @xmath2217 of @xmath2212 we will denote with @xmath2398 the completion of the universal linear functor @xmath2392 .
99 b. bartlett , c. douglas , c. schommer - pries , j. vicary - _ modular categories as representations of the 3-dimensional bordism 2-category _ - https://arxiv.org/abs/1509.06811[arxiv:1509.06811 ] [ math.at ] , 2015 c. blanchet , f. costantino , n. geer , b. patureau - mirand - _ non - semisimple tqfts , reidemeister torsion and kashaev s invariants _ - advances in mathematics , volume 301 , 1 october 2016 , pages 178 c. blanchet , n. habegger , g. masbaum , p.
vogel - _ topological quantum field theories derived from the kauffman bracket _ - topology , volume 34 , issue 4 , 1995 , pages 883927 f. costantino , n. geer , b. patureau - mirand - _ quantum invariants of 3-manifolds via link surgery presentations and non - semi - simple categories _ - journal of topology , volume 7 , number 4 , 2014 , pages 10051053 m. de renzi - _ quantum invariants of 3-manifolds arising from non - semisimple categories _ - bulletin of the chelyabinsk state university , number 3 , mathematics , mechanics , informatics , volume 17 , 2015 , pages 2640 - https://arxiv.org/abs/1703.07319[arxiv:1703.07319 ] [ math.gt ] n. geer , j. kujawa , b. patureau - mirand - _ generalized trace and modified dimension functions on ribbon categories _ - selecta mathematica , volume 17 , number 2 , 2011 , pages 435504 n. geer , j. kujawa , b. patureau - mirand - _ ambidextrous objects and trace functions for nonsemisimple categories _ - proceedings of the american mathematical society , volume 141 , 2013 , 29632978 n. geer , b. patureau - mirand - _ topological invariants from non - restricted quantum groups _ - algebraic & geometric topology , volume 13 , issue 6 , 2013 , 33053363 n. geer , b. patureau - mirand , v. turaev - _ modified quantum dimensions and re - normalized link invariants _ - compositio mathematica , volume 145 , issue 1 , january 2009 , pages 196212 n. geer , b. patureau - mirand , a. virelizier - _ traces on ideals in pivotal categories _ - quantum topology , volume 4 , issue 1 , 2013 , pages 91124 n. reshetikhin , v. turaev - _ invariants of 3-manifolds via link polynomials and quantum groups _ - inventiones mathematicae , volume 103 , issue 1 , 1991 , pages 547597 c. schommer - pries - _ the classification of two - dimensional extended topological field theories
_ - http://arxiv.org/abs/1112.1000v2[arxiv:1112.1000v2 ] [ math.at ] , 2014 v. turaev - _ quantum invariants of knots and 3-manifolds
_ - berlin , boston : de gruyter , 2016 v. turaev , a. virelizier - _ on 3-dimensional homotopy quantum field theory i _ - international journal of mathematics , volume 23 , number 9 , 2012 , pages 1250094 | in this paper we provide the general theory for the construction of 3-dimensional etqfts extending the costantino - geer - patureau quantum invariants defined in @xcite .
our results rely on relative modular categories , a class of non - semisimple ribbon categories modeled on representations of unrolled quantum groups , and they exploit a 2-categorical version of the universal construction introduced by blanchet , habegger , masbaum and vogel .
the 1 + 1 + 1 tqfts thus obtained are realized by symmetric monoidal functors defined over 2-categories of admissible cobordisms decorated with colored ribbon graphs and cohomology classes and taking values in 2-categories of complete graded linear categories . in particular
our construction extends the family of graded tqfts defined for unrolled quantum @xmath0 by blanchet , costantino , geer and patureau in @xcite to a new family of graded etqfts .
the non - semisimplicity of the theory is witnessed by the presence of non - semisimple graded linear categories associated with critical 1-dimensional manifolds .
= 1 |
decays of pseudoscalar mesons to light leptons are suppressed in the standard model ( sm ) by angular momentum conservation . in particular , the sm width of @xmath4 decays ( with @xmath5 ) is @xmath6 where @xmath7 is the fermi constant , @xmath8 and @xmath9 are meson and lepton masses , @xmath10 is the decay constant , and @xmath11 is the corresponding ckm matrix element . within the two higgs doublet models ( 2hdm of type ii ) , including the minimal supersymmetric one , the charged higgs boson ( @xmath12 ) exchange induces a tree - level contribution to ( semi)leptonic decays proportional to the yukawa couplings of quarks and leptons @xcite . in @xmath13 , it can compete with the @xmath14 exchange due to the helicity suppression of the latter . at tree level ,
the @xmath12 exchange contribution is lepton flavour independent , and for @xmath15 leads to @xcite @xmath16 ^ 2.\ ] ] here @xmath17 is the @xmath12 mass , @xmath18 is the ratio of the two higgs vacuum expectation values , and @xmath19 is an effective coupling . a plausible choice of parameters @xmath20 , @xmath21 leads to @xmath22 relative suppression of @xmath23 decays , and @xmath24 suppression of @xmath25 decays with respect to their sm rates .
however , searches for new physics in the decay rates are hindered by the uncertainties of their sm predictions .
in particular , interpretation of the measurements of the ratio @xmath26 in terms of constraints on ( @xmath17 , @xmath18 ) phase space is currently limited by lattice qcd uncertainties @xcite . on the other hand ,
the ratio of kaon leptonic decay widths @xmath27 , where the notation @xmath28 is adopted for @xmath29 decays , is sensitive to loop - induced lepton flavour violating ( lfv ) effects via the @xmath12 exchange @xcite : @xmath30 where the mixing parameter between the superpartners of the right - handed leptons @xmath31 can reach @xmath32 .
this can enhance @xmath0 by @xmath33 relative without contradicting any presently known experimental constraints , including upper bounds on the lfv decays @xmath34 with @xmath35 . unlike the individual @xmath36
decay widths , the ratio @xmath27 is precisely predicted within the sm due to cancellation of hadronic uncertainties @xcite : @xmath37 where @xmath38 is an electromagnetic correction due to the internal bremsstrahlung ( ib ) process .
the sensitivity to lfv and the precision of the sm prediction make @xmath0 an excellent probe of lepton universality .
the current world average ( based on final results only ) @xmath39 is dominated by a recent kloe result @xcite .
a precise measurement of @xmath0 based on a part ( 40% ) of the data sample collected by the cern na62 experiment in 2007 is reported here .
this is an update of an earlier result obtained with the same data sample @xcite .
the na48/2 experimental setup @xcite has been used for the na62 200708 data taking .
experimental conditions have been optimized for the @xmath40 measurement .
the beam line is designed to deliver simultaneous unseparated @xmath41 and @xmath42 beams derived from the sps 400 gev/@xmath43 primary protons .
however , the muon sweeping system was optimized for the positive beam in 2007 , and the sample used for the present analysis was collected with the @xmath41 beam only .
positively charged particles within a narrow momentum band of @xmath44 gev/@xmath43 are selected by an achromatic system of four dipole magnets with zero total deflection , pass through a muon sweeping system , and enter a fiducial decay volume contained in a 114 m long cylindrical vacuum tank . with about @xmath45
primary protons incident on the target per sps pulse of about @xmath46 s duration , the secondary beam flux at the entrance to the decay volume is @xmath47 particles per pulse , of which 5% are kaons ( @xmath41 ) .
the fraction of beam kaons decaying in the vacuum tank at nominal momentum is @xmath48 .
the transverse size of the beam within the decay volume is @xmath49 mm ( rms ) , and its angular divergence is negligible . among the subdetectors located downstream the decay volume , a magnetic spectrometer , a plastic scintillator hodoscope ( hod ) and a liquid krypton electromagnetic calorimeter ( lkr ) are principal for the measurement .
the spectrometer , used to detect charged products of kaon decays , is composed of four drift chambers ( dchs ) and a dipole magnet .
the hod producing fast trigger signals consists of two planes of strip - shaped counters .
the lkr , used for particle identification and as a veto , is an almost homogeneous ionization chamber , @xmath50 deep , segmented transversally into 13,248 cells ( @xmath51 each ) , and with no longitudinal segmentation .
a beam pipe traversing the centres of the detectors allows undecayed beam particles and muons from decays of beam pions to continue their path in vacuum .
a minimum bias trigger configuration has been employed .
the @xmath52 trigger condition consists of coincidence of hits in the two hod planes ( the @xmath53 signal ) , loose lower and upper limits on dch hit multiplicity ( the 1-track signal ) , and lkr energy deposit @xmath54 of at least 10 gev .
the @xmath55 trigger condition requires a coincidence of the @xmath56 and 1-track signals downscaled by a factor @xmath57 .
the analysis strategy is based on counting the numbers of reconstructed @xmath52 and @xmath58 candidates collected concurrently .
thus the analysis does not rely on the absolute beam flux measurement , and several systematic effects ( e.g. due to charged track reconstruction and @xmath56 trigger efficiencies , time - dependent effects ) cancel at first order . due to the significant acceptance and background dependence ,
the measurement is performed independently in 10 bins of lepton momentum covering a range from 13 to 65 gev/@xmath43 .
the first bin spans 7 gev/@xmath43 , while the others are 5 gev/@xmath43 wide .
the selection conditions have been optimized separately in each momentum bin .
the ratio @xmath0 in each bin is computed as @xmath59 where @xmath60 are the numbers of selected @xmath36 candidates @xmath61 , @xmath62 are the numbers of background events , @xmath63 is the geometric acceptance correction , @xmath64 are the lepton identification efficiencies , @xmath65 are the trigger efficiencies , @xmath66 is the global efficiency of the lkr readout , and @xmath57 is the @xmath58 trigger downscaling factor . to evaluate the acceptance correction and the geometric parts of the acceptances for background processes , a detailed monte carlo ( mc ) simulation including beam line optics , full detector geometry and material description , magnetic fields , local inefficiencies of dch wires , non - working lkr cells ( 0.8% of channels ) and temporarily masked lkr cells is used .
particle identification , trigger and readout efficiencies are measured directly from data .
charged particle tracks are reconstructed from hits and drift times in the spectrometer .
track momenta are evaluated using a detailed magnetic field map .
clusters of energy deposition in the lkr are found by looking at the maxima in the digitized pulses from individual cells in both space and time , and accumulating the energy within a radius of 11 cm .
shower energies are corrected for energy outside the cluster boundary , energy lost in non - working cells , and cluster energy sharing .
due to the topological similarity of @xmath52 and @xmath58 decays , a large part of the selection is common for the two modes : ( 1 ) exactly one reconstructed particle of positive electric charge geometrically consistent with originating from a kaon decay is required ; ( 2 ) extrapolated track impact points in the dchs , hod and lkr must be within their geometrical acceptances ; ( 3 ) track momentum must be in the range ( 13 ; 65 ) gev/@xmath43 , where the lower limit assures the efficiency of the @xmath67 gev trigger condition ; ( 4 ) no lkr energy deposition clusters with energy @xmath68 gev and in time with the track are allowed unless they are consistent with being produced by the track via direct energy deposition or bremsstrahlung ; ( 5 ) the reconstructed decay vertex longitudinal position must be within the nominal decay volume ; ( 6 ) distance between the charged track and the nominal kaon beam axis must be below 3.5 cm .
the following two principal criteria are used to distinguish @xmath52 from @xmath58 decays .
kinematic identification is based on constraining the reconstructed squared missing mass in positron ( muon ) hypothesis @xmath69 , where @xmath70 and @xmath71 are the four - momenta of the kaon ( defined as the average one monitored with @xmath72 decays ) and the lepton ( under the @xmath73 or @xmath74 mass hypothesis ) . the limits @xmath75 and @xmath76 have been optimized taking into account the resolution and backgrounds , and vary among lepton momentum bins in the ranges ( 0.013 ; 0.016 ) and ( 0.010 ; 0.014 ) ( gev/@xmath77)@xmath78 , respectively .
lepton identification is based on the ratio @xmath79 of track energy deposition in the lkr to its momentum measured by the spectrometer .
tracks with @xmath80 , where @xmath81 for @xmath82 gev/@xmath43 and @xmath83 otherwise , are identified as positrons .
tracks with @xmath84 are identified as muons .
( -22,170)(a ) ( -22,82)(b ) kinematic separation of @xmath52 from @xmath58 decays is achievable at low lepton momentum only ( @xmath85 gev/@xmath43 ) , as shown in fig .
1a . at high lepton momentum , the @xmath55 decay with the muon mis - identified as positron ( @xmath86 , as shown in fig .
1b ) due to ` catastrophic ' bremsstrahlung in or in front of the lkr is the largest background source . in order to measure the mis - identification probability @xmath87 , a muon sample free from the typical @xmath88 positron contamination due to @xmath89
decays has been collected : a @xmath90 thick lead ( pb ) wall covering @xmath91 of the geometric acceptance was installed in front of the lkr during a period of data taking .
the positron component in a sample of muon candidates consistent with originating from @xmath55 decays , traversing the pb wall , with @xmath92 gev/@xmath43 and @xmath86 is suppressed to a negligible level ( @xmath93 ) by energy losses in the pb wall
. however , the muon passage through the pb wall affects the measured @xmath94 via two principal effects : 1 ) ionization energy loss in pb decreases @xmath87 and dominates at low momentum ; 2 ) bremsstrahlung in pb increases @xmath87 and dominates at high momentum . to evaluate the corresponding correction factor @xmath95 , a dedicated geant4 based @xcite mc simulation of muon propagation downstream the spectrometer involving all electromagnetic processes , including muon bremsstrahlung @xcite , has been developed .
the relative systematic uncertainties on @xmath87 and @xmath96 obtained by simulation are estimated to be @xmath97 , mainly due to the simulation of cluster geometry and calibration ( measured and simulated @xmath98 are shown in fig . 2 ) . however , the error affecting their ratio is significantly smaller ( @xmath99 ) due to partial cancellation of uncertainties .
[ fig : pmue ] the @xmath58 background contamination has been computed to be @xmath100 using the measured @xmath94 corrected by the simulated @xmath101 , and correcting for the correlation between the reconstructed @xmath102 and @xmath79 .
the uncertainty comes from the limited size of the data sample used to measure @xmath94 ( 0.16% ) , the uncertainty @xmath103 ( 0.12% ) , and model - dependence of the @xmath102 vs @xmath79 correlation ( 0.08% ) .
the @xmath55 decay also contributes to background via the @xmath104 decay in flight .
energetic forward daughter positrons compatible with @xmath52 kinematics and topology are suppressed by muon polarization effects @xcite .
radiative corrections to the muon decay @xcite lead to a further @xmath105 relative background suppression .
the background contamination has been estimated to be @xmath106 .
the structure - dependent ( sd ) @xmath107 process @xcite , not suppressed by angular momentum conservation ( more specifically , its @xmath108 component corresponding to positive photon helicity ) , represents a significant background source .
a recent measurement of the @xmath109 differential decay rate @xcite has been used to evaluate the background contamination to be @xmath110 .
the dominant uncertainty comes from the uncertainty on the rate of @xmath111 decay , which has been increased by a factor of 3 with respect to that reported in @xcite , as suggested by a stability check of @xmath0 with respect to a variation of the @xmath112 limit .
the beam halo background is induced by halo muons undergoing @xmath104 decays in the vacuum tank , or being mis - identified as positrons .
it has been measured directly by reconstructing the @xmath113 candidates from a @xmath42 data sample collected with the @xmath41 beam ( but not its halo ) blocked , and a special data sample collected with both beams blocked .
the control sample is normalized to the data in the region @xmath114 @xmath115 populated predominantly by beam halo events .
the ` cross - talk ' probability to reconstruct a @xmath116 candidate due to a @xmath42 decay with @xmath73 emission ( @xmath117 , @xmath118 , @xmath119 , where @xmath120 denotes the @xmath121 dalitz decay @xmath122 ) is at the level of @xmath123 and is taken into account .
the halo background contamination has been estimated to be @xmath124 , where the uncertainty comes from the limited size of the control sample and the uncertainty of its normalization .
the beam halo is the only significant background source in the @xmath55 sample , measured to be @xmath125 with the same technique .
the number of @xmath36 candidates is @xmath126 ( about four times the statistics collected by kloe ) and @xmath127 .
the @xmath128 distributions of data events and backgrounds are presented in fig .
3 ; backgrounds in the @xmath52 sample integrated over lepton momentum are summarized in table 1 .
[ fig : mm2e ] lc source & @xmath129 + @xmath55 & @xmath100 + @xmath130 & @xmath106 + @xmath131 & @xmath110 + beam halo & @xmath124 + @xmath132 & @xmath133 + @xmath134 & @xmath133 + total background & @xmath135 +
the ratio of geometric acceptances @xmath136 in each lepton momentum bin has been evaluated with a mc simulation .
the radiative @xmath107 ( ib ) process is simulated following @xcite with higher order corrections according to @xcite .
lepton tracking inefficiency due to interactions with the spectrometer material is included into the acceptance correction , and its simulation has been validated with the data .
the main sources of systematic uncertainty of the acceptance correction are the limited knowledge of beam profile and divergence , accidental activity , and the simulation of soft radiative photons .
a separate uncertainty has been assigned due to the limited precision of the dch alignment .
a sample of @xmath137 positrons selected kinematically from @xmath138 decays collected concurrently with the main @xmath36 data set is used to calibrate the energy response of each lkr cell , and to study @xmath139 with respect to local position and time stability ( in the kinematically limited momentum range @xmath140 ) .
a sample of electrons and positrons from the @xmath141 @xmath142 decays collected during a special short ( 15h ) run with a broad momentum band @xmath143 beam allows the determination of @xmath139 in the whole analysis momentum range .
the measurements of @xmath139 have been performed in bins of lepton momentum ; separate measurements have been performed for several identified groups of lkr cells with higher local inefficiencies .
the inefficiency averaged over the @xmath52 sample is @xmath144 , where the uncertainty takes into account the statistical precision and the small differences between @xmath41 and @xmath143 measurements .
the efficiency of the @xmath56 trigger condition has been measured using @xmath55 events triggered with a special control lkr signal : integrated over the @xmath55 sample , it is @xmath145 .
owing to its geometric uniformity , and the similarity of the @xmath52 and @xmath55 distributions over the hod plane , it mostly cancels between the @xmath52 and @xmath55 samples , and the residual systematic bias on @xmath0 is negligible .
the inefficiency of the 1-track trigger for @xmath36 modes is negligible .
the trigger efficiency correction @xmath146 is determined by the efficiency @xmath147 of the lkr energy deposit trigger signal @xmath67 gev , which has been measured to be @xmath148 in the first lepton momentum bin of ( 13 ; 20 ) gev/@xmath43 , and to be negligible in the other momentum bins .
the corresponding uncertainty on @xmath0 is negligible .
.summary of the uncertainties on @xmath0 . [ cols="<,^",options="header " , ] energetic photons not reconstructed in the lkr may initiate showers by interacting in the dch or beam pipe material , which causes the dch hit multiplicities to exceed the limits allowed by the 1-track trigger condition .
this suppresses the @xmath109 background by about 10% relative ( varying over the positron momentum ) .
evaluation of the 1-track inefficiency for @xmath149 partially relies on simulation ; its uncertainty has been propagated into @xmath0 .
the global lkr readout inefficiency has been measured using an independent readout system to be @xmath150 and stable in time .
the independent measurements of @xmath0 in the 10 lepton momentum bins and the average over the bins are displayed in fig .
4 . extensive stability checks in bins of kinematic variables , against variation of selection criteria and analysis procedures have been performed .
the uncertainties of the combined result are summarized in table 2 .
the result is @xmath151 this is the most precise measurement to date ; it is consistent with the sm expectation
. 9 w .- s .
hou , phys .
d48 ( 1993 ) 2342 . g. isidori and p. paradisi , phys .
b639 ( 2006 ) 499 .
m. antonelli _
et al . _ , arxiv:1005.2323 .
a. masiero , p. paradisi and r. petronzio , phys .
d74 ( 2006 ) 011701 . v. cirigliano and i. rosell , phys .
99 ( 2007 ) 231801 .
f. ambrosino _
et al . _ ,
j. c64 ( 2009 ) 627 .
erratum - ibid .
c65 ( 2010 ) 703 .
e. goudzovski , arxiv:1005.1192 .
et al . _ , nucl .
instrum . methods a574 ( 2007 ) 433 .
s. agostinelli _
et al . _ ,
. methods a506 ( 2003 ) 250 .
kelner , r.p .
kokoulin and a.a .
petrukhin , phys .
. nucl . 60
( 1997 ) 576 .
l. michel , proc .
a63 ( 1950 ) 514 .
a. arbuzov , a. czarnecki and a. gaponenko , phys .
d65 ( 2002 ) 113006 .
j. bijnens , g. ecker and j. gasser , nucl .
b396 ( 1993 ) 81 . s. weinberg , phys . rev
. 140 ( 1965 ) b516 . c. gatti , eur .
j. c45 ( 2006 ) 417 . | a precision test of lepton universality by measurement of the ratio @xmath0 of @xmath1 to @xmath2 decay rates was performed using a sample of 59963 @xmath1 candidates with 8.8% background contamination collected by the cern na62 experiment .
the result @xmath3 is in agreement with the standard model expectation . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.